A partnership with academia

Building knowledge for trade and development

Vi Digital Library - Text Preview

Regionalism and Trade Facilitation: a Primer

Working paper by Maur, Jean-Christophe / World Bank, 2008

Download original document (English)

This paper investigates when trade facilitation reform should be undertaken at the regional level. First, looking at both efficiency and implementation considerations, it confirms the perception that the regional dimension matters. Investigating where efficiency gains can be made, this research explains why national markets alone fail to produce the full scale economies and positive externalities of trade facilitation reform. Second, because trade facilitation policies need to address coordination and capacity failures, and because of the operational complexity challenge, the choice of the adequate platform for delivering reform is crucial. The lessons are that regional trade agreements offer good prospects of comprehensive and effective reform and can effectively complement multilateral and national initiatives. However, examples of implementation of trade facilitation reform in regional agreements do not seem to indicate that regional integration approaches have been more successful than trade facilitation through specific cooperation agreements or other efforts, multilateral or unilateral. Customs unions may be an exception here, and the author suggests reasons why this could be the case.

Policy ReseaRch WoRking PaPeR 4464

Regionalism and Trade Facilitation:

A Primer

Jean-Christophe Maur

The World Bank
Development Research Group
Trade Team
January 2008




























Produced by the Research Support Team


The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development
issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the
names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those
of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and
its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.

Policy ReseaRch WoRking PaPeR 4464

This paper investigates when trade facilitation reform
should be undertaken at the regional level. First, looking
at both efficiency and implementation considerations,
it confirms the perception that the regional dimension
matters. Investigating where efficiency gains can be
made, this research explains why national markets alone
fail to produce the full scale economies and positive
externalities of trade facilitation reform. Second, because
trade facilitation policies need to address coordination
and capacity failures, and because of the operational
complexity challenge, the choice of the adequate
platform for delivering reform is crucial. The lessons are

This paper—a product of the Trade Team, Development Research Group—is part of a broader project on trade facilitation
and development supported through a Trust Fund of the U.K. Department for International Development. Policy Research
Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at JC-Maur@

that regional trade agreements offer good prospects of
comprehensive and effective reform and can effectively
complement multilateral and national initiatives.
However, examples of implementation of trade
facilitation reform in regional agreements do not seem to
indicate that regional integration approaches have been
more successful than trade facilitation through specific
cooperation agreements or other efforts, multilateral or
unilateral. Customs unions may be an exception here,
and the author suggests reasons why this could be the

Regionalism and Trade Facilitation: A Primer*, †

Jean-Christophe Maur‡

Keywords: trade facilitation; customs; regional trade agreements; public goods

* This work is part of a World Bank research on trade facilitation and regional integration experiences
conducted with John Wilson and Patrick Messerlin and supported through a trust fund of the U.K.
Department of International Development (DFID). The views expressed in this paper should solely be
attributed to the author and not be construed as representing the views of DFID or the World Bank.
The findings, interpretations, and conclusions expressed in this paper are entirely those of the author.
They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries
they represent.
† The author would like to express his warm thanks to Peter Holmes, Gerard McLinden, Patrick
Messerlin, Ben Shepherd, Matthew Stern, Chris Stevens, and John Wilson for invaluable comments
and inputs. All errors remain the author’s sole.
‡ Economic Adviser, DFID and Research Fellow, Groupe d’Economie Mondiale. jc-maur@dfid.gov.uk.

Trade facilitation reform is the sum of efforts undertaken at the national, regional and

multilateral level designed to reduce trade transaction costs. Multilateral discussions on trade

facilitation have stepped up recently with negotiations in the WTO: until then trade

facilitation was addressed in the GATT in a relatively modest fashion. Other multilateral

instruments getting into more detail, such as the Kyoto Convention, have not mustered

enough momentum for reform. In parallel to multilateral efforts, recent regional trade

agreements have gradually incorporated trade facilitation dimensions (Moïsé, 2002).4

However, to date, a majority of trade facilitation efforts are undertaken unilaterally, the latter

including by low and middle income countries that were thought to lack the capacity to do


While multilateral, regional and national interventions might be competing to achieve

identical aims, they more likely coexist because of their complementarities, and by virtue of

the specific comparative advantages of each approach. However, in discussions about trade

facilitation reform, the regional dimension has not been the object of as much attention as

national and, recently, multilateral reform. Besides, when regional approaches to trade

facilitation reform have been envisaged, for instance in the context of development banks

regional work, or in regional trade agreements, the rationale for the regional dimension of

reform is not always clearly expressed. We also suspect that in many instances a narrow focus

on standalone reforms has prevailed over more holistic approaches. Of course there are


This paper focuses on investigating more thoroughly how regional initiatives can contribute

to trade facilitation reform. As international trade and the costs associated to it involves

activities beyond a single border, the question of where the optimal level of policy

intervention lies – unilateral, regional or multilateral – needs to be examined. This requires

the application of two tests. First a “market failure” test, to assess whether regional public

intervention will deliver a welfare maximizing reform. Because markets are imperfect,

governmental intervention is sometimes needed to deliver the optimal social outcome. When

market failures cannot be remedied at the national level, addressing this becomes a

transnational issue and collective action is needed; regional solutions shall be sought when the

failing markets correspond to some well-defined set of nations.

4 We should note with different motives in sight than the ones underpinning the WTO negotiations.
5 An example is Senegal, who devised their own basic electronic single window system. Over 80 countries have
also implemented ASYCUDA electronic customs modules provided by UNCTAD.


Secondly, a “subsidiarity” test needs to be applied, by which actions to achieve a given policy

objective should be taken at the lowest level of government capable of effectively addressing

the problem at hand (Sauvé and Zampetti, 2000). Ideally this level of action should

correspond to the level affected by the need to provide the regional good, meaning that the

political jurisdiction matches the economic domain of benefits. Thus the most appropriate

participants will partake in the provision of regional trade facilitation and transaction costs

will be economized (Arce and Sandler, 2002).

Our working hypothesis is that in some instances regional solutions, by contrast to purely

national or multilateral ones, will offer the best prospect for meaningful trade facilitation

reform. In essence, this paper explores when regional trade facilitation reform should be

undertaken (the economic case), but also when or how this could be undertaken (the

implementation issue).

With what some perceive as the limits of multilateral trade negotiations, in particular in

relation to development and implementation issues, eyes turn again towards regional

solutions. This research does not put itself in this perspective, but rather aims at

understanding better how regionalism - a force that cannot be denied - can ideally

complement and contribute to the facilitation and liberalization of trade.

This paper sets itself in the direct continuation of the broad investigation into regionalism by

the World Bank, a research that was summarized in Schiff and Winters (2003). The specific

topic of trade facilitation, although mentioned in various outputs of this research program did

not however receive full treatment. In an earlier work the authors provide insight into the

relevance of regional trade agreements for several policies, among them transport, itself an

element of trade facilitation. They note, as we do in detail below, that “… in the presence of

economies of scale or inter-country externalities, market solutions are generally sub-optimal,

and failing to cooperate can be very costly . However, regional cooperation is not the same as

regional integration, and, indeed, there is generally rather little connection between them”

(Schiff and Winters, 2002).6 This remark alludes to the two main dimensions of regional trade

agreements investigated in this paper. First, from an economic perspective, under what

circumstances regional initiatives are optimal given market failures affecting the supply of

trade facilitation services? Secondly do regional approaches offer better and more cost-

effective prospects than other institutional solutions to achieve the objective of trade

facilitation reform?

6 On the substantive differences between regional trade and cooperation agreements, see Devlin and Estevadeordal



We start by defining what we mean by trade facilitation. There is no universal understanding

of what trade facilitation is (Wilson et al., 2002), reflecting differences, as well as some

evolution, in views of what should be the reforms undertaken to reduce the cost of trading. In

simple terms trade facilitation can be thought as the simplification of the trade interface

between partners. This trade interface is composed in a broad sense of compliance to

government rules by traders, enforcement by authorities of these rules (including taxes),

exchange of information, financing, insurance, ICT and legal services, transport, handling,

measurement and storage. We focus therefore in the rest of this paper on this broad

conception of trade facilitation as the private and public interventions that help goods cross


Government interventions in all these aspects of the trade interface affect the magnitude of

transaction costs incurred by traders. For example, the sheer diversity of government

regulations, and the replication of their enforcement causes duplication (e.g. compliance cost

with two different standards: Baldwin, 2000) and friction costs (e.g. time lost because of

repeated loading and unloading of merchandise) that regional harmonization and cooperation

could conceptually help address (Schiff and Winters, 2003: 82).

The relevance of geography for the tangible and intangible dimensions of the supply


Country borders create costly obstacles to international trade. The empirical reality of the

“border effect” is demonstrated by the gravity model of international trade (for a recent

overview see: Anderson and Wincoop, 2002). Trade flows between pairs of countries are

proportional to their Gross Domestic Product and inversely proportional to the trade barriers

between them. It is well known from this literature that trade transactions are determined by

geography, and more broadly of non-policy elements that draw two countries closer: It is

customary to include indicators of geographic proximity such as distance and common

borders and of broader indicators of proximity such as commonality of language, legal

systems, history, etc. Some barriers to trade facilitation are associated with these “natural

barriers” which have a lot to do with geography.

Components of the international trade interface are intangible (payments for instance), and

others are tangible (physical transport). Looking first at the tangible aspects of the

international supply chain, it is relatively straightforward to identify components affected by


geography: transport; storage; and physical inspection and presentation of documentation at

border agencies. Taking the example of transport, international road shipping involves border

crossing, and transit through neighboring countries. Road and rail transporters have to meet

possibly different local legal regimes and standards. Different transport standards, such as

varying maximum axle loads permissible for trucks (Namibia, Zambia, Botswana),7 or

changing rail gauges width for trains necessitates unloading and reloading and disrupt the

supply chain.8 Compliance with different regulations at each border increases compliance

costs and add to the time spent there.

Enforcement of border agency controls is highly localized, often near border crossings

(although not all controls are). When crossing several borders, costs have to be incurred many

times if customs do not cooperate (at the border of the originating and destination country,

and of countries of transit). Arvis et al. (2007) talk of “triple clearance time” because of

duplicating lengthy clearances on most transit corridors in least developed countries. The cost

of duplication is also magnified by the number of regulations affecting trade implemented by

border agencies, including entry visa requirements, technical and phytosanitary standards,

security checks, tax levying, etc.9 Such requirements involve producing paper documentation

at border posts, storage in bonded warehouses and physical inspections. The tangible

elements of international trade transactions therefore strongly suggest that costs are

geographically localized. This may require local solutions: such as the sharing of border


However, a feature of modern trade facilitation is the elimination of costly physical

operations. Thus the tangible nature of trade transactions is being reduced: paper work is

replaced by electronic documentation - de-linking the production of documentation with the

physical flow of goods (National Board of Trade, 2003) - modern risk management and non-

intrusive techniques reduce considerably the need for physical inspection. One consequence is

that transport costs become less related to geographical distance.

7 Röschlau (2003).
8 Lack of regulatory interconnection of rail systems creates also disruptions. Before its collapse in 1977, the East
African Community (Tanzania, Kenya and Uganda) had an integrated rail network. In the 1980s the Kenya
Railways Corporation was transporting twice the volume of freight that it now carries.
=228424&Projectid=P079734 for more info.
9 De Wulf & Sokol (2005) give this example: “the United States Customs and Border Protection enforces laws on
agriculture, alien and naturalization, banks and banking, census, commerce and trade, conservation, copyrights,
crimes and criminal procedures, customs duties, food and drugs, foreign relations, internal revenue, intoxicating
liquors, money and finance, navigation, patents, postal service, public building and property, public land, railroads,
shipping, telegraphs and telephones, territories and insular possessions, transportation, war and national defense,
and international treaties, statutes, and agreements”.
10 We discuss this below in the text.


Does this imply that the regional dimension does not matter for intangibles aspects of

international trade? Not necessarily. The border effect also holds for services (Kimura and

Lee, 2006). Portes and Rey (2005) show that cross-border equity transaction flows, which,

because of their intangibility, one could have presumed not to be affected by transaction

costs, are subject to similar determinants as goods flows. Geography is therefore an issue.

Regional proximity indeed favors better information flows between traders, shared cultural

references, and importantly for trade facilitation, practices and common systems developed

through time.

Implications for trade facilitation reform

What does this mean for trade facilitation reform? First, since proximity is associated with the

existence of location-based transaction costs, this could imply that groups of countries who

incur common local transaction costs could try to address them jointly. In a way this means

that there are countries with whom it is more “natural” to cooperate with to implement trade

facilitation than others. Some economies of scale are geography dependent, such as physical

transport networks.

However, the attractiveness of geographical proximity for trade does not hold for many other

aspects of international cooperation dimensions of trade facilitation, and the “natural partner”

may be one far away. Other dimensions of trade facilitation relate to intangible costs. One

could surmise that for groups of countries that enjoy already high volumes of trade, and that

presumably already share good knowledge of the trade facilitation impediments arising in

their trade, would potentially be better placed to undertake reform together than with other


In summary, the relevance of proximity can be established for trade facilitation, but it

provides relatively little guidance in terms of rationale for regional cooperation except

perhaps a presumption that there is some role for regional intervention. Regional cooperation

should then be guided by the identification of efficiency gains realized at the scale of the


There are two main dimensions to the benefits of transnational cooperation, corresponding to

the specific market or institutional failures faced by countries: first, the realization of

economies of scale, including by elimination of duplication and increasing competition;

second through the avoidance of negative and creation of positive externalities among

neighbors. Obviously, in order to make sense from a regional economic perspective, cost


reduction must either be more efficiently undertaken regionally than say multilaterally or

belong to costs categories that are incurred only regionally (i.e. on a geographical basis). The

question of delivery of regional public goods is particularly relevant in this respect.

Realizing socially optimal economies of scale

With many interventions and parties in the international transport of goods, and possibly the

crossing of several borders, there is plenty of scope for cost duplication for trade operators.

An important portion of these costs being fixed, eliminating any duplication will enable

efficiency gains for firms, but also allow smaller scale operators to access export markets, an

important aspect for developing countries.11 Additionally, because some of the procedures

and services that facilitate trade involve large fixed and possibly sunk costs, full economies of

scale in the administrative procedures and services to international trade transactions may not

be realized at the country level, especially in small and poor countries.

Duplication arises because similar requirements must be met repeatedly, but also because

national rules differ, therefore also increasing search costs and associated uncertainty,12 and

creating further opportunities for rent seeking and corruption. COMESA’s regional carrier’s

licensing system illustrates how duplication can be tackled. The regional license avoids

having to pay for multiple licenses (Schiff and Winters, 2003: 81). Likewise, inspection of

goods, if carried out in different places on each side of one border delays trade. Different

national regulatory requirements force traders to meet two standards instead of one: in

Tanzania registration requirements for agro-chemical pesticides are burdensome and subject

to high fees, despite the fact that Tanzania’s market for such pesticides is small and

equivalent and more efficient products are already registered and tested in neighboring Kenya

(Tanzania DTIS, 2005).13

Solutions to reduce duplication costs may involve harmonization, forming a common market

within a customs union, mutual assistance among authorities (an important facilitating

11 World Bank (2005: 85) quotes for instance the cost of certification of organic nut production in Moldova for
export to Germany which can amount to $18,000 per year, a not insignificant amount for firms in poor countries.
12 Arvis et al., 2007 examine the large impact associated with uncertainty along the supply chain created by non
harmonised regulations.
13 The Tanzania certification and testing agency pesticides “… charges relatively high fees to register an agro-
chemical and also requires three years of field testing. It does not recognize the testing done and registration of
chemicals in neighbouring countries, including Kenya. Hence, there are a broad range of newer, more effective
and safer chemicals which do not get registered in Tanzania because of the high cost and which are prevented from
being legally imported from Kenya or other neighbouring countries. The chemical registration revenue imperative
of TPRI thus appears to take precedence over a feasible solution of mutual recognition of other (including more
rigorous) testing and registration systems.”


practice for customs valuation),14 and mutual recognition of rulings (e.g. for transit

operations) and of certification and testing (e.g. for standards). Such cooperation is not

necessarily regional or bilateral and can also be achieved multilaterally. It may, however, be

the case that a large part of these extra costs be better addressed at the regional level because

of the political economy and complexity of such arrangements (such as mutual recognition) is

most probably better managed among a limited number of countries. Secondly,

implementation of cost reduction measures will often involve some form of regional

cooperation on the ground: for example, the creation of joint border posts enables neighboring

countries to share facilities, learn from each other and carry joint inspections thus potentially

reducing cost and time spent at the border.15

Economies of scale can also be realized on the administrative procedures and private services

delivering trade facilitation. It is however unclear to what extent there is a scale barrier to

efficient border administration, including modern practices such as single windows and risk

management. The cost of collecting customs taxes in Rwanda is relatively low, amounting to

2.5% of receipts in 2005 with capital costs representing only a small fraction (2.6%) of these

costs.16 It is also unclear to what extent compliance with new security measures imposed by

large developed countries pose a new challenge in that respect. Overall there is good evidence

that customs reform projects can be self-sustaining through increase in revenues brought by

facilitation (Moïsé, 2005) suggesting that economies of scale are exhausted at the country

level.17 Arguably this tends to rather relate to still relatively modest levels of reform, and in

particular do not involve many elements of deep integration with trading partners. Although

customs operations do not seem to be overly subject to important economies of scale in most

cases, one can nevertheless argue that there are important costs associated with the

surveillance of borders. A motivation for Norway, Sweden and Finland to sign cross border

cooperation agreements (starting in 1960) was “division of labor”; i.e. to share the cost of

individually manning the 1,630 km long border between Norway and Sweden, and the 739

km long border between Norway and Finland.18 Small administrations may not be able to

14 For a discussion, see Chapter 8 in De Wulf & Sokol (2005).
15 The benefits of joint border posts should however not be overstated. Practical implementation has proven
problematic as the incentives for the border agency of each country to cooperate on joint inspection may not exist
as import and export controls remain very different as incentives to control them (the emphasis is generally on
imports and their contribution to tax revenue).
16 Source: Rwandan Customs, http://www.rra.gov.rw/en/about/performance.pdf.
17 In Mozambique, the investments made during the initial stages of the program were recouped within 14 months
from additional revenue receipts. In Peru, increased revenue collection has helped solve many funding problems of
the administration.
18 See the communication by Norway to the WTO Trade Facilitation Negotiating Group on Border Agency
Cooperation, TN/TF/W/48, 9 June 2005. In 1995 Norway calculated the savings associated with the two
agreements: 10 new customs offices would have had to be opened on the Norwegian side of the border.
100 new customs officers would have had to be employed. This would have cost about 8 million USD in
additional investment costs and 8 million USD in recurring annual costs for the customs authorities for new


afford all the material and infrastructure necessary. For instance, in the current WTO

negotiations on trade facilitation, some members have called for “small vulnerable

economies” to undertake a regional approach to the implementation of some expected WTO

commitments that will require capacity building and made a specific submission for regional

trade facilitation enquiry points.19

Beyond customs operations, many developing countries are too small (and have too small

markets) to offer the full range of standards conformity assessment, which could thus benefit

from regional integration (World Bank, 2005: 90). Setting up regional accreditation bodies or

opening regional markets for accreditation bodies could be a way to provide cheaper and

better testing, building on scale economies and comparative advantage. Scarcity of technical

skills20 is indeed another reason why regional approaches can make sense for countries facing

serious shortages in these skills, which for modern trade facilitation techniques can become

an issue.

Backbone services provide crucial inputs in trade transactions: finance and insurance,

transport and logistics, handling, measurement and communication services. The delivery of

these services for trade transactions can require a scale of production beyond the national

borders.21 Insurance and financial services (letters of credit, guarantees, insurance, etc.) are

key input to the capacity to trade internationally: national operators in developing countries

may not provide them or at non-competitive prices. According to EBRD (2003) national

banking systems do not pool enough capital22 to underwrite trade transactions. Financial

guarantees for payments are often not available for small firms to allow them to export.23

Similarly, fixed costs and geographic factors confer natural monopoly characteristics to some

modes of transport (rail and maritime particularly). Ghana’s Growth and Poverty Reduction

buildings, salaries, etc. Economic operators would have incurred an estimated 39 million USD additional annual
costs due to longer waiting time and double stops at the border.

19 Communication from Barbados, Fiji, Papua New Guinea and the Solomon Islands to the WTO Negotiating
Group on Trade Facilitation, 7 July 2006, document TN/TF/W/129.
20 An essential piece of the architecture for the enforcement of technical regulations and SPS measures in
international commercial exchanges is accreditation, which offers an internationally recognized guarantee that
national processes of assessment of standards conformity can be relied on (Holmes et al., 2006 offer a good
overview of the question). In Sub-Saharan Africa there was until recently only one accredited expert, located in
South Africa, able to grant this accreditation. Now three additional experts have been trained, still in South Africa.
This confirms the view by ILAC-UNIDO (2003) that markets for accreditation and certification bodies in many
developing countries may be too small.
21 Arvis et al. (2007) for instance report how advanced logistics services are inhibited by lack of trade facilitation:
it is often impossible, note the authors, to maintain multi-country inventories or to dodge first clearance and then
re-export to the gateway country.
22 Actually payment guarantee systems require less working capital than payments in advance (which are required
when there is no guarantee) and thus help smaller agents to access international trade.
23 This has prompted the EBRD and the IFC to create international risk sharing funds in order to provide small
enterprise with access to trade finance. The risk sharing funds help international banks (confirming banks) cover
the political and commercial risk faced by local issuing banks when they cover international trade transactions.


Strategy (Ghana National Development Planning Commission, 2005) envisions regional

cooperation for cross-border road infrastructure development. Regional transport hubs help

realize economies of scale.24 For freight transport, the emergence of multi-modal hubs

(generally located near important existing infrastructure: air, sea or rail) generate important

economies of scale (through higher utilization of infrastructure), and efficiency gains (through

competition between modes of transport) compared to point-to-point routes (Müller-Jentsch,

2002). Transport hubs depend as much, and probably more from the liberalization of regional

transport services (such as liberalization of cabotage or air traffic rights) than the availability

of infrastructure. Indeed, transport hubs tend to be geographically mobile, suggesting the

secondary importance of infrastructure as a determinant of their location.

Lastly, regional approaches to trade facilitation can also aim to create competition. First,

among regional operators in the international trade transport and logistics chain; this will

reduce the cost of trading and increase the availability of services to exporters and importers,

thus contributing to trade creation. Arguably, better than regional competition is competition

with the world. There are nevertheless arguments in favor of a regional approach. First, it

does not have to be discriminatory. Also, as seen earlier, transport competition is increased by

the establishment of regional hubs. The negotiation of bilateral liberalization agreement in air

and maritime transport is more straightforward than multilaterally. Mattoo and Fink (2002)

indeed find that more efficient bargaining may be possible in a pluri-lateral context than in the

multilateral context for services: there is less concern that outsiders will be able to free-ride

on the reciprocal exchange of concessions than if there were a general MFN obligation.

Increased competition in trade related services is also largely dependent on some level of

regulatory cooperation, which is generally more feasible and in many cases more desirable

among a subset of countries than globally. Competition dimensions matter particularly in the

context of transit corridors.

Coastal countries can be placed in a situation of monopoly for transit services: an example is

the transport of oil products to Uganda by the Kenya Pipeline Corporation, which in the

absence of rail and road transport competitors can charge excessive prices (Uganda DTIS,

2006).25 Competition among transit corridors is also desirable and regional agreements

between more than two countries can create the conditions for the establishment of transit

agreements competing against each other. The World Bank (2005) reports that transit

agreements are often governed by unilateral or bilateral frameworks and such arrangements

24 For instance air transport hubs avoid having empty cargo on incoming or outgoing freight, a problem for small
non diversified economies.
25 Until 1996 this monopoly was even reinforced by government regulation forbidding the transport of petroleum
by road operators. This restriction is now lifted.


are not conducive to competition. Participants in “tour de role” and 50-50 sharing type

agreements oppose regional agreements because they want to stave off competition.

Competitive pressures also hold for government agencies. By offering institutional

mechanisms through which border agencies can exchange information and benchmark each

other’s performance, regional agreements offer more transparent regulatory competition that

can foster more efficient border management. However, this is to be balanced with the other

consideration that economies of scale and specialization justify one approach for the region.26

Regional negative and positive externalities

Regional agreements can serve as a policy coordination mechanism to help prevent individual

countries to opt for national strategies that are fall short of optimal global outcomes. For

instance, countries of transit trade are often tempted to use trade restricting policies such as

seeking to set revenue maximizing fees on transit, imposing compulsory transit routes and

check points, or the use of mandatory securitized convoys. Fees and requirements will be

above the cost of services provided (this includes the use of roads, provision of security, etc.)

or strictly necessary for secure transit.27 In the worst cases motives behind these policies are

protectionists; often, this is the outcome of absence of consideration28 of negative

externalities imposed on neighbors. Such risk is particularly important when alternative

transit routes are few as this often happens in Africa.29 Domestic transport infrastructure

constraints often have regional implications, justifying from an economic point of view port

or regional airport hubs. Landlocked countries depend on the quality of the infrastructure of

eir neighbors.30



d is

congested, transit bond regimes are financially burdensome, rail transit does not offer a


A finding from the diagnostic on transport and trade facilitation in Uganda (Uganda DTI

2006) is that the most important transport and trade facilitation issues are outside of

country’s direct control. Tanzania and Kenya, its coastal neighbors offer poor trade

facilitation: the port of Mombasa where 95% of Uganda’s external trade traffic is handle

26 The Tanzania DTIS (2005) raises this question about having regional accreditation organisations. In this case,
the benefits of joint approach seem high enough to justify foregoing regulatory competition.
27 McTiernan (2006) for instance reports that Benin and Togo charge very high fees for transit which incites
transport from Lagos to Accra to be done by ship.
28 Not merely oversight or neglect, but lack of incentives on the country of transit to internalise the costs of more
efficient transit.
29 A counter example is Bolivia, which has several access roads to the sea (Schiff and Winters, 2002)
30 For instance poor road conditions and border crossing procedure between Kenya and Uganda at Malaba have
made rail and water transport across Lake Victoria a competitive alternative; or because the route between Durban
and Kampala is more reliable, some traders of high value goods prefer it to much shorter routes to Dar Es Salaam
or Mombasa (Caron and Reichert, 2001).


competitive alternative to poor road transport and expensive pipeline transport.31 As Schiff

and Winters (2002) note, this type of externality is often asymmetric, with landlocked

countries standing to gain a lot from better transit, when the gains for the coastal partner are

much smaller (improved access to the internal market). In practice, landlocked countries have

not gained much from participation in regional trade agreements, and this is likely because

important trade obstacles have remained (Yang and Gupta, 2005).

Once again, standards and phytosanitary measures provide a further example of possible

market failure. Weak or absent enforcement of SPS in one country can mean that negative

consequences can spill over to neighbors, as is the case with Tanzania, where highly

contagious bovine diseases are not fully under control. In this context Tanzania has agreed

with SADC neighbors to a 5-year program of vaccination, surveillance and control of animal

movements (Tanzania DTIS, 2005).

As well as getting rid of negative externalities, creating positive externalities such as network

effects may substantiate regional intervention. Transport, electronic and other information

systems networks play an increasingly important role in trade facilitation reform. There are

positive externalities for neighboring countries to join existing networks rather than

developing own systems or multiplying bilateral channels of communication and exchange of

information. The European Union has developed several networks initiatives around

electronic transit systems (NCTS), satellite information (GALILEO), or Trans European

Networks on Transport. This concern is also reflected in trade action plans in developing

countries such as in the Uganda’s recent DTIS which puts emphasis on the use of EDI

interchange at the regional level, the development of a regional cargo tracking system and the

interconnection of East African Community’s customs electronic systems (Uganda DTIS,

2006). The same issue of interconnection is noted in the case of Mozambique. Mozambique’s

proprietary customs electronic system is incompatible with Asycuda++, used by

Mozambique’s Southern African neighbors, and on which they intend to build their exchange

of electronic information on transit cargo (Mozambique DTIS, 2004). Transport hubs,

mentioned earlier, also create positive externalities, such as access to multi-modal transport


31 Arvis et al. (2007) add that because of railways poor performance and unpredictability (Tanzania Railways
Corporation has an error margin of 4 to 5 days predicting the arrival of any shipment), road transit from Kenya to
Northern Tanzania has increased by 20% over the last five years. This also explains why 75% of Rwandan trade
transits through Kenya while 50% transited through Tanzania only 3 years ago.


Box 1. How regional cooperation can help: DTIS recommendations about SPS
for Tanzania

The 2005 Diagnostic Trade Integration Study conducted in Tanzania under the
Integrated Framework finds that a regional approach and cooperation for SPS and
quality standards can:

• increase trade through harmonization of standards and mutual recognition
of conformity systems (avoiding duplication);

• spread knowledge and good practices (creating positive information

• more effectively manage trans-boundary risks (avoiding negative

• serve to realize economies of scale and scope in the delivery of conformity

• better enable regional enterprises to comply with extra-regional standards
where collaboration facilitates international accreditation;

The DTIS recommendations prioritize the following areas of regional cooperation:

• Streamlining of regulations (simplification and improved transparency) and
mutual recognition;

• Resource pooling (‘centers of excellence’ for the testing, registration or
other monitoring of inputs or outputs for specific products such as
pesticides, condoms or cosmetics);

• Multi-country collaboration for problem solving (surveillance and
contingency planning, but also research, pilot programs and training).

Source: Tanzania DTIS (2005: 91-98) and author

Positive externalities also arise – beyond the mere realization economies of scale described

earlier – from the provision of international finance and insurance. International provision of

such services offers the possibility to mutualize risks across a region, and contribute to

positive network effects, such as linking banks that usually do not do business with each other

and diffusing skills through the network. The principle of mutualization is applicable for other

trade related financial instruments specifically relevant to trade facilitation and transit such as

guarantees for payment of taxes and insurance. COMESA introduced the Yellow Card or

Third Party Regional Motor Vehicle Insurance scheme which allows traders to purchase

insurance covering transport in the region (Arvis, 2005). The region also plans to set up a

regional transit bond scheme. Regional guarantees (to secure the payment of duty and taxes)

can address the failure of national organizations to set up such systems, because they often do

not have the sufficient size, have only access to underdeveloped national financial services,

and international insurers are not willing to face the political and commercial risks of

developing markets. In Uganda, the cost customs bonds is estimated to add up to 4% to

import and export costs and a recommendation of the Integrated Framework diagnostic study

is to use a regional approach to reduce their incidence (Uganda DTIS, 2006). Arvis (2005)

argues that the lack of a regional customs guarantee explains why transit initiatives to

replicate the success of the European TIR system have failed so far.


We conclude this section by noting that there is a risk for regional initiatives to create their

own negative externalities. This, for instance, can occur when countries take part in RTAs

with policies going against trade facilitation objectives. The obvious example is the need for

rules of origin in the absence of a common external trade regime, which represent specific

difficulties in for customs enforcement (World Bank, 2005: 68). Another problem can be

created by the incentives to harmonize RTA members’ practices, ending up in adopting

higher standards than strictly necessary.32 Thirdly, there is also the risk of incoherence with

trade facilitation objectives of the policies agreed when signing the RTA: when adopting the

East African Community common external tariff, Uganda had to raise its tariff on vehicles

from 9% to 25% (customs duty and import commission), which increases the cost of transport

in Uganda (Uganda DTIS, 2006). Finally, regional trade facilitation should not result in trade

diversion effects. There are two ways in which this can happen. The most serious one is when

trade facilitation measures have their own discriminatory effect. For most trade facilitation

measures this is not the case, but it is nevertheless a possibility for several important policies,

such as mutual recognition and transport infrastructure policies, in the latter case if the

improvement of one mode of transport (road for instance) is privileged over others. More

benignly, in a second best world, trade facilitation may simply augment existing distortions if

these are left untouched.


A public good perspective

The benefits from regional trade facilitation display public or quasi-public goods properties.

Public goods indeed vary in their degree of “publicness” (Sandler, 2006) and thus display

different properties. One reason why it is important to understand these properties is because

this has a bearing on understanding better how the benefits from regional trade facilitation

reform ought to be delivered.33

A public good is a good giving rise to two distinct market failures. The first failure arises

because of non excludability: providers of the good cannot prevent others from free riding by

consuming it at no cost. The second failure is caused by non rivalry: the consumption of one

32 This is a question that is difficult to assess in practice, but if one takes the example of regional trade agreements
signed by the European Union, they make clear reference to harmonization to European standards (see Maur,
2005). Otsuki, Wilson and Sewadeh (2001) give an example in how European SPS measures can be particularly
33 The growing literature on regional public goods offers a general discussion of this. See: Sandler (1998; 2006);
Estevadeordal, Franz and Nguyen (2004); Cook and Sachs (1999); Ferroni (2002); Arce and Sandler (2002); Kaul
et al. (2003), and Kanbur et al. (1999). Stålgren (2000) offers a review of the literature. See also Rufin (2004) on
transport and communication infrastructure, and Holmes et al. (2006) on standards.


unit of the good does not diminish the quantity available for consumption to others, meaning

that once a public good is provided, all can enjoy it at no or very low cost.

A quick examination then tells us that most of the benefits provided by regional trade

facilitation are not pure public goods. They are indeed almost all characterized by near total

excludability. This is clearly the case for transit corridors, transport infrastructure, financial

and communication infrastructure, networks, authorized traders regimes, and collocation of

customs services. The only dimension of trade facilitation that seems to be a pure regional

public good relates to communicable diseases. For the other regional public good dimensions

of trade facilitation, the full exclusion characteristic has actually positive implications.

Excludability indeed means that fees can be charged to finance their supply. It seems also that

once provided, trade facilitation goods are mostly non-rival in nature: new and improved

procedures or systems can be enjoyed by all at little or no marginal cost (although in some

specific cases, discrimination might still have to be borne in mind). In essence regional trade

facilitation is a mostly a “club good”.

Secondly, depending on the type of public good, the capacity of countries, individually or as a

group, to affect the supply of the public good varies: what in the literature is called

aggregation technology (Hirschleifer, 1983). In the case of trade facilitation, the specific

difficulties for the delivery of regional trade facilitation goods are quite varied. In the first

instance, a regional good could in theory be provided by any country within a region, and thus

it is desirable for countries to coordinate over this, otherwise one risks to have too much of

one good. Airports and ports hubs, cost duplication elimination are two illustrations where

coordination is optimal.34 A second, different example is the integrity of a customs union,

which depends on the member with the weakest customs enforcement capacity (this is

discussed below).35 In this case, the failure is linked with lack of capacity. Incentives are big

for customs union members to provide the weakest link with the technology to enforce

customs disciplines at the higher standard. Third case in hand: networks, but also many other

aspects of regional trade facilitation, are affected by the sum of efforts of the countries


In the all these cases, the contribution of each country to the supply of regional trade

facilitation is to some extent substitutable for the contribution of others and thus creates free

riding incentives. These have to be addressed if undersupply is to be avoided, and requires

34 “Best shot” and “better shot” categories in Kanbur et al. (1999) typology.
35 “Weakest link” and “weaker link” categories (Kanbur et al., 1999).
36 This is to build the network. Note on the other hand that the integrity of the network is affected by the
contribution of its weakest link.


some form of institution building (Arce and Sandler, 2002). However, the type of aggregation

technology matters as contributions to the total effort are not necessarily equally distributed,

which alters the free riding incentives of each country. An implication is that despite having

in common the need for regional intervention, various components of trade facilitation reform

cannot be delivered using similar approaches. An important remark in this respect is that the

role of individual countries belonging to a region will vary depending on what type of

regional public good that needs to be delivered.

The nature of regional initiatives

The regional nature of the various aspects of trade facilitation reform suggests that regional

initiatives are in a unique position to help tackle trade facilitation issues (World Bank, 2004,

2005). There are numerous institutional public and private arrangements that have the ability

to deliver these regional goods. The most common are regional trade integration agreements

and regional cooperation agreements. But other arrangements can also supply regional public

goods, such as public private partnerships (in several transit corridors), regional development

institutions, or NGOs (Sandler, 2006). The focus here is primarily on regional trade

agreements, which for reasons discussed below, but also more generally in the literature

(Devlin and Estevadeordal, 2004) appear to offer the greatest potential for efficient delivery.

Because trade facilitation involves a wide range of regulatory activities, it makes sense to

concentrate on institutions that are largely public, such as regional cooperation agreements,

when we discuss alternatives to regional trade integration.

A fact is that regional trade agreements (except for customs unions) have not included much

in the way of trade facilitation efforts (Moïsé, 2002), at least until recently; Bin and

Misovicova (2007) note that the number of agreements covering trade facilitation in Asia and

the Pacific has significantly augmented in the recent years and that 34 out of 102 signed

RTAs now include some trade facilitation provisions. The institutional setting of regional

agreements seems well suited to pursue a trade facilitation agenda. Trade Facilitation indeed

requires not only the elimination of distortionary and inefficient rules and practices, but

mostly to carry on an ambitious and positive agenda of reform by implementing

internationally compatible modern legislation, systems and skills. In essence trade facilitation

is about deep integration.


How RTAs can facilitate trade facilitation

The World Bank (2005) lists several areas of facilitation in which they view RTAs can lead to

improvement: alignment of customs codes with international standards; simplification and

harmonization of procedures (e-documents and single document); alignment of tariff

structures with the HS; transparency; effective implementation of the WTO valuation

agreement; joint work towards customs integrity; establishment of joint border posts; and

joint training centers. A regional approach to some of the items in this list (alignment with

international standards, transparency) does not obviously stem from any the economic

advantages discussed above, implying that there is not necessarily a direct correspondence

between the economic optimal level of intervention and the jurisdictional one: the subsidiarity

question. The interest of taking such reforms forward in regional agreements is thus not be

only because of pure economic reasons, but also because regional trade agreements offer

specific advantages over other forms of international agreements (multilateral or other

regional forms of cooperation) and unilateral initiatives.

We examine the following characteristics of RTAs: i) a forum to exchange concessions across

a broad array of sectors; ii) the access to mechanisms of cooperation, including in some

instances financial transfers and capacity building; iii) a mechanism for political commitment;

and iv) a more efficient approach to harmonisation and implementation.

RTAs as a forum of issues

First, deep RTAs offer the possibility of comprehensive trade facilitation, involving reform in

several sectors of the economy that can be incorporated in the new generations of RTAs.

Deep integration agreements present the opportunity to include sectors that are not well

covered multilaterally, while also providing efficient enforcement mechanisms (we treat this

below). The wider remit of RTAs compared to multilateral approaches (binding such as the

WTO, or other international organizations with less enforceability such as the WCO, which

focuses only on customs) is reflected in initiatives such as APEC, which uses of a broader

definition of trade facilitation reform compared to the relatively narrow approach taken in the

WTO (Wilson et al., 2002). By incorporating policies for which there is no actual or possible

prospect multilateral liberalization, some regional agreements offer increased scope for

meaningful trade facilitation reform. A good illustration of this is the adoption of flexible and

harmonized policies on visas and opening of services, dimensions usually out of reach of a

multilateral agreement, but which can be part of regional discussions; ECOWAS, for instance,

suppressed visas between member countries. The political economy of RTAs makes it easier


to deal with migration issues because countries can exchange commitments on natural

persons movements (whereas not possible in GATS).37

By tackling many dimensions of trade facilitation at the same time, regional trade agreements

also give the scope of exploiting natural complementarities between the different elements of

trade facilitation reform.38 A particular challenge of reform is to get all the agencies involved

in border control to work to a common objective of facilitating trade. This is often not

happening. By having the policy areas implemented by these agencies (such as SPS and

standards) covered in a same trade agreement offers scope for agreeing to common aims

linking these policies. In theory the WTO could offer similar advantages, but there the

multiplicity of diverging country interests behind each issue makes any attempt at defining a

coherent approach across them difficult and reduced to the common high level principles of

non discrimination, transparency or due process, but still a far cry from delivering coherent

implementation of these policies.

Conceivably all the facets of trade facilitation reform could equally be included in either

RTAs or specifically designed cooperation agreement. However, RTAs generally offer the

scope of a wide spectrum of policies across which trading-off concessions (Devlin and

Estevadeordal, 2004), including non-economic dimensions.39 Besides, because of this breadth

of scope, regional trade agreements can also guarantee better commitment. In theory any

attempt to deny any trade facilitation concession by the imposition of other trade barriers (e.g.

tariffs) should be more complicated as these are part of the agreement. By the same token,

enforcement of trade facilitation measures will be guaranteed by the possibility for partners to

withdraw any other concession. Obviously there many limits to this happening in practice,

starting with the fact that despite being broad in scope, many existing regional agreements are

more shallow than deep, thus offering still quite a lot of flexibility to members to escape



Regional trade agreements are not infrequently complemented by sharing of resources and

redistribution mechanisms among partner countries, including the supply of financial and

37 Although this is only likely when partner countries are have similar levels of development and patterns of
comparative advantage that make movement of natural persons relatively balanced. Bin and Misovicova (2007)
find that provisions on mobility of business persons are present in about one third of RTAs containing trade
facilitation provisions in Asia and the Pacific.
38 For instance regional guarantee systems help establish global standards for documentary credity (EBRD, 2003)
and thus generate information that can be used for other purposes.
39 At the same time this increases the complexity of negotiating across a wide range of issues.


technical assistance. Trade facilitation reforms can be demanding, both in expertise and

material.40 Regional efforts offer in addition the possibility of benchmarking (for example the

regional program for Trade and Transport Facilitation in South Eastern Europe, De Wulf and

Sokol, 2005: 137) the sharing of good practice. Several regional agreements have a program

of regional capacity building. This is the case of APEC and COMESA. APEC has developed

a program of technical assistance for which members have drawn a collective and individual

Country Assistance Plans covering 16 areas.41 The European Union devotes rather

considerable sums and efforts to assisting neighbor countries with which it signs association

agreements (OECD, 2005). In particular, as discussed earlier, when the provision of better

trade facilitation at the regional level is impaired by the lack of capacity of a few members,

with implications as a whole for a regional system, regional groupings will assist the delivery

of joint assistance (acting as a coordination mechanism and sharing the costs among


Even in the absence of redistribution arrangements, RTAs potentially create beneficial access

to external financial resources as they may increase the credibility and ability of the regional

group to offer loan collateral for instance (Devlin and Estevadeordal, 2004). Regional

guarantee systems would benefit from this.

Trust and institutions

It is also well-known that regional trade agreements act as trust building mechanism, favoring

interactions between officials and exchange of information (Schiff and Winters, 1998). Trust

is a vital aspect of trade facilitation cooperation, as it helps mitigate risk – and thus reduce

physical constraints on the transport of goods such as inspections or requirements to abide to

certain requirements such as compulsory routes – through increased confidence in shared

information and systems. While regional trade agreements have a good track record as

enabling trust building across partner countries’ administrations, attempts to involve

businesses, for instance through public private partnerships have been much less successful:

40 This question is the subject of some debate: it is argued that developing countries are actually more advanced
than commonly thought and that a large share of essential trade facilitation measures would not be that costly and
that the main hurdle is political. This is for instance the conclusion reached by McLindern (2006), arguably in the
narrow context of WTO negotiations. On the other hand, there is also evidence that ambitious customs reform
(often as part of revenue reform) mobilises very significant donor support over several years as in the instance of
South Eastern Europe.
41 APEC Sub-Committee on Customs Procedures (2006). CAP Assessment/Evaluation Matrix. September 2006.


for instance the European Union has tried to build ambitious public-private partnerships in the

context of its European transport network policy, with mixed success.42

In the context of regional trade agreements, customs cooperation committees are often

established to discuss enforcement issues and also help diffusing disputes (World Bank, 2005:

89). More informal expert groups have also been established in regional integration context,

such as the EU Florence process on infrastructure, which has been influential for promoting

reform (Rufin, 2004). Also, through harmonization of instruments and increased

transparency, it becomes easier to understand what trade partners are up to. The establishment

of trust is central in achieving devolution of responsibilities to partner countries such as in

mutual recognition and regional guarantee schemes, but also in establishing shared facilities.

Trust not only matters at the technical level, but of course also at the political one. An

illustration of the link between trade facilitation and international trust, albeit not in the

context of an RTA is given by the recent decision by India and China to reopened to trade the

Himalayan pass of Nathu-La, after four decades of military tension.43 The trust dimension

takes an added importance with enhanced concerns about security after the terrorist attacks of

11 September on the United States. Better security involves better border control, which can

benefit from regional cooperation.

Regional trade integration implies the building of regional institutions that can take forward

some policies on behalf of its members. RTAs offer a cost-saving institutional architecture

(Devlin and Estevadeordal, 2004; Sandler, 2006) through which the demand for regional

public goods can be more easily aggregated. The redistribution mechanisms discussed above,

but also the cooperation mechanisms established through a RTA will help both achieving

better cooperation (limiting free riding in particular) and capacity, all central dimensions for

the delivery of public goods. It is also often thought that regional institutions are better placed

to carry forward international harmonization agendas (World Bank, 2005; Consilium Legis,

2003 also make this argument in the case of transit corridors). Regional representation can

also be a way to increase the bargaining power of its constituents in international negotiating

forums such as standard setting organizations (an example of this is for instance discussed in

EC, 2001 in relation to air and maritime transport standards). Finally, as discussed above,

pooling scarce resources can mean that regional institutions will become more efficient.

42 Another way to create ownership with businesses is by giving them access to dispute settlement under the RTA,
as NAFTA and CAFTA do for investment. Similar solutions could be envisaged in relation to trade facilitation,
offering the possibility for the private sector to challenge governments that unlegitimately restrict their business.
43 http://news.bbc.co.uk/1/hi/world/south_asia/5093712.stm


Incentives for regional reform: are RTAs naturally complements of trade facilitation?

Interestingly, RTAs have contributed to create new impediments to trade which require more

sophisticated trade facilitation measures as administration of border formalities becomes more

complex because of the need to discriminate between preferential and non-preferential trade

(De Wulf and Sokol, 2005). It takes more time to process goods covered by regional

agreements than other (Roy and Bagai, 2005). Thus, while making trade more complex,

regional agreements have created new trade facilitation needs with the enforcement of

preferential rules of origin. All this probably explains why most RTAs, to the exception of a

few deep RTAs, when they include trade facilitation aspects focus on questions of origin

(Moïsé, 2002) and their enforcement. Too often, RTAs have focused on these elements alone

rather than explicitly making of trade facilitation an element of regional integration.

On the other hand, Europe views customs (and probably border processes in general) as an

important tool in favoring regional integration and promotion of preferential links (EC, 2003).

There are indeed incentives, triggered by the trade creation effects of regional agreements,

which raise the salience of other aspects of regional cooperation (World Bank, 2005: 92) such

as trade facilitation. One has however to add that trade diversion may as well diminish the

relative importance of trade with the excluded parties to the regional agreement. One can for

instance see that the trade facilitation related policies of Europe have been largely inward

looking until most recently. This is not so much an issue when the trade volume

complementarity operates with trade facilitation policies that are not discriminatory in essence

(such as international harmonization). However it could become one with policies of

cooperation and mutual recognition or adoption of any standards that exclude certain

categories of traders.

The focus on enforcement questions in existing RTAs limits the scope of these agreements on

trade facilitation to few dimensions, which essentially aim at simplifying the job of the

customs authorities when dealing with goods under preferential access. This includes

transparency requirements with frequent reference to GATT article X (Moïsé, 2002), the

adoption of documentation standards (such as the Single Administrative Document in the EU)

to facilitate the access to information and cooperation between customs authorities for fact

finding. Harmonization is not much on the agenda, except for better mutual understanding of

day-to-day operations.

Secondly, when RTAs involve more than two partners, the question of transit management is

raised. If three countries grant each other preferential access, goods exchanged between any


two members must be able to transit through the third one without added charges and

impediment to trade that undo the preferential treatment. Thus regional trade agreements

provide a strong incentive to push forward transit agreements, either within the RTA itself or

in parallel. As discussed in section 3, despite abundance of provisions on transit,

implementation remains an issue.

The third facilitation-related element historically prominent in RTAs deals with rules on

technical standards and phytosanitary requirements. In some occasions, such as within the

European Union and the Mercosur, an ambitious regulatory agenda has been pursued.

However, in most cases standards provisions have been much more modest, including in

RTAs among advanced economies. In the case of the European Union, beyond the upward

harmonization for the internal market, the most ambitious attempts at facilitation of trade on

standards have been through standalone mutual recognition agreements (such as the one with

the USA) and are limited to conformity assessment. One lesson that seems to emerge from

regional cooperation on standards is that the nature of regional cooperation on standards

depends on the specific capacity of the trade partners such as preparedness to perform

conformity assessments, and the institutional setting of the agreement, where depending on

how strong institutions are, more or less active harmonization or recognition routes can be

followed (World Bank, 2005: 88).

The recent generation of regional trade agreements seems however to increasingly incorporate

additional trade facilitation content. Evidence of this is reported for regional agreements in

Asia that include provisions covering transparency of laws and rulings, use of ICT and e-

commerce, freedom of transit, mobility of business people, facilitation of transport and

logistics, and facilitation of payment and trade finance (Bin and Misovicova, 2007).

Thus the complementarity between RTAs and what is often their first main objective, the

reduction of tariffs and quotas, and some aspects of trade facilitation reform create an

incentive to bring trade facilitation cooperation under the umbrella of an RTA rather than in a

standalone agreement. As noted by Devlin and Estevadeordal (2004), this complementarity is

likely to increase with commercial integration, thus expanding the scope of trade facilitation

intervention in the context of RTAs. This aspect is highlighted by the case of Customs



The special case of customs unions

Arndt et al. (2007) suggest that a benefit of opting for a customs union model over a simple

free trade agreement could be to register more progress on “deep integration” issues such as

trade facilitation or liberalization of services. We find that this is most likely to be indeed the

case for trade facilitation.

The preservation of tax revenues is very much at the forefront of customs and other trade

facilitation related concerns in regional trade agreements. This is in this respect a notable

difference between cooperation agreements (which are not necessarily signed in the context

of tariff reductions) and RTAs. For developing countries and even more least developed ones

trade taxes are a big share of all government revenue (Keen and Simone, 2004). This is why

emphasis has most often only been on the few aspects of customs that would enable to secure

trade tax revenues while allowing preferential trade.

The incentives related to optimizing revenue collection are modified in a customs union,

which we first define broadly as a regional trade agreement with a common external tariff and

no tariffs among its members. 44 In a “true” customs union, revenues will be collected at the

initial port of entry in the customs union. That is revenue is collected by each member on

behalf of the union and subsequently either redistributed or spent by common institutions.

This substantive difference between customs union and other regional trade agreements bears

important consequences on the scope for trade facilitation reform at the regional level.

First, customs unions do not require the implementation of rules of origin among members.45

They also either eliminate or diminish the need for transit bond regimes for goods destined to

markets within the union. Both are significant facilitations of trade. Further, forming a

customs union implies further incentives for trade facilitation reform in member countries to

harmonize their regime, starting with adopting common customs legislations, classification,

and tariff rates.

Like in any other preferential trade arrangements there are incentives for trade deflection in

customs union, to take advantage of borders in the region where protection is the lowest.

Because tariffs are uniform, the cause of trade deflection is solely due to non-tariff barriers,

therefore putting a specific emphasis on them. Among them, lack of trade facilitation in one

44 The World Customs Organization (1995) defines a customs union as the union of two or more customs
territories sharing a common tariff, where customs duties and restrictive regulations of commerce have been
abolished within the customs union.
45 At least for “true” customs unions that collect revenue at entry.


partner provides incentives for private operators to concentrate their trading operations in the

most efficient member of the union: “port shopping” is what happened in Europe (EC, 2003).

Another characteristic of customs unions is that there is no possibility to change unilaterally

the external tariff applied to third countries, and eventually compensate tariff revenue losses

caused by trade diversion. Such revenue losses are one of the chief motives for governments

to take action.46 Additionally when revenue is collected by individual members on behalf of

the Union, and then shared under some revenue sharing agreement - as it is in SACU -

implies that there is mutual confidence in the enforcement capacity of other union members:

trade facilitation measures and upgrading of union members border management is an

element of confidence building.

Therefore in order to preserve the integrity of a union, members have a strong incentive to

take a joint approach to issues and build capacity at the weakest points of entry of the customs

territory.47 These “race to the top” incentives can promote facilitation reform in some

members. This also means that the prognosis for the provision of weakest link type public

goods is increased in customs unions, which is confirmed by their generally more ambitious

trade facilitation undertakings.48 Members of the COMESA have for example adopted several

regional trade facilitation initiatives, including a single document for customs and are in the

process of establishing a regional bond system (see box 2 for further examples in COMESA).

Deeper integration within the Customs Union adds to incentives for trade facilitation. In the

context of a more integrated markets such as the EU single market, the removal of internal

borders has meant not only a transfer of sovereignty for tariff revenue collection to the Union,

but also of all other border controls, and thus a transfer of authority at the European

Commission level, as well as further incentives to build the capacity of the weakest members

in the absence of national border controls for goods transiting through other members of the

Single European Market (EC, 1989).

Note however, that the incentive is rather one of harmonization and enforcement than

facilitation of trade in the strict sense. In this context there is a clear distinction between trade

among members and trade with third countries. While intra-union trade is most likely to see

more benefits from the removal of controls, the harmonization of regulatory requirements,

46 Ineffective protection of the domestic industry is the other reason why regional agreement members want to
avoid trade deflection. One can however assume that governments will be looking after their direct interests (tax
collection) very closely in any case.
47 See Keen (2003).
48 According to Arce and Sandler, regional institutions are already well placed to provide weakest link type public
goods compared to global institutions because being closer to the problem they can identify more easily the
“laggards”. As the weakest link problem is one of capacity, the idea seems here that for implementation purposes,
regional institutions will be more efficient.


and mutual cooperation, effects are likely to be much more ambiguous with countries outside

the Union. Border reform under customs union is only synonymous of genuine trade

facilitation to the extent that it makes trade easier among customs union partners and with

external partners. It is true that the adoption of a unique rule within the union creates

immediate benefits to third countries such as: an automatic reduction of duplication costs for

transiters and trading partners dealing with more than one customs union member and access

to a broader market once the fixed costs associated to border control (such as accessing

information about customs procedures) are paid. On the other hand, there is a distinct risk that

process of cooperation among union members in the implementation of their procedures such

as MRAs will give union economic agents an added competitive hedge against traders outside

the union that will not have access to these facilitation tools. Another risk of discrimination is

the adoption of higher standards and stricter procedures by union members on trade from

third parties leading for instance to standards harmonization and added controls at the border

to guarantee integrity of borders on behalf of other union members.

Box 2. COMESA’s regional initiatives on trade facilitation

The Common Eastern Market for Eastern and Southern African States (COMESA)
was formed in 1981 and the COMESA free trade area launched in October 2000. The
20 member countries of COMESA now plan to launch the COMESA Customs Union
in December 2008.

Trade facilitation has been an important aspect of regional integration in COMESA.
The following trade facilitation measures have been developed (not necessarily to all
member states):

• Common tariff nomenclature
• Common valuation system
• Protocol on rules of origin
• COMESA single customs declaration document (COMESA-CD)
• Protocol on transit and trade facilitation introducing licensing of transit

carriers and harmonization of axle load controls
• Regional customs bonds guarantee scheme was launched in 2006 enabling the

payment of a single bond for the region
• Licensing of clearing agents and formation of a regional Freight Forwarder

• Protocol of third party motor vehicle insurance scheme (Yellow Card scheme,

to which 13 countries participate)
• Joint border controls (Chirundu port for Zambia-Zimbabwe; Malada border

for Uganda-Kenya)
• Implementation of common standards
• Capacity building with development of customs training modules

Source: COMESA (2005)

One should also note that most customs unions do not follow the model of collection of

revenues at the port of entry, but rather according to the final point of consumption. Arndt et

al. (2007) survey 9 customs unions and find that only the EU and SACU collect revenue at


entry. This means for the other customs unions the use of complex bond procedures to ensure

that goods entering the customs union are indeed taxed at their point of destination, as well as

possibly rules of origin.49 Trust and more developed institutions are also much less a feature

of such arrangements, as are incentives to cooperate to upward harmonization.


Bergsten (1997), referring to APEC, lists trade facilitation as one of the five possible

definitions of “open regionalism”. Trade facilitation measures are indeed rarely preferential in

nature, except when they providing lower customs fees, simplified origin marking

requirements and mutual recognition agreements of conformity (Moïsé, 2002). Other reforms

undertaken in the name of trade facilitation in RTAs are de jure non-discriminatory and thus

their benefit should also extend to non-RTA trade partners (Schiff and Winters, 2003; Maur,

2005). The possibility however remains that through the adoption of specific standards, an

artificial advantage is granted to parties better acquainted than others with such standards (a

hidden motive behind upward harmonization claims promoted by developed countries). This

is an issue for technical barriers and phytosanitary standards. This probably less the case for

harmonization of customs procedures, essentially modeled on internationally agreed standards

(Moïsé, 2002). Implementation issues and necessary cooperation, including mutual

recognition of practices however raise the spectrum of discrimination against countries

outside the regional trade agreements.

What channel for regional trade facilitation? RTAs and cooperation agreements

As noted by Arvis et al. (2007) the lack of reform is not because of lack of legal instruments.

Regional dimensions of trade facilitation can be addressed through sector-specific bilateral or

plurilateral cooperation agreements rather than cross sectoral RTAs (thus confirming the view

that RTAs are not a technical necessity per se, cf. Hoekman and Kostecki, 2000). Regional

transit agreements are an example of such cooperation agreements.50 Transit arrangements

between countries are numerous. Taking the case of Africa, N’Guessan (2003) points out that

transit systems are particularly deficient. Only 30% of the regional transit in WAEMU is

conducted under the regional transit agreements, the remaining 70% being subject to bilateral

rules. Regional corridor agreements in Sub-Saharan African are largely not operational for

most of the transit traffic, and are superseded by national, non-harmonized, overlapping and

49 I thank Matthew Stern for drawing my attention to this fact. For instance in the EAC, a transitory regime will
ensure that collection of taxes is first made at the final port of destination, as opposed to the port of arrival
(Uganda DTIS, 2006).
50 There are several other examples of (generally bilateral) sector specific agreements to facilitate trade: customs
cooperation agreements and mutual recognition agreements for instance.


discriminatory provisions such as: compulsory customs escort, non-harmonized transit

charges, or specific country documentation for transit. Administrative burden adds to the

regulatory burden, with lack of coordination between different agencies in charge of

controlling transit goods (customs, police, sanitary controls). Inefficiency of these agencies

multiplies the costs. The conclusion thus seems that neither bilateral sectoral agreements nor

regional agreements in the context of regional integration have delivered trade facilitation

benefits in Africa.

One reason why bilateral transit cooperation agreements have not delivered could be the

absence of incentives by countries to internalize regional externalities: the asymmetry of

incentives for landlocked and border countries noted by Schiff and Winters (2002). However,

and more importantly, the lack of political commitment of individual countries behind

market-based reforms probably remains the biggest impediment. An example is absence of

liberalization in the transport sector denying the benefits of better transport links and

corridors. Poor procedures in individual countries of transit, linked to weak political

incentives to reform, remain often the main obstacle to efficient regional transit regime. An

interesting case in hand is Djibouti, which has made important port infrastructure investments

to increase its capacity to serve neighboring countries from its hinterland. To date, the

capacity of the port terminals remains in some instances largely underutilized. The reason is

not lack of demand: the Djiboutian port authority has had to increase its dry port capacity to

store increasing amounts of cargo laying in waiting. Slow border clearance from neighboring

Ethiopian authorities is the explanation. Cargo ships have to stay longer than necessary,

preventing full utilization of the terminals. Slow clearance from Ethiopian authorities creates

a negative externality on port activities in Djibouti, probably an important source of revenue

for a small economy like Djibouti. This demonstrates incidentally that the policies of

landlocked countries can also negatively impact on coastal neighbors.

The political commitment problem can be better addressed in the broader setting of RTAs,

than in other regional institutional settings. The anchoring of reform can be stronger, and the

cost of non-implementation of one obligation can jeopardize the agreement as a whole, and

thus is much higher. Dispute settlement measures tend to be stronger in regional trade

agreement than in sectoral agreements (World Bank, 2005: 86 provides the example of

standards). This could explain why the alternative to incorporating transit processes into or in

parallel regional agreement seems to have gained credence in the past decade, when several


agreements have been signed with provisions on regional transit.51 Indeed, many regional

agreements have incorporated rules and developed instruments to facilitate transit such as the

EAC, COMESA or SADC. However, actual transit facilitation has been disappointing mainly

because of poor implementation (Arvis, 2005 quoting UNCTAD, 2001). With the exception

of the European TRIE agreement (Transit Routier Inter Etats) most fail to address

implementation problems and have confined themselves to broad policy recommendations.

The poor record on transit illustrates the more general failure of most regional institutions to

deliver tangible trade facilitation reform. The EU, and to a lesser extent the APEC being

among the exceptions. In Africa, Mc Tiernan (2006) reports that only COMESA and the

trans-Kalahari corridor (an ad hoc transit cooperation agreement) have provoked changes in

customs practice.52 Interestingly, COMESA, the EU and APEC are three non traditional cases

compared to other RTAs. The poor performance of RTAs may seem unexpected. : since

RTAs have clearly not tried to take a rule setting role - left to international organizations53 -

they could have been expected to take a more active role in implementation. Moreover, that

RTAs have kept clear of the trade facilitation standard setting agenda is a good thing as it

guarantees MFN treatment: the modern standards adopted are accessible to all. Because they

are defined internationally, and thus one can presume that domestic interests are not

influencing their design, international standards should also guarantee National treatment.

The fact that RTAs have not delivered reform in practice needs therefore be questioned

further. A first remark is that the role of binding rules for implementation is perhaps

overstated. Finger (2006) reminds us that there is not necessary a need for legal obligation to

spark trade facilitation reform as many countries have already shown willingness to do so.

One should however wonder whether the willingness to undertake regional initiatives is the

same. APEC has no enforcement mechanism and commitments are entirely voluntary. APEC

has emphasized its focus on policy integration and common principles of reform, while

organizing the delivery of technical assistance.

A second observation relates, as discussed earlier, to the revenue preservation incentives in

regional integration. Policies in the transit country are likely to be motivated by tax revenue

preservation and since revenue collection is often not perceived as complementary to trade

facilitation, a plethora of controls will be the preferred option. Weak regional institutions also

51 Most regional trade agreements incorporate Transport and Trade Facilitation Agreements types (World Bank,
52 Although the trans-Kalahari corridor progress have been hampered by a unilateral decision by Botswana to
increase road user fees, which resulted in a decrease of traffic (World Bank, 2005).
53 The World Customs Organisation Kyoto Convention for customs, the International Civil Aviation Authority
(ICAO), the International Road Transport Union (IRU), the International Maritime Organisation (IMO), etc.


mean that implementation does not necessarily follow transit agreement that have been

negotiated. On the contrary, in a true customs union because revenues are collected on behalf

of the union, transiting countries have some incentives this time to facilitate transit trade for

countries that collect part of their revenue.

Another difference with customs union is often the strength of their institutions which allows

them to carry more ambitious reforms. Looking at the examples of the EU and COMESA it

looks certainly the case that bold regional facilitation initiatives have been tabled like the

regional bond scheme in COMESA, implemented in 2006, and numerous policies in Europe

such as trans-national information and transport networks, or standards harmonization.

Whether such policies have resulted in effective facilitation of trade is a question that is

difficult to answer. While many of these initiatives seem desirable, their implementation

remains largely unstudied and so is their impact. Besides there is also some evidence that

these policies have been more complex to implement that initially thought, short of

expectations in some cases (transit in COMESA), and that in the instance of the European

Union, many of the higher international standards remain to be implemented.

A last remark is the presence of large advanced countries as a driving force (both political and

material) behind reform. APEC (Japan, Australia), COMESA (South Africa), and of course,

European countries. As Schiff and Winters (2003) argue, trade facilitation as a dimension of

policy integration requires much more than non-discrimination. It needs the rapprochement of

policies and enforcement; it also needs effective implementation. Making policies more

compatible undoubtedly involves strong political will to fight against the vested interests in

border agencies.54 It also requires a different institutional setting that classical exchange of

trade concessions. In short, stronger and more permanent institutions than those of

multilateral trade negotiations are needed because implementation of trade facilitation

reforms will require at least coordination, and more likely, as discussed, harmonization and

mutual recognition of rulings. Regional cooperation agreements and RTAs seem more likely

to deliver institutions that will foster integration, mostly because transaction costs are smaller

among a few countries than multilaterally (and in general for the delivery of public goods,

Devlin and Estevadeordal, 2004; Sandler, 2006); in particular regional agreements seem to

offer the flexibility needed for the design of such institutions and for informal cooperation.

RTAs further offer compared to sector-specific trade facilitation agreements the benefit of

stronger political commitment and linkages with several policies facilitating trade. Finally,

RTAs offer enforceability and resources for implementation. Enforceability of reform

54 This is for instance a clear finding from a recent study on the needs, priorities and costs for the implementation
of a future WTO Trade Facilitation Agreement (McLinden, 2006).


commitment does not seem to be such an issue in the case of trade facilitation as the example

of voluntary commitments with APEC show.55 Implementation of reform on the other hand

seems a major challenge for most countries, including richer ones. Availability of funds, time

and expertise is required for the most ambitious reforms. Again, RTAs including advanced

countries seem well suited to fulfill capacity building needs, provided that they offer the right

mix of financial assistance and expertise.

The complementarity of regional solutions with multilateral and national ones

Regional solutions to trade facilitation should not always be thought as substitutes to

multilateral, national or other regional interventions but also as complement. Transit corridors

should probably gain from the involvement of regional trade agreements on top of corridor

specific institutions such as regional management committees, and more national

arrangements (Consilium Legis, 2003).

It is also in several instances a fact, that in order to become operational, regional cooperation

has to rest on international rules and institutions. This is because the need for guarantees

extending beyond the regional level: we gave earlier the example of accreditation, which

needs to be guaranteed internationally by one of two global organizations: the International

Accreditation Forum and the International Laboratory Accreditation Organization (Holmes et

al., 2006). Likewise the integrity of TIR system is guaranteed by the UN Economic

Commission for Europe and the International Road Transport Union (Arvis et al., 2007).

Another dimension of complementarity of regional integration agreements and donor

organizations is likewise illustrated by the Trade and Transport Facilitation in South East

Europe program, a joint venture between the EU and the World Bank and bilateral donor

countries.56 Of course, the private sector, already mentioned, which may operate nationally,

regionally or globally, is another desirable stakeholder.

Which RTA? How the size of partner countries matters

Many aspects discussed so far are country specific. Country or country grouping

characteristics have therefore significant implications for regional trade facilitation. Least

Developed countries face a set of constraints that require particular attention. First, economic

size, meaning that economies of scale are less likely to be internalized at the national level,

55 Another enforcement issue is between government and private operators, and the need to have fair and
consistent enforcement of rules, and the right to appeal procedures for traders.
56 http://www.seerecon.org/ttfse/


thus suggesting that small countries should team up together to deliver regional trade

facilitation instruments (witness the regional enquiry point proposed by Barbados). Size also

implies that the bargaining position of small countries with more powerful neighbors leads to

suboptimal outcomes for them. The obvious example is transit. LDCs are over-represented

among landlocked countries. A solution to this size asymmetry would perhaps be for the

small country to trade-off concessions in an RTA context by offering market access (in

services sector for instance) against facilitated transit and better enforcement of rules so that

transit is not discriminated against (this includes competition in transport). Over-dependence

on tax revenues is another characteristic of LDCs compared to more advanced developing and

developed countries. This dependence creates strong incentives against trade facilitation, and

in particular against sharing any border responsibility with neighboring countries. In this

respect, an asymmetric regional cooperation may be an advantage, as partnering with a large

country that is less dependent on trade tax revenue will mean more political support for trade

facilitation reform.

This raises the question of whether there are specific partners with whom to preferably

implement trade facilitation reform? There is the complementarity between some aspects of

trade facilitation and the volume of goods trade: transport infrastructure springs to mind.

Modernization of trade procedures involves one off fixed costs that will be recouped more

easily over larger volumes of trade. This obviously applies to the case of transit for

landlocked countries where often a few transit routes will represent a very large proportion of

all external trade.

The volume of trade criterion however may fail to address the geographical-determined

dimensions of trade facilitation as for many countries trade with neighbors is actually too low,

thus shifting the focus of regional trade where transaction costs are already low, and possibly

reinforcing existing distortions (Schiff, 2001).57 A trade potential criterion makes more sense

and thus incorporates the geography determinants of trade facilitation discussed.

A different question is whether regional cooperation would not be better with partners

representing a large proportion of world trade? This complementarity can arise for some

norm-based dimensions of trade facilitation reform: it makes more sense to harmonize

standards with big world traders as this also means harmonization with this partner’s trade

partners. In the context of the positive externalities described earlier, it may make sense to

57 See for instance Al-Atrash and Yousef (2000) on Arab trade. Another example is the share of trade between
India and its neighbours.


integrate with hubs at the centre of large trade networks.58 In a world where production

processes become more fragmented, this seems the path to follow.59 The share of world trade

criteria points out for developing countries towards seeking trade facilitation reform in the

context of agreements with Northern economies. This presents the added benefit of access to

higher modern standards, technical assistance and capacity building: an important

consideration for weak link type public goods. Actually, evidence suggests that RTAs

involving developed economies contain more detailed and sophisticated trade facilitation

provisions (Bin and Misovicova, 2007).

The above arguments need however to be qualified with the possible negative effects of

RTAs mentioned earlier, in particular if trade facilitation measures boost trade diversion

effects. This seems to reinforce in our view the importance of promoting regional trade

facilitation measures in the context of regional trade agreements that are not trade diverting to

start with: an additional argument for choosing partners with a large share of the world’s


Finally, there is also the risk that countries with low capacity may not be able to share much

with countries with sophisticated policies in place, or be invited to implement measures

beyond what is strictly necessary for them.

Disentangling these dimensions is difficult as there is no automatic prescription as to where

the most benefits are concentrated and whether any piecemeal approach to trade facilitation is

advisable. Regarding the latter, I think better not. This thus suggests that a better framework

for addressing trade facilitation issues might need to rely on either a multilateral approach

with regional extensions or a hub-an-spoke approach such as the one envisaged in the

European Partnership Agreements.


Regional interventions matter. Regionalism, specifically regional trade agreements, has

undoubtedly a role to play to help trade facilitation reform. We find several reasons why this

should be the case:

58 As pointed out by one reviewer, does it make sense for Mercosur to have an agreement on multi-modal transport
(signed in 1994) among its countries but none with the EU and the US, its main trade partners?
59 This echoes a point made by Devlin and Estevadeordal (2004) who suggest that such “hub and spoke”
arrangements allow getting a greater geographical reach and realising economies of scope in the delivery of similar
multiple public goods to different regions.


- The existence of regional public goods, because of the presence of economies of scale

and externalities at the regional level, for instance in the exchange of electronic

information, or the mutualization of guarantee systems and transit.

- The “local” nature of several elements of the trade logistics chain and the inherent

efficiency of reforming them at the regional level.

- The adequacy of regional trade agreements, if appropriately designed, for

implementation of regional tools facilitating trade: this is because they are well

designed to address both coordination and capacity constraints facing the provision of

regional trade facilitation public goods. This means that regional trade agreements

can deliver several, different, and complementary trade facilitation public goods.

- Another advantage of regional trade agreements is when they incorporate deep

integration dimensions. They are better suited for addressing complex regulatory

liberalization. In addition, because they can be broad in scope across many policies,

trade-offs between the many components of the trade logistics chain can be made,

and offer better safeguards against violations of commitments.

Regionalism can deliver more. This said, there is yet scant evidence of success of regional

agreements promoting effective trade facilitation. The reasons why this is the case are still not

well identified and would warrant further research. Political economy consideration certainly

play a very important role here; one of the reasons behind low political will could well be the

reliance on trade tax revenue, a very sensitive issue for poor country especially. There might

thus be strong resistance to facilitate trade with regional partners. Incentives for better transit

arrangements are quite difficult to address in this respect.

What scope for regional cooperation? RTAs need to shift their focus from reciprocity to

policy integration and from a narrow vision of border enforcement to one of policy of

integration into world markets. This entails a broader coverage of trade facilitation issues as

whole (recent trade agreements seem to move in that direction).

In particular, trade facilitation objectives should be integrated in RTAs with a clear objective

to make them also more open to third country trade. As we know too well, trade diversion

means that regional trade agreements are not necessarily guaranteeing welfare gains for

participants in the agreement, and likely to generate welfare losses for countries excluded.

Regionalism is also not necessarily a stepping stone towards multilateral trade liberalization.

Trade facilitation reform can be a force for good in this context if it reduces trade transaction

costs on trade from all origins. While we offer no clear conclusion as to whether facilitation is

itself a stepping stone, it indeed looks like a useful and relatively straightforward policy


(probably easier to implement from a political point of view than tariff reform) to moderate

the trade diversion effects of preferential tariff reduction.60

We also saw that the diagnostic about the need regional trade facilitation reform is going to

differ depending on the absolute and relative size of countries under consideration. While

some criteria helping to discern how regional trade facilitation could be delivered were

identified, further work is needed to understand when the political economy incentives (in a

way the demand for regional public goods) can be catered for so as to result in optimal supply

of regional public goods.61

There is also little evidence on the impact of the most successful regional integration

agreements to guide policy makers on which regional trade facilitation policies may be

desirable. In particular, the study of existing regional agreements could bear some lessons on

whether there is a case for gradual approach, for prioritizing specific trade facilitation

regional public goods, leaving others for later and building upon that basis as seems to have

been the case in Europe.

Strong institutions are required. The crucial role played by institutions governing regional

trade liberalization efforts must be emphasized. Their role seems to matter as much – if not

more – in the implementation of the soft law aspects of reform than of hard rules. We

highlighted the importance of cooperation and joint regulatory design to deliver regional trade

facilitation solutions. This also highlights an area of comparative advantage of regional

approaches over multilateral ones. Weak institutions also are not well equipped to manage

heavy implementation challenges.

Regional institutions also need to be strong enough to push what are difficult reforms:

customs unions seem relatively more efficient. Emphasis should be shifted from rules to

implementation, with flexibility in mind, and possibly complementarity with other national or

multilateral approaches. This may mean strengthening regional specialist organization,

eventually in partnership with the private sector, which have tended to be more successful in

cooperation agreement or maintaining working groups of experts. Finally, Customs unions

have fared better than other forms of regional cooperation in many respects. Admittedly,

customs unions are few, and forming a customs union requires a serious amount of political

will on all sides: these are naturally good conditions for regional trade facilitation reforms too.

60 Helping the RTA to maintain the level of trade with third parties to its pre integration level, otherwise known as
the Kemp-Wan (1976) proposition.
61 For instance how exactly is coordination delivered? What makes countries that would compete with each other
for the supply of trade facilitation infrastructure change their mind?



Al-Atrash, Hassan and Tarik Yousef, 2000. “Intra-Arab Trade: Is It Too Little?”, IMF
Working Paper No. 00/10, Washington: International Monetary Fund.

Anderson, James E. and Eric van Wincoop, 2002. “Borders, Trade and Welfare”, in Dani
Rodrik and Susan Collins (eds.), Brookings Trade Forum 2001, Washington DC: Brookings
Institution, 2002.

Arce, Daniel G., and Todd Sandler, 2002. Regional Public Goods: Typologies, Provision,
Financing, and Development Assistance. Stockholm: Almqvist and Wiksell International for
Exert Group on Development Issues, Swedish Ministry of Foreign Affairs.

Arndt, Channing, Xavier Cirera, Lawrence Edwards, Frank Flatters, Robert Kirk, Mareike
Meyn, Thomas Rutherford, Matthew Stern, and Christopher Stevens (2007). Evaluation of An
Appropriate Model for a SADC Customs Union, Report commissioned by the SADC
Secretariat, final draft, 7 August 2007.

Arvis, Jean-Francois, Gael Raballand and Jean-Francois Marteau, 2007. “The Cost of Being
Landlocked: Logistics Costs and Supply Chain Reliability”, World Bank Policy Research
Working Paper 4258.

Arvis, Jean-Francois, 2005. “Transit and the Special Case of Landlocked Countries” chapter
11 in Luc De Wulf and Jose B. Sokol (eds) Customs Modernization Handbook, Washington
DC: World Bank.

Baldwin, Richard, 2000. Regulatory Protectionism, Developing Nations and a Two-Tier
World Trade System, Brookings Papers on Economic Activity.

Bergsten, Fred, 1997. "Open Regionalism", World Economy, 20: 545-65.

Bin, Peng and Maria Misovicova, 2007. Trade Facilitation Provisions in Regional Trade
Agreements in Asia and the Pacific, mimeo UN Economic Commission for Asia and the

Caron, Jim and Heidi Reichert, 2001. East Africa Subregion: Enhancing Transportation
Management to Foster U.S. Agricultural Trade Opportunities, Washington DC: Agricultural
Marketing Service, US Department of Agriculture.

COMESA, 2005. COMESA’s Practical Experience on Trade Facilitation in Relation to
Regional Economic Integration and Development, paper presented at the “Expert Meeting on
Trade Facilitation as Engine for Development”, Geneva Switzerland, 21-23 Septembre 2005.

Consilium Legis, 2003. Transport and Trade Facilitation: East and Southern Africa: Review
of Present Problems and Reform Initiatives, final report volume 1, Washington DC: The
World Bank.

Cook, Lisa D. and Jeffrey Sachs, 1999. Regional public goods in international assistance, in
Kaul, I., Grunberg, I. and Stern, M., (Eds.), Global public goods: international cooperation in
the 21st century, 436−449. New York: Oxford University Press.

Devlin, Robert and Antoni Estevadeordal, 2004. “Trade and Cooperation: A Regional Public
Goods Approach”, in Antoni Estevadeordal, Brian Frantz and Tam Robert Nguyen (eds.)
Regional Public Goods, Washington DC: Interamerican Development Bank and ADB.


De Wulf, Luc and Jose B. Sokol, 2005. Customs Modernization Handbook, Washington DC:
World Bank.

European Bank for Reconstruction and Development (EBRD), 2003. Regional Trade
Facilitation Programme: Fostering Transition Through Documentary Credit, Evaluation
Department, EBRD.

European Commission (EC), 1989. Working Document of the Commission Services on the
Customs Union in the Context of the Internal Market, SEC(89) 1942, Brussels: European

European Commission (EC), 2003. Communication from the commission to the council, the
European parliament and the European economic and social committee a simple and
paperless environment for customs and trade, COM(2003) 452, Brussels: European

Ferroni, Marco, 2002. “Regional Public Goods in Official Development Assistance”,
INT/INTAL Working Paper, Washington D.C.: Inter-American Development Bank.

Finger, Michael J., 2006. “Aid for Trade: How We Got Here, Where We Might Go”, ILEAP
Background Brief No. 10.

Ghana National Development Planning Commission, 2005. Growth and Poverty Reduction
Strategy (GPRS II) (2006-2009), Republic of Ghana, November.

Hirshleifer, Jack, 1983. “From Weakest-Link to Best-Shot: The Voluntary Provision of Public
Goods.” Public Choice, 3:271-286.

Hoekman, Bernard and Michel Kostecki, 2000. The Political Economy of the World Trade
System, Oxford: Oxford University Press.

Holmes, Peter, Leornado Iacovone, Rungroge Kamondetdacha, and Lara Newson, 2006.
“Capacity Building to Meet International Standards as Public Goods”, UNIDO Working

Kaul, Inge, Pedro Conceição, Katell Le Goulven and Ronald U. Mendoza, 2003. Providing
Global Public Goods: Managing Globalization, Oxford: Oxford University Press.

Kanbur, Ravi, Todd Sandler, and Kevin M. Morrison, 1999. “The future of development
assistance: Common pools and international public goods”, Overseas Development Council
Policy Essay No. 25, Washington, DC.

Keen, Michael, 2003. Changing Customs: Challenges and Strategies for the Reform of
Customs Administration, Washington DC: International Monetary Fund.

Keen, Michael, and Alejandro Simone, 2004, “Tax policy in Developing Countries: Some
Lessons from the 1990s, and Some Challenges Ahead” in Sanjeev Gupta, Benedict Clements,
and Gabriela Inchauste (eds.): Helping Countries Develop: The Role of the Fiscal Policy.
Washington DC: International Monetary Fund.

Kemp, Murray C. and Henry Y. Wan Jr, 1976. “An Elementary Proposition Concerning the
Formation of Customs Unions”, Journal of International Economics, 6: 95-97.

Kimura, Fukunari and Hyun-Hoon Lee, 2006. The Gravity Equation in International Trade in
Services, Review of world economics, 142(1): 92-121.


Mattoo, Aadittya and Carsten Fink, 2002. “Regional Trade Agreement and Services: Policy
Issues”, World Bank Policy Research Working Paper 2852.

Maur, Jean-Christophe, 2005. “Exporting Europe’s Trade Policy”, The World Economy,
28(11): 1565-1590.

Mc Linden, Gerard, 2006. Needs, Priorities and Costs Associated with Technical Assistance
and Capacity Building for Implementation of a WTO Trade Facilitation Agreement: A
Comparative Study Based on Six Developing Countries, Washington DC: World Bank.

Mc Tiernan, Andrew, 2006. Customs and Business in Africa: A Better Way Forward
Together, draft report prepared for Business Action for Improving Customs Administration in
Africa (BAFICAA).

Moïsé, Evdokia, 2002. The Relationship Between Regional Trade Agreements and
Multilateral Trading System: Trade Facilitation, OECD Working Party of the Trade
Committee TD/TC/WP(2002)17/FINAL.

Moïsé, Evdokia, 2005. “Trade Facilitation Reforms in the Service of Development: Country
Case Studies”, OECD Trade Policy Working Paper No. 12, Paris: OECD.

Müller-Jentsch, Daniel, 2003. “Transport Policies for the Euro-Mediterranean Free-Trade
Area: An Agenda for Multimodal Transport Reform in the Southern Mediterranean”, World
Bank Technical Paper No. 527.

National Board of Trade, 2003. Trade Facilitation From a Developing Country Perspective,
Kommerscollegium/National Board of Trade: Stockholm.

N’Guessan N’Guessan, 2003. La problématique de la gestion intégrée des corridors en
Afrique, Document d’analyse SSATP No. 3F, Programme de politiques de transport en
Afrique subsaharienne Banque mondiale et Commission économique pour l’Afrique.

OECD, 2005. A Review of Past Technical Assistance and Capacity Building Initiatives for
Trade Facilitation, OECD DAC Project on Trade Facilitation phase I, Paris, OECD.

Otsuki, Tsunihero, John Wilson, and Mirvat Sewadeh, 2001. “What price precaution?
European harmonisation of aflatoxin regulations and African groundnut exports”, European
Review of Agricultural Economics, 28(3): 263-284.

Portes, Richard and Hélène Rey, 2005. The Determinants of Cross-Border Equity Flows,
Journal of International Economics, 65: 269-296.

Röschlau, F. 2003. “Double Standards Cause Regional Transport Nightmares”, e-corridor,
Walvis Bay Corridor Group, number 3.

Roy, Jayanta and Shweta Bagai, 2005. “Key Issues in Trade Facilitation: Summary of World
Bank/ EU Workshops in Dhaka and Shanghai in 2004”, Policy Research Working Paper
3703. World Bank, Development Research Group, Trade, Washington, D.C.

Rufin, Carlos, 2004. “Regional Public Goods and Infrastructure”, in A. Estevadeordal, B.
Frantz and T. R. Nguyen (eds), Regional Public Goods: From Theory to Practice,
Washington DC: Inter-American Development Bank.

Sandler, Todd, 1998. “Global and regional public goods: A prognosis for collective action”,
Fiscal Studies 19: 221-247.



Sandler, Todd, 2006. “Regional Public Goods and International Organizations”, Review of
International Organizations, 1(1), March 2006.

Sauvé, Pierre and Americo B. Zampetti, 2000. “Subsidiarity Perspectives on the New Trade
Agenda”, Journal of International Economic Law, 3(1): 83-114.

Schiff, Maurice, 2001. Will the “Natural Trading Partner” Please Stand Up? Journal of
Economic Integration, 16(2): 245-62.

Schiff, Maurice and L. Alan Winters, 1998. “Regional Integration as Diplomacy”, World
Bank Economic Review, 12: 271-295.

Schiff, Maurice and L. Alan Winters, 2002. “Regional Cooperation and the Role of
International Organizations and Regional Integration”, Policy Research Working Paper 2872.
World Bank, Development Research Group, Trade, Washington, D.C.

Schiff, Maurice and L. Alan Winters, 2003. Regional Integration and Development,
Washington DC: The World Bank.

Stålgren, Patrick, 2000. “Regional Public Goods and the Future of International Development
Co-operation: A Review of the Literature on Regional Public Goods”, Expert Group on
Development Issues Working Paper 2000: 2, Ministry for Foreign Affairs, Stockholm,

Mozambique DTIS, 2004. Removing Obstacles to Economic Growth in Mozambique: A
Diagnostic Trade Integration Study, volume 2 main report, Integrated Framework, Geneva:

Tanzania DTIS, 2005. Tanzania: Diagnostic Trade Integration Study: Volume 1, Integrated
Framework, Geneva: WTO.

Uganda DTIS, 2006. Uganda: Diagnostic Trade Integration Study: Volume 1, Integrated
Framework, Geneva: WTO.

Wilson, John S., Catherine Mann, Yuen Pau Woo, Nizar Assanie, and Inbom Choi, 2002.
Trade Facilitation: A Development Perspective in the Asia Pacific Region, APEC.

World Bank, 2004. The World Bank’s Trade Logistics Initiative: From Country Awareness to
Global Advocacy, from International Debate to Local Implementation, September 2004.

World Bank, 2005. “Beyond Trade Policy Barriers: Lowering Trade Costs Together”,
Chapter 4 in Global Economic Prospects 2005: Trade, Regionalism and Development,
Washington DC: The World Bank.

World Customs Organization (WCO), 1995. Glossary of International Customs Terms,
Brussels: WCO.

Yang, Yongzheng and Sanjeev Gupta, 2005. “Regional Trade Agreements in Africa: Past
Performance and the Way Forward”, IMF Working Paper WP/05/36.