(PDF) Continental Free Trade Area Compensation Mechanisms (Draft)

Compensation Mechanisms across the CFTA Member

States

Dr Nick Charalambides

Imani Development

September 2017

Draft

DISCLAIMER

This study has been prepared with the financial assistance of the European Commission and

was conducted through the GFA Consulting Group GmbH in partnership with Nathan

Associates and DNA Economics. The views expressed herein are those of the consultants

and therefore in no way reflect the official opinion of the European Union.

Table of Contents
Abbreviations........................................................................................................................................
Foreword iii
Acknowledgements.............................................................................................................................
Executive Summary.............................................................................................................................
Introduction.....................................................................................................................................
1 Fiscal revenue loss and broader concerns over liberalisation....................................
2 Accommodating concerns in the CFTA.........................................................................
3 The focus of the paper.....................................................................................................
The Fiscal Impact of Tariff Liberalisation in an FTA.....................................................................
1 The literature review........................................................................................................
1.1 Methodologies used............................................................................................4
1.2 Assumptions, Risks and Methodological Issues.................................................5
1.3 Brief review of the data used...............................................................................5
2 Fiscal Impact of Trade Liberalisation: summary of findings........................................
2.1 The simulated potential impact of the CFTA.......................................................6
2.2 Estimating the potential CFTA fiscal impact........................................................7
2.3 Insights on revenue loss from the broader literature...........................................8
2.4 Revenue recovery...............................................................................................9
3 Conclusions on tariff revenue loss and recovery.......................................................
Fiscal loss compensation mechanisms and adjustment facilities............................................
1 Brief discussion of the literature reviewed..................................................................
2 FLCMs as distinct from Adjustment/Regional Development Funds..........................
3 Summary of Findings.....................................................................................................
3.1 Financing Structure...........................................................................................11
3.2 Management, governance mechanisms........................................................... 12
3.3 Funding Allocation Criteria................................................................................13
3.4 Operational Period............................................................................................ 13
Conclusion and recommendations..............................................................................................
Bibliography..................................................................................................................................
Annexes18
1 Fiscal Impact of Tariff Liberalization: Literature Table...............................................
2 Tabulation of Findings: Characteristics of Fiscal Loss Compensation 
Mechanisms in Regional Trade Agreements...................................................................................
i





 !"#!!$

%&'(&

ii

1 Introduction

The Continental Free Trade Area (CFTA) will bring together 54 African countries with a

combined market of $3.4 billion of over one billion people. By promoting trade integration,

the CFTA can play a major role in structural transformation and inclusive growth across

Africa.

Currently the potential for regional trade is going largely unrealised1. There is great untapped

potential owing to limited trade occurring between regional economic communities (RECs).

At continental level, the biggest economies trade at non-preferential tariffs. As a result, they

often have better market access when exporting under the United States of America’s (USA)

African Growth and Opportunity Act (AGOA) or the EU’s Everything but Arms (EBA) and

Economic Partnership Agreements (EPAs) than to each other. Within the RECs, only the

EAC has achieved full tariff liberalisation. And, despite significant progress, non-tariff barriers

have remained a challenge2. Deeper and wider integration in Africa is needed. As stated by

President Kenyatta, “There cannot be a good reason why it is easier for us to trade with

Asia, Europe and the America’s rather than with fellow Africans”3.

1.1 Fiscal revenue loss and broader concerns over

liberalisation

While recognising the potential of the CFTA to drive growth, the AUC also recognises that

tariff reductions can have a significant impact on smaller, poorer countries. There are

concerns regarding (a) adjustment costs and (b) de-industrialisation/ constraints to

industrialisation.

(J. Francois, 2011) characterises the costs of adjustment for the private and the public

sectors as follows:

Private adjustment costs:

oLabour

Unemployment

Lower wages

Obsolescence of skills

Training costs

Psychological costs

oFirms

Obsolete machinery

Underutilised capital

Transaction costs of investment in new sectors

Public Sector costs:

oLower trade taxes

oSocial safety net expenditure

oImplementation costs of trade reform

While trade liberalisation is a key driver of efficiency gains and can be an important tool for

development and industrialisation in developing countries (UNECA, 2015) there are

1 Trade Integration and Structural Transformation, UNECA 2015.

2 Assessing Regional Integration in Africa, VII, UNECA, 2016.

3 Towards One African Market, AU, June 5 2015.

1

concerns that premature liberalisation has resulted in de-industrialisation4 and inadvertently

put smaller, poorer countries on a lower growth path (Shafaeddin, 2010). Trade liberalisation

can contribute to de- industrialisation by enabling a surge in lower price imports. Lower tariffs

result in foreign companies grabbing too much of the market too quickly for negatively

affected companies to be able to respond (and improve their own competiveness). Concerns

about trade liberalisation and industrialisation also relate to the difficulties of young firms in

new sectors – not able to develop the experience and market share to be able to grow when

they face competition with larger and more mature foreign companies.

1.2 Accommodating concerns in the CFTA

The range of concerns relating to adjustment costs and de-industrialisation can be

accommodated in the CFTA through well-established trade policy instruments and support

measures, as set out in the table below.

Challenge Trade Policy Instrument

(short term)

Support Measure

Adjustment Costs

Private Sector:

Import competition

resulting in

unemployment,

lower wages and

bankruptcy.

-

Safeguard measures: to

protect against surges in

imports that will damage a

domestic sector, tariffs can be

temporarily imposed, subject

to conditions.

Within a specified time period, the

affected sector will need to be

supported in becoming more

competitive (as protection is

temporary).

Where sectors prove not to be

competitive, assistance is focused

on retraining labour and facilitating

exit and re-investment in growth

sectors.

Unfair competition

being used by

exporters to

enhance their

market position.

Anti- dumping: to protect

against exporters’ anti-

competitive practices, anti-

dumping duties can be

imposed if a domestic sector

is being harmed (or

potentially harmed). These

duties can be imposed

subject to conditions.

Over the longer term the anti-

competitive practices must be

proved, or duties will be

refunded.

Assistance can be provided to

enhance co-operation on

competition and competition policy.

Public Sector

Lower trade taxes

Fiscal loss compensation:

Governments of smaller

economies can be directly

compensated for their loss in

tariff revenue, on a temporary

basis.

Initial rationalisation of tariff regime

(reducing exemptions and

translating quotas to tariffs) can

mitigate the initial impact.

Over the longer term, adjustment

needs to be part of a wider tax

4 Though see also http://www.ifpri.org/blog/myth-de-industrialisation-sub-saharan-africa.

2

Challenge Trade Policy Instrument

(short term)

Support Measure

There is also the option of

allowing affected countries to

re-impose, subject to

conditions, tariffs as a

temporary measure.

reform to shift the tax base to other

sources. The use of VAT to replace

lost tariff revenue has been

particularly prevalent.

Social safety net

expenditure

Either cushioned in the short

run by a dedicated fund or

often by donor financing.

Priority given to expenditure

on social safety nets during

adjustment.

Generally assumed that

expenditure needs are transitional.

Implementing costs

of trade reform.

For smaller, poorer countries

adopting reform, adjustment

support in the form of aid for

trade is often required.

Institutional development and

reform is often required to ensure

reforms are sustainable.

De-industrialisation

Import competition

resulting in the

industrial sector

contracting.

Safeguard measures: to

protect against surges in

imports that will damage a

specific industrial sector,

tariffs can be temporarily

imposed, subject to

conditions.

Within a specified time period, the

affected sector will need to be

supported in becoming more

competitive (as protection is

temporary).

Import competition

constraining the

development of

identified strategic

industrial sectors.

Infant industry protection:

tariffs can be imposed to

protect emerging sector,

subject to conditions and as

agreed.

Within a specified time period, the

infant industry will need to be

supported to become competitive

(as protection is temporary).

1.3 The focus of the paper

The purpose of this study is to assess the scope for addressing one of the concerns for the

FTA – lower trade tax revenue. It looks at the potential of that fiscal loss, and reviews Fiscal

Loss Compensation Mechanisms (FCLM) in Africa and more broadly.

The challenge of fiscal loss is particularly relevant in the context of the CFTA given that

several Member States rely on trade taxes for a large part of tax revenues. Addressing the

fiscal challenge is, of course, very important for continuing expenditure on human

development. But it is also important for the implementation of liberalisation commitments.

Several countries have delayed tariff phase downs because of fears of revenue loss (OECD,

2013), (SATH, 2012)

3

2 T h e F i s c a l I m p a c t o f T a r i f f L i b e r a l i s a t i o n

in an FTA

2.1 The literature review

The studies selected were chosen because of: relevance to the CFTA; developing-country

focus; use of recent datasets; author (widely cited, affiliated to leading institutions, etc.);

specific referral from our client and/or affiliate network; and effective synthesis of existing

work.

Six of the studies deal expressly with tariff revenue impacts of the CFTA. Others review the

experience with tariff liberalisation in Africa and more broadly. They capture the impact of

both unilateral and regional liberalisations.

2 . 1 . 1 M e t h o d o l o g i e s u s e d

The studies include both ex post and ex ante analysis – the majority, however, are ex-ante.

Ex-ante studies assess the likely impact of different scenarios for tariff reduction. They do so

using models of the entire economy (general equilibrium) or a sector (partial equilibrium).

The studies reviewed examine various liberalisation scenarios, of varying timeframes, with a

range of partial and general equilibria models. For example, Jensen & Sandrey (2016) use a

CGE model to simulate fiscal impacts of 3 liberalisation scenarios against a baseline year of

2025; while Mevel and Karingi (2012) use a similar approach to simulate full liberalisation

impact against a baseline of 2022. All studies of the CFTA by necessity use a computable

equilibrium approach as the FTA is still under negotiation.

Ex-post econometric studies try to assess the actual impact of an FTA on, inter alia, fiscal

revenue. Cage and Gadenne (2014) provides a broad historical lens on post-tariff

liberalisation recoveries5 while (Kassim, 2016) and (Keen & Mansour, 2009) review the

experience of countries in Sub Sharan Africa.

A few studies produced synthesized analysis, for example, Cirera, et al. (2011), Gathii

(2016), Gathii, et al. (2017), Bilal, et al. (2012) and OECD (Grosso, et al., 2001).

Different approaches are more useful for answering different questions – or indeed different

aspects of the same question. When it comes to estimating impacts on tax revenue, for

example, CGE results allow for comparisons of different tariff reduction scenarios, which in

turn can isolate impacts on tax revenue in different simulations. This makes CGE evidence

preferable, as econometric evidence is likely to select the impact of simultaneous

interventions affecting tax revenue (Cirera, et al., 2011). Simulation models estimating total

fiscal revenue with actual collected tariff revenue - for example, the World Bank’s TRIST tool

– are also more reliable that those measuring tariff revenue in isolation using statutory rates

and recorded trade flows.

Econometric evidence is, however, very useful in assessing historical impacts of different

tariff cuts, and recoveries. Some capture the longer term impact of liberalisation (Kassim,

2016), while others look directly at tax performance in the affected countries (Keen &

Mansour, 2009).

Another study (Brafu-Insaidoo & Obeng, 2008) focusses on testing the efficiency of the trade

tax administration system and revenue leakage as these affect the ability of states to

implement effective tariff reforms. They focus on two tax system measures: buoyancy and

5 The authors constructed a panel dataset of tax revenues and government expenditures in developing countries (1945- 2006) from 110 episodes of decreases

in tariff revenue

4

elasticity6. High buoyancy and elasticity of the tax system indicates an efficient customs

administration system, which in turn indicates fiscal resilience to tariff reductions (even in

cases where governments rely heavily on these revenues).

2 . 1 . 2 A s s u m p t i o n s , R i s k s a n d M e t h o d o l o g i c a l I s s u e s

Comparing findings of different simulation studies is rarely straightforward. Liberalisation

scenarios differ, as do data sources, elasticities of demand for imports, or level of

aggregation in data, etc. (Bilal, et al., 2012).

There are also various assumptions and associated risks that drive results and affect

estimations7. The following concerns should be borne in mind when weighing results and

their use in decision making8:

Using statutory duties as reported in the standard trade databases such as TRAINS

and WTO IDB (rather than applied duties) will lead to overestimation of the impact of

reducing tariffs on trade flows and on revenues. Revenues predicted by statutory

tariffs will almost certainly be higher than actual collected tariff revenue.

Static estimates of aggregate revenue effects – while useful in understanding broad

effects – need to be interpreted in the context of a dynamic policy environment.

There is a general assumption that higher trade tariffs are protectionist rather than

fiscally motivated (i.e. a sign of narrow domestic tax bases or limited institutional

capacity for domestic tax collection). However, the underlying economic structures of

both fiscal and protectionist goals will influence effective tariff reform.

Tariff revenue losses only represent a part of the fiscal revenue impact of tariff

reductions. Customs duties are just one tax measure that apply to trade (others

include value-added sales and excise taxes).

Although useful, many CGE studies carry the assumption of a frictionless reallocation

of labour and other factors across sectors, which is an oversimplification not always

supported by the econometric evidence.

Studies that use trade flows to proxy trade liberalisation or use dummies9 to capture

trade reform episodes (around 67% of specifications) tend to find positive effects on

employment, tax revenue and labour reallocation. While robust, for current purposes

trade flows are a poor proxy for tariff reductions in some cases.

2 . 1 . 3 B r i e f r e v i e w o f t h e d a t a u s e d

Bilal, et al. (2012) note that data used in (especially general equilibria) estimations will

strongly influence both results and their interpretation. The source should be up to date,

disaggregated, and containing actual (not proxied) tariff revenues. The table below highlights

data sources of the included studies.

6 Buoyancy measures growth in duty revenue as a result of income growth (i.e. effects of tax base expansion and related changes), while elasticity measures

control for automatic or natural economic growth.

7 For example, a lower risk assumption would be one would influence the magnitude of an effect, but not necessarily the direction. A higher risk assumption

would be one that, when changed, materially influences whether an effect is positive or negative.

8 Concerns collectively raised in (Cirera, Willenbockel, & Lakshman, 2011), (Kowalski, 2005), (Bilal,

Dalleau, & Lui, 2012), (Brenton, Hoppe , & von Uexkull, 2007), (Mevel & Karingi, 2012), and

(Jensen & Sandrey, 2015).

9 A dummy variable is used in econometrics to detect within the presence or absence of an event, in our case a trade liberalisation episode, can help explain

e.g. changes in tax revenue. To continue with the example, a dummy will take the value of 1 for countries that have liberalised trade, and 0 for countries that

have not. Data is then used to see if this ‘dummy’ is significant in explaining tax revenue for the countries assessed.

5

Table 1: Data Sources

Reference Database

(Jensen & Sandrey, 2015) GTAP version 9.2, base year 2011

(Mevel & Karingi, 2012) GTAP version 7, base year 2004; MAcMap-HS6

version 2 database

(UNCTAD, 2015) UNCTADstat database (2014)

(UNCTAD, 2016) UNCTADstat database (2015)

(Cage & Gadenne, 2012) Baunsgaard & Keen (2010), Government

Finance Statistics data; and Mitchell (2007)

(Grosso, et al., 2001); (Cirera, et al., 2011);

(Gathii, 2016); (Gathii, et al., 2017); (Stephens,

et al., 2015); (Bilal, et al., 2012)

Meta-analysis; various ex-post and ex-anti

econometric and CGE studies; different

approaches used to overcome comparability

problems

(Sandrey, 2013) ITC; WTO; SACU; SARS; GTA trade and export-

related databases

(Mangeni, et al., 2014) COMESA Secretariat Statistics Unit

(Brenton, et al., 2007) Official customs data (incl. duties paid, duty

exemption data, by product and country of origin)

for 4 COMESA countries where available and

detailed

(Brafu-Insaidoo & Obeng, 2008) Combined annual data from various sources:

IMF database, the World Bank database, United

Nations’ Commodity Trade Statistics, Ghana

Statistical Services, Customs, Excise and

Preventive Services, and the Ministry of Finance

and Economic Planning

(Kowalski, 2005) PE Model: TRAINS (2000-2002); GE Model:

GTAP 2001; Tariff rates: ITC/CEPII MacMapps

database

(Keen & Mansour, 2009) Uses a new tax dataset for the period 1980 to

2006, disentangling tariff from commodity tax

revenue and distinguishing between resource

related and other revenues.

2.2 Fiscal Impact of Trade Liberalisation: summary

of findings

2 . 2 . 1 T h e s i m u l a t e d p o t e n t i a l i m p a c t o f t h e C F T A

The CFTA will generate greater trade and welfare gains over the longer run with

opportunities for greater employment (UNCTAD, 2016). (Jensen & Sandrey, 2015) estimate

a welfare gain of $7.3 billion by 2025, with the largest gains going to South Africa ($5.7bn),

Nigeria ($2bn) and Kenya ($1.3bn). (Mevel & Karingi, 2012) are much more conservative,

suggesting real income gains for African countries of 0.2 per cent or $296m with an increase

in total African trade by 4% or $25.3bn. Total intra-African trade would grow by 52.3% or

$34.6n if improvements in trade facilitation are realized within the CFTA, a further US$ 85

6

billion would be added to intra-African trade – increasing the share of intra African trade in

total trade by a further 5% by 2022.

At the aggregate level, the CFTA is expected to amount to a relatively small trade shock

since intra-African trade accounts for only 13.6 per cent of total African imports and the

CFTA is expected to contain exclusion lists and safeguard provisions (Gathii, 2016). Given

that, on average, for LICs in Sub Saharan Africa, customs and import duties account for 2%

of GDP and for 15% of total tax revenue (Mansoor, 2014), the impact may still be significant.

As a rough estimate, taxes could fall by 0.25% of GDP.

The impact on tariff revenue is expected to vary. (UNCTAD, 2015) suggests that, at the level

of the RECs, tariff revenue (not total taxes) will decline by 11.5% for ECCAS, 9.6% for

ECOWAS and 7.8% for COMESA. (Jensen & Sandrey, 2015) find big potential differences in

the impact on tariff revenue at the country level. The effect is strongest on Zimbabwe,

Angola, DRC, Nigeria, Kenya, Tanzania, Ghana, Ethiopia and Mozambique. For many of the

larger countries (Egypt, Morocco, Tunisia, and Senegal) tariff losses are not an issue. Their

analysis suggests that SACU would see tariff revenues increase over time – but this is

driven by a growth effect on the economy which sucks in imports from the rest of the world

and thereby increases tariffs.

2 . 2 . 2 E s t i m a t i n g t h e p o t e n t i a l C F T A f i s c a l i m p a c t

To get to ball park figures for the potential fiscal impact of the CFTA, we have:

(a) Estimated the total taxes from customs and duties to be an average of 1.9% of GDP

(using estimates from the OECD (2015) and the World Bank (2013));

(b) Estimated the total tariff revenue reduction from the CFTA to be (i) 10% of total taxes

from customs and duties and (ii) 15%, as the upper limit of the impact (assuming trade also

grows through reduction in non-tariff restrictions (Mevel & Karingi, 2012)

We have assessed these estimates on the baselines of (a) All Africa (b) All of Sub Saharan

Africa (SSA) and (c) all of SSA excluding the three biggest economies of Nigeria, South

Africa and Angola.

We take the estimate for SSA excluding Nigeria, South Africa, and Angola as the upper-limit

for fiscal compensation as the smallest and poorest economies will be prioritised in any

compensation scheme (see next section).

On this basis, the upper limit for fiscal compensation would have been $1.1bn to $1.7bn in

2015 and $1.5 to $2.3bn in 2016.

$ Billion 2015 2016

GDP

All Africa $2,240 $2,130

SSA $1,600 $1,500

SSA exc. Nigeria, South Africa and Angola $600 $800

Taxes from customs and duties (1.9% GDP)

All Africa $43 $40

SSA $30 $29

SSA exc. Nigeria, South Africa and Angola $11 $15

Potential tax affected (10% trade tariff free)

7

All Africa (%GDP) $4 (0.2%) $4 (0.2%)

SSA $3. $3

SSA exc. Nigeria, South Africa and Angola $1.1 $1.5

Potential tax affected (15% trade tariff)

All Africa (%GDP) $6.4 (0.3%) $6.1 (0.3%)

SSA $4.6 $4.3

SSA exc. Nigeria, South Africa and Angola $1.7 $2.3

Sources: OECD, World Bank, UNCTAD trademap

2 . 2 . 3 I n s i g h t s o n r e v e n u e l o s s f r o m t h e b r o a d e r l i t e r a t u r e

Drawing on the broader literature review, tariff reductions, all else equal, will result in fiscal

revenue loss in the short run. Given the relative dependency of emerging economies on

trade taxes, the initial impact can be significant. (Cage & Gadenne, 2012) found trade taxes

fell by 4 GDP percentage points during episodes of tariff reductions. However, the overall

and longer term impact depends on a range of factors. (Kassim, 2016) found that, for

selected countries in Sub Saharan Africa, a 10% reduction in tariffs is associated with a

4.3% fall in trade tax revenue.

The extent to which liberalisation is accompanied by a rationalisation of tariff structure and

exemptions is often very important. (Brenton, et al., 2007) find that it is important to control

for tariff exemptions in simulating revenue loss – in their analysis of COMESA countries

they find that removing exemptions would more than compensate tariff reductions in the

cases of Malawi and Zambia (resulting in an increase in revenue despite the lower statutory

tariff). Improved administration can also play an important role in reducing tax loss – for

example, inefficient border procedures cost the EAC an estimated 5% of revenue (OECD,

2016).

Countries can offset tariff revenue losses by raising other domestic taxes. But this can be

challenging. (Baunssgaard & Keen, 2010) find that low income countries, on average,

recover $25c per dollar of lost trade taxes directly through domestic taxes, while high and

middle income countries quickly replace trade taxes with domestic tax revenue. (Brafu-

Insaidoo & Obeng, 2008) find that for Ghana a 1% reduction in the average tariff rate results

in a 0.74% revenue loss. However, (Keen & Mansour, 2009), in a study of 41 countries in

SS Africa during the period 1980 to 2005, find that most countries that lost trade tax

revenues recover the loss, and in many cases have been able to increase total taxes.

Twenty one of the 30 LICs reviewed had higher tax revenues than before liberalisation while

9 did not replace the trade tax loss. Nine of the MICs increased tax revenue, while 2 did not.

The introduction of VAT in the 1990s accounted for over 60% of tax increases.

In terms of concerns over the impact of tariff reduction on society, (Cage & Gadenne, 2014)

find that for the LICs struggling to recover lost taxes, fiscal expenditure on human

development objectives recovers within 3 years because of donor assistance.

Ultimately, the longer-term impact of liberalisation on government revenue depends not only

on tax adjustment and reform, it also depends on the indirect effects of liberalisation on

trade, employment and growth. (Kassim, 2016) finds that while some countries in Sub

Saharan Africa struggled to directly replace trade tax revenue in the short run, trade

liberalisation was generally associated with an increase in the total tax take in the long run;

(Gaalya, 2015) also finds trade openness to have a positive impact on tax revenue in the

8

case of Uganda. This was because of reforms converting quotas to tariffs and also the boost

in trade overall.

In the context of the CFTA, the impact of regional agreements has a slightly different

dynamic relative to unilateral liberalisations. The analytical framework to assess the impact

of an FTA on a countries economic welfare and tariff revenue was established by Viner

(1950). The assessment rests on the key concepts of trade creation and trade diversion.

Trade creation takes place when an FTA results in greater trade among Member States

without disrupting trade with third parties. While tariff revenue from Member States is likely to

decline (if tariffs fall to zero) other indirect taxes (such as VAT) will increase because of

higher trade volumes. And tariffs will continue to be collected on imports from third parties.

Trade creation also increases economic efficiency and should result in higher growth and

direct taxes over the longer term.

Trade diversion occurs when trade between FTA members grows at the expense of trade

with third parties. With trade diversion, tariff revenue is lost not only on Members, but also on

third party imports that are now displaced. And with no real trade growth - other indirect and

direct taxes do not increase to help replace the lost revenue.

This dynamic means that there are often winners and losers as shown by (Stephens, et al.,

2015) and (Jensen & Sandrey, 2015).

2 . 2 . 4 R e v e n u e r e c o v e r y

The most comprehensive assessment of the time taken for revenue recovery is found in

(Cage & Gadenne, 2014). In their assessment of 96 episodes of trade policy related

reductions10 on trade tax revenue between 1970 and 2006, the time taken to recover tax

revenue typically takes 10 years for MICs and 12 years for LICs. However, this average

masks important variations in experience. Over 50% of MICs and over 40% of LICs affected

recover their tax revenue within 5 years (see table).

Table 2: Time for Fiscal Recovery: 96 trade related trade tax losses (1970-2006)

Time to recovery MICs LICs

Under 5 years 51.5% 42.4%

Under 10 years 59.3% 56.2%

Under 20 years 70% 60%

Source: Cage & Gadenne, 2014

(Keen & Mansour, 2009) do not explicitly look at the time to recover from lost tariff revenue,

but their assessment of the tax performance of 41 Sub Saharan Africa countries suggests

that tax revenue recovers, on average, in a three to five-year period.

Experience with government revenue mobilisation (World Bank IEG, 2017) suggests that

programmes supporting tax revenue recovery are generally successful with results often felt

within a three to eight year window. For example, Paraguay raised revenue 15.6% of GDP in

2006 to 17.8% by 2009 and Orisa State, India, increased revenue from 5.6% of state GDP in

2000 to 9.2% by 2008. In Sub Saharan Africa, (Keen & Mansour, 2009) found that VAT,

introduced in the 1990s, drove 66% of the increase in total revenue in the 41 countries

studied; accounting for even more in non-resource rich countries. (Keen, 2012) and

10 They look at a total of 130 episodes of trade tax reductions.

9

(Prichard, 2016) strike a note of caution regarding the efficiency gains for VAT and the scope

to greatly increase revenue from this source without reforms to address the informal sector.

2.3 Conclusions on tariff revenue loss and recovery

The CFTA is likely to result in welfare gains of up to $7bn and increases in the share

of intra African trade in total trade by 5%; by 10% if the CFTA is successful in trade

facilitation.

Much of the benefit is expected to go to the big players such as South Africa, Nigeria

and Kenya.

The initial size of the trade shock is limited by the current low levels of intra African

trade (13.6%) and the exemptions on liberalisation that the Member States are likely

to maintain. We would expect total taxes to fall by roughly 0.2%- 0.3% of GDP, on

average, because of the CFTA.

For Sub Saharan Africa excluding the three major economies of Nigeria, South Africa

and Angola, we would expect lost taxes to be between $1.1bn - $1.7bn on the basis

of 2015 figures, and $1.15bn - $2.3bn on the basis of 2016 figures.

The impact of tariff liberalisation varies. At the level of the RECs, tariff revenue (not

total taxes) could decline by 11.5% for ECCAS, 9.6% for ECOWAS and 7.8% for

COMESA. At country level, the effect is likely to be strongest on Zimbabwe, Angola,

DRC, Nigeria, Kenya, Tanzania, Ghana, Ethiopia and Mozambique.

Given the importance of tariff revenue in total taxes, the challenge of revenue loss

needs to be taken on board. Note that for many of the larger countries (Egypt,

Morocco, Tunisia, and Senegal) tariff losses are not an issue.

Experience with tariff liberalisation suggests that the majority of countries, including

in Sub Saharan Africa, recover the tax lost from tariff reductions within a 5 to 10 year

period. And in many cases tax revenue is higher within 10 years of the start of

liberalisation. In Sub Saharan Africa the main source of replacement revenue has

been VAT.

But a significant minority of LIC are not able to replace lost revenue. In these

countries government expenditure has been maintained in large part through Donor

assistance. These countries require adjustment support over the longer term to

ensure liberalisation contributes to sustainable development.

3 F i s c a l l o s s c o m p e n s a t i o n m e c h a n i s m s

and adjustment facilities

While the majority of MICs and LICs recover lost tax revenue within 5 to 10 years of tariff

reductions, the burden in the short run can still be substantial. Unilateral liberalisation has

resulted in short run tax losses of between 2 and 5% of GDP for several Sub Saharan

African countries. However, we can expect the average tax loss from the CFTA to be lower,

at an average 0.2% of GDP. That said, the challenge for certain countries will still be

significant. Their implementation of tariff phase downs could be made less disruptive and

more sustainable with short run compensation for fiscal loss. Therefore, international

experiences with Financial Compensation Loss Mechanisms (FCLM) are reviewed next, to

the extent that information is available. Furthermore, if the experience of the last 3 decades

is anything to go by, it is likely that a small subset of countries will struggle to recover

revenue even over the long term. The paper therefore reviews, to a very limited extent,

international experience with regional adjustment funds.

10

3.1 Brief discussion of the literature reviewed

The studies selected for the review have a strong bias towards financial loss compensation

mechanisms (FLCM) associated with regional trade agreements (RTAs) within Africa, while

examples from the ACP-EU EPAs and non-African RTAs are also included to widen the net.

The review compares important structural, technical and institutional characteristics of

compensation mechanisms, including those of Africa’s main RTAs (CARICOM, COMESA,

ECCAS, ECOWAS, SACU, and WAEMU/UEMOA). Characteristics captured include

financing structures and resource mobilization strategies; distribution, management, and

governance mechanisms; fund allocation criteria; operational periods; and phase-downs.

Where possible, commentary is provided on risks associated with implementation and

general recommendations or lessons that emerge.

The literature on FLCMs is limited, in particular with regards to the effectiveness and impact

of the schemes. The few extant evaluations have generally focused on the need to move

support from compensation to adjustment (Olympio & Wangwe, 2010).

The review also, briefly, touches on adjustment facilities. Here there is a greater appreciation

of challenges and success factors than for the FLCM.

3.2 FLCMs as distinct from Adjustment/Regional

Development Funds

FLCMs have generally been implemented through broader regional integration support

programmes that go beyond compensation. Compensation to Rwanda and Burundi for the

adoption of the EAC Customs Union and Common External Tariff was implemented by

COMESA through the Regional Integration Support Mechanism (RISM) programme, which

also supported infrastructure development and broader adjustment objectives. In CEMAC,

fiscal compensation is allocated 40% of funds from the Fonds de Développement de la

Communauté (FODEC) while 60% is to target regional integration projects (including

infrastructure). The ECOWAS Regional Development Fund (ERDF) was responsible for

lending to support regional infrastructure projects as well as fiscal compensation.

However, FLCMs are distinct from the adjustment or regional development funds through

which they are often implemented. This is because: (a) they are strictly time bound to

accommodate the immediate impact of tariff liberalisation on government revenue and (b)

they do not target specific activities or beneficiaries. Adjustment funds can also be temporary

in nature but generally target implementing trade policy reform and cushioning its impact

through e.g. retraining of displaced labour. (In principle, adjustment funds can also be used

to support fiscal reform programmes to help governments replace lost tariff revenue.

However, such support is usually provided through comprehensive programmes that

address government revenue and expenditure management.) Support to adjustment,

including implementation, is provided through such programmes as SADC’s Trade Related

Facility (TRF).

Regional Development Funds are generally more focused on regional infrastructure. For

example, the AfDB’s Development Fund (ADF) supports power generation and regional

transport initiatives; and the Fondo para la Convergencia Estructural del MERCOSUR

(FOCEM) also finances regional projects in energy and transport. However, the lines

between adjustment and regional development funds are blurred – an important component

of all regional development funds reviewed was to support the competitiveness of smaller

member states and to promote social cohesion (in different forms).

11

3.3 Summary of Findings

3 . 3 . 1 F i n a n c i n g S t r u c t u r e

FLCM:

oFLCMs generally, rely on a community levy on imports from third countries,

varying from 0.4% to 1%. Where countries are members of two blocs

employing a FLCM, both levies are imposed, So countries in UEMOA and

ECOWAS impose combined levies of 1.5%, those in ECCAS and CEMAC

combined levies of 1.4%. The discriminatory nature of the community levy

raises the issue of WTO compatibility (See box).

oIn the case of CARICOM, an LDC suffering fiscal loss can, subject to the

endorsement of the COTED, re-impose tariffs on intra-regional trade and

keep the resulting revenue.

oIn the case of COMESA’s Regional Integration Support Mechanism (RISM),

compensation was paid to Rwanda and Burundi primarily from Donor funds

(Member States contributed less than 5% to RISM).

Adjustment/ Regional Development Facilities (A/RDF):

oMostly rely on a combination of core funding from Donors and other

institutions such as development banks.

oEffective resource mobilization is very important for a Fund to accomplish its

objectives.

Member States payments have often been a challenge, sometimes

because of political crisis (UEMOA/ECOWAS) and sometimes

because of concerns over the potential effectiveness of the

mechanism they are being called on to support.

Different approaches have been taken to try and secure regular

payment. For example, ECCAS has placed Central banks in charge of

ensuring Member State’s contributions.

WTO compatibility of a community levy

A Community Levy on imports from non CFTA trade partners may raise concerns among

WTO members. This is because such a levy risks violating the rules that govern the

formation of a preferential trade agreement.

There are two options for notifying an FTA– under Article XXIV or the Enabling Clause.

Article XXIV(5)(b) requires that duties and other trade regulation applicable to 3rd countries

should not be higher or more restrictive than prior to the formation of the FTA. There is also

a requirement to show that the levy is essential to the creation of the CFTA. The Enabling

Clause provides for the notification of preferential trade agreements involving developing

countries. It requires that the agreement not “raise barriers to or create undue difficulties for

the trade of any other contracting parties”.

The Community levy would need to be justified whether the CFTA is notified under Article

XXIV or the Enabling Clause. The Enabling Clause would be preferable as it has less

stringent requirements in comparison to Article XXIV.

The levy would be open to challenge at the WTO. While a community levy of 0.2% to 0.4% is

unlikely to create “undue difficulties”, the significance of the levy could also be assessed in

terms of systemic ‘principles’ (others may copy CFTA members, creating a snowball effect).

Ultimately CFTA members will need to decide whether to risk the challenge, and whether the

purported gains are worth it.

12

3 . 3 . 2 M a n a g e m e n t , g o v e r n a n c e m e c h a n i s m s

There are different structures for management and governance of both FLCM and

Adjustment Facilities.

oThere is strong Ministerial and Secretariat involvement in the CARICOM and

COMESA FLCMs, and in the COMESA AF and SADC TRF.

oECOWAS, UEMOA and CEMAC involve development banks directly.

Subpar management of adjustment and regional funds has in some cases

contributed to poor resource collection and ineffective implementation.

oThis challenge prompted the reform of the ECOWAS fund to include a strong

financial institution capable of mobilizing resources for financing investment

(ERIB) and development (ERDF).

oSimilarly, UEMOA set up the Regional Integration Support Fund (RISF) to

better manage and finance development projects.

3 . 3 . 3 F u n d i n g A l l o c a t i o n C r i t e r i a

FLCM

oCompensation has generally been calculated on the basis of either:

intraregional trade shares (in proportion to their shares in total intraregional

imports) or incurred revenue losses (the loss in revenue associated with the

non-application of MFN tariffs to partner country trade). The latter is more

transparent as the loss can be calculated on the basis of readily available

information (MFN tariffs and import data).

oCompensation is restricted to the least developed or most disadvantaged in

the FTA.

Adjustment/ Regional Development Facilities (A/RDF):

oThere are a range of allocation criteria and approaches, guided by the scope

and principles established for the facility. They are generally focused on the

least developed countries and aim at ensuring that the benefits of regional

integration are shared – either through increasing competitiveness or

supporting social cohesion. There is also a strong element of reducing the

burden of implementing regional agreements.

oChallenges relate to a range of factors, but Member State engagement has

been key. Where mechanisms or facilities have been passive – relying on

Member States alone to define their needs and draw down on the funds,

activity has been limited or at best slow paced. For example, the financing

agreement for SADC’s Trade Related Facility (TRF) was signed in 2014, but

projects were only agreed on in July 2017 because of the challenges Member

States experienced in conducting needs assessment and project

development.

oExperience has shown the value of establishing management units. The

COMAid Unit, established to better oversee programme co-ordination and

management of COMESA RISM, has contributed to significantly increased

programme activity and Member State engagement. And the outreach and

technical assistance provided to Member States by the Facility Support Unit

of the SADC TRF was important in ensuring the programme got off the

ground.

3 . 3 . 4 O p e r a t i o n a l P e r i o d

FLCM: the FLCM are generally time bound to address the immediate period of tariff

liberalisation. They ‘cover the gap’ in fiscal revenue as broader fiscal adjustment

takes place. Some of the examples have a phased approach (UEMOA, ECOWAS) to

13

link compensation more directly to each step in the tariff phase down, while others

are more akin to once off lump sum payments (COMESA/ EAC RISM). In contrast,

CARICOM’s compensation arrangements are not time bound,

oUEMOA RISF progressively reduced tariffs and paid compensation on actual

loss of revenue: Compensation covered a nine-year period, while the

mechanism was operational for 6 years.

In the first period (1996-1997) UEMOA tariff rates reduced from 20%

to 14% so compensation of 6% was made to states with loss of

revenue [second period: tariff 8% compensation 12%; third period:

tariff 4% compensation 16%]. This continued to 2000, when the

customs union was achieved.

From then, compensation changed, reducing year on year: 100% in

2000-2002, 80% in 2003, 60% in 2004, 30% in 2005, 0% 2006. IE CM

in place for 9,5 years (1 July 1996 to 31 December 2005).

In 2002, Benin received over 373m CFA Francs as compensation for

revenue lost in 1998-2000.

oThe ECOWAS FLCM was operational for four years.

Compensation paid was 100% of loss in 2002, 80% in 2003, 60% in

2004, 30% in 2005, 0% of loss from 1 January 2006.

Niger received CFA 20 billion for losses between 1995 and 2002

oThe COMESA EAC RISM

The operational period of the FCLM was determined more by the

development programming cycle than the rhythm of tariff liberalisation.

Under calls for proposals, Burundi received Euro12.7m and Rwanda

Euro22.6m to compensate for potential fiscal losses as a result of

applying the EAC Customs Union.

oCAROCOM

CARICOM takes a different approach. Financial compensation is not

time bound.

But financial compensation is “paid’ though the re-imposition of tariff

for a limited duration. There is therefore no need for a regional system

to manage a compensation mechanism.

Adjustment/ Regional Development Facilities (A/RDF): While several FCLMs have

been executed within an A/RDF, the Funds are targeted at longer duration activities.

Making these funds operational takes time, be it in Latin America, the Caribbean or

Africa,

oFOCEM was established in 2005 and became operational in 2007. In 2015

the fund was extended by a further 10 years and provided with additional

financing.

oThe SADC TRF was established in 2014 but only became operational in

2017. The programme extends up to 2019.

oThe CARICOM Development Fund was established in 2008 with its first

Country programme approved in 2010. The CDF will run indefinitely, subject

to finance.

4 Conclusion and recommendations

The CFTA is likely to result in welfare gains of up to $7bn and increases in the share of intra

African trade in total trade by 5% (by 10% if the CFTA is successful in trade facilitation).

Much of the benefit is expected to go to the big players such as South Africa, Nigeria and

Kenya.

Because intra African trade is only around 14% of total trade, and because Member States

are likely to maintain many exemptions to liberalisation, the initial impact of the CFTA on

14

total taxes will be roughly 0.2% to possibly 0.5% of GDP, on average. The impact on

revenue will vary by region and country. At country level, the effect is strongest on

Zimbabwe, Angola, DRC, Nigeria, Kenya, Tanzania, Ghana, Ethiopia and Mozambique.

For Sub Saharan Africa, excluding the three major economies of Nigeria, South Africa and

Angola, we would expect lost taxes to be between $1.1bn - $1.7bn on the basis of 2015

figures, and $1.15bn - $2.3bn on the basis of 2016 figures.

A Community Levy of 0.4% on imports from 3rd parties would be sufficient to finance a FLCM

to cover the tax losses. Such a levy would have raised $1.7bn in 2015 and $1.6bn in 2016,

sufficient on the basis of a two-year average (see table).

2015 2016

Potential Compensation requirement for tariff losses

SSA exc. Nigeria, South Africa and Angola

$1.1 - $1.7bn $1.5 - $2.3bn

Imports (all CFTA members) with the Rest of World $444bn $406bn

Finance raised by a 0.2% Community Levy $0.89bn $0.81bn

Finance raised by a 0.4% Community Levy $1.78bn $1.62bn

Experience with tariff liberalisation suggests that the majority of countries, including in Sub

Saharan Africa, recover the tax lost from tariff reductions within a 5 to 10-year period. And in

many cases tax revenue is higher within 10 years of the start of liberalisation.

The timeframes of the ECOWAS and UEMOA FLCM, operational for between 4 to 6 years

and covering up to 9.5 years of fiscal losses, would be sufficient for most CFTA countries.

Given the potential management challenges in applying a CFTA wide FLCM,

In terms of operationalising a FLCM, its narrow focus and limited duration would point to a

dedicated, stand alone, facility which could be relatively easy to manage and staff. This

would shelter its performance from the growing pains of any Adjustment or Regional

Development Facility that might accompany the CFTA.

However, even a dedicated facility would still need to raise funds. So even if it is delinked

from a Fund for operational purposes, it is still likely to be dependent on the decisions made

with regards to how the CFTA institutions will raise and channel finance.

There is also the approach of CARICOM to consider. This allows countries to re-impose

tariffs for a strictly limited period. Such an approach has several practical advantages. It

does not require a new institution or financing mechanism and can be easily applied and

monitored. It would also avoid the risks of a Community Levy (on third parties) in terms of

WTO compatibility. However, it is the least attractive from an economics perspective

because businesses operate with less uncertainty – they cannot be guaranteed the market

acces provided under the CFTA as Government may re-impose tariffs.

Experience also suggests that a minority of countries will struggle to replace lost tax revenue

even over the longer term. These countries would require adjustment assistance. However,

given the experience of operationalising adjustment facilities, the management and

governance challenges would need to be reviewed and assessed more deeply before a

decision is made on a CFTA adjustment facility. At what level would such a fund operate? At

REC level (FOCEM)? To what extent would it be accessible to individual member states

(RISM. TRF, CDF)? What would be the focus – social cohesion, regional infrastructure? And

how would it link to other programmes supporting regional integration – including the e.g.

ADF and the Africa Infrastructure Facility (AIF)? What would be the arrangements for

15

financial management – a dedicated development bank or different regional development

banks or an Aid management unit?

5 Bibliography

AFDB, 2005. ECCAS RIASP: 2005-09, s.l.: s.n.

Baunssgaard, T. & Keen, M., 2010. Tax Revenue and (or?) Trade Liberalsiation. Journal of

Public Economics, Volume 94.

Bilal, S., Dalleau, M. & Lui, D., 2012. Trade Liberalisations and Fiscal Adjustments: The

Case of EPAs in Africa, Maastricht: ECDPM.

Brafu-Insaidoo, W. G. & Obeng, C. K., 2008. Effect of Import Liberalization on Tariff

Revenue in Ghana, Nairobi: AERC Research Paper 180.

Brenton, P., Hoppe , M. & von Uexkull, E., 2007. Evaluating the Revenue Effects of Trade

Policy Options for COMESA

Countries: the Impacts of a Customs Union and an EPA with the European Union, s.l.:

World Bank.

Cage, J. & Gadenne, L., 2012. The Fiscal Cost of Trade Lib eralization, Paris: PSE.

Cage, J. & Gadenne, L., 2014. Tax Revenue, Dwvelopment and the Fiscal Cost of Trade

Libreralisation, 1792 - 2006, s.l.: Harvard.

Cirera, X., Willenbockel, D. & Lakshman, R., 2011. What is the evidence of the impact of

tariff reductions on employment and fical revenue in developing countries? A Systematic

Review, London: EPPI Centre, University of London.

Correa, F. d. A., 2010. A Legal and Institutional Anlayis of FOCEM, the MERCOSUR Fund

for Structural Adjustment, s.l.: s.n.

ECDPM, 2016. Political Economy of Regional Integration in Africa: What Drives and

Constrains Regional Organisations, s.l.: s.n.

ECOWAS, 2003. Protocol Relating to the Application of Compensation Procedures for Loss

of Revenue Incurred by ECOWAS Member States as a Result of the Trade Liberalisation

Scheme, Abuja: ECOWAS Secretariat.

Flatters, F. & Stern, M., 2005. Implementing the SACU Revenue-Sharing Formula: Customs

Revenues, s.l.: USAID South Africa.

Gaalya, M., 2015. Trade Liberalisation and Tax Revenue Performance in Uganda. Modern

Economy, 6(2).

Gathii, J., 2009. African Regional Trade Agreements as Flexible Legal Regimes. Albany Law

School Working Paper Series, Volume No. 20.

Gathii, J. T., 2016. Scoping Study: Designing the Continental Free Trade Area (CFTA): An

African Human Rights Perspective, s.l.: UNECA.

Gathii, J. T., Burnett, K., Nshimbi, C. C. & Dommen, C., 2017. Report: The Continental Free

Trade Area (CFTA) in Africa - A Human Rights Perspective, Geneva: ECA & FES.

16

Grosso, M. G., Edmonson, C. & Walkenhorst, P., 2001. Regional Integration: Observed

Trade and Other Economic Effects, Paris: OECD.

Healy, F. & Baker, P., 2014. Evaluation of the Reional Integration Support Mechanism

(RISM), Lusaka: COMESA.

Jensen, H. G. & Sandrey, R., 2015. The Continental Free Trade Area - A GTAP

assessment, Stellenbosch: TRALAC.

Kassim, L., 2016. The Revenue Implications of Trade Liberalisation in Sub Saharan Africa;

Some New Evidence. Universtity of Kent School of Economics Discussion Papers.

Keen, M., 2012. Taxation and Development - Again. Volume WP/12/220.

Keen, M. & Mansour, M., 2009. Revenue Mobilisation in Sub Saharan Africa: challanges

from Globalisation. Volume WP/09/157.

Kowalski, P., 2005. Impact of Changes in Tariffs on Developing Countries'Government

Revenue. OECD Trade Policy Working Papers (no. 18).

Mangeni, F. et al., 2014. Key Issues in Regional Integratin: Volume III, s.l.: COMESA.

Marape, D., 2006. Report of the Study on Operational Modalities of the COMESA Fund, s.l.:

COMESA.

Mevel, S. & Karingi, S., 2012. Deepening regional integration in Africa: A computable

general equilibrium assessment of the establishment of a Continental Free Trade Area

followed by a Continental Customs Union. Kigali, UNECA.

OECD, 2013. The Development Dimension: Succeeding with Trade Reforms - the Role of

Aid for Trade, s.l.: s.n.

Olympio, J. & Wangwe, S., 2010. Mid-Term Evaluation of the Regional Integration Support

Mechanism, s.l.: COMESA.

Prichard, W., 2016. Reassessing Tax and Develoopment Research: A New Data Set, New

Findings and Lessons for Research. World Development , Volume 80.

Sandrey, R., 2013. An analysis of the SADC Free Trade Area, Stellenbosch: TRALAC.

SATH, 2012. 2012 Audit of the Implementation of the SADC Protocol on Trade, s.l.: s.n.

Stephens, C., Irfan, M., Massa, I. & Kennan, J., 2015. The Impact of Free Trade Agreements

between Developed and Developing Countries on Economic Development in Developing

Countries, London: DFID Overseas Development Institute.

UNCTAD, 2015. Elements of Modalities for the Continental Free Trade Market Access

Negotiations on Tariffs: Some Key Issues for Consideration, Geneva: UNCTAD.

UNCTAD, 2016. African Continental Free Trade Area: Policy and Negotiation Options for

Trade in Goods, Geneva: UNCTAD.

UNECA/AU, 2006. Assessing Regional Integration in Africa II, s.l.: s.n.

Walkenhorst, P., 2006. Compensating Lost Revenue in Regional Trade Agreements. World

Bank Trade Note 28.

World Bank IEG, 2017. Tax Revenue Mobilisation: Lessons from World Bank Group Support

for Tax Reform, s.l.: s.n.

WTO, 2013. Trade Policy Review for CEMAC, s.l.: s.n.

17

6 A n n e x e s

6.1 Fiscal Impact of Tariff Liberalization: Literature Table

The table below sets out provides an overview of the papers assessed for the study of fiscal impact.

Table 3: Papers assessed for the study of fiscal impact

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

2017 Report: The

Continental

Free Trade

Area (CFTA) in

Africa - A

Human Rights

Perspective

(Gathii, et al.,

2017)

CFTA Continental Ex-ante impact

assessment

initiated during the

preparatory period

of the CFTA

negotiation

process, this study

provides a unique

opportunity to

present the

findings and

analysis during

the negotiations

and thus to

contribute to a

positive, robust

and human-rights

consistent

At the aggregate level, the

CFTA is expected to

amount to a relatively small

trade shock since intra-

African trade accounts for

only 13.6 per cent of total

African imports and the

CFTA is expected to

contain exclusion lists and

safeguard provisions. That

said, continental

liberalisation will still

contribute to a reduction in

tariff revenues, particularly

for African countries that

are heavily reliant on trade

tariffs on intra-African

imports as a source of

Given that governments have

obligations to mobilise resources

for human rights purposes –

such as education and social

protection - the full breadth of

implications of tariff reductions

must be considered with utmost

care.

18

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

outcome government revenue.

2016 African

Continental

Free Trade

Area: Policy

and

Negotiation

Options for

Trade in

Goods

(UNCTAD,

2016)

CFTA Continental Builds on the

econometric and

analytical work on

evaluating intra-

African trade flows

and the potential

impact of the

CFTA by UNECA,

the IMF, the World

Bank, and

UNCTAD

LR: significant welfare

gains, output and

employment expansion,

and intra-African trade

growth; SR: countries are

likely to bear some tariff

revenue losses which may

not be distributed uniformly

across the African

continent.

Unequal sharing of costs and

benefits may hinder the

negotiation processes and

implementation of the agreement

(if sufficient flanking measures

and flexibilities that ensure more

even allocation of burdens and

benefits among AU member

states are not in place).

Complementary policies to

internal trade liberalization also

NB to offset: such as an

adjustment plan for re-training of

employees and re-tooling of

industries to new growth areas

2016 Scoping Study:

Designing the

Continental

Free Trade

Area (CFTA):

An African

Human Rights

Perspective

(Gathii, 2016) CFTA Continental The methodology

comprised a

combination of

desk research,

including an

extensive review

of trade

integration,

agriculture,

employment and

trade, and semi-

structured

interviews with

experts on HRIA

and key

Recommends the following:

• A development fund from

which funding to compensate for

losses from liberalization

commitments of the CFTA would

come;

• Longer transition periods to

adjust to import competition for

countries that need them;

• Appropriate timelines to

harmonize national tariffs with

the new CFTA tariff to level out

revenue losses particularly for

19

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

stakeholders in

Africa. The HRIA

scoping

assessment

research also

included

determining the

existence of

similar HRIAs, the

main

achievements and

their key

challenges.

least developed country

members;

• Mechanisms to ensure

equitable sharing of gains from

liberalization in a fair and

proportionate way to African

Union members.

2015 The

Continental

Free Trade

Area - A GTAP

assessment

(Jensen &

Sandrey,

2015)

CFTA Africa-wide

scenario:

continental;

Willing

Participants

scenario: the

five countries in

the Southern

African

Customs Union

(SACU), four in

the East African

Community

(EAC), Malawi,

Zambia and

Egypt in the

Tripartite FTA

region, Nigeria

CGE model-based

analysis of 2025,

modelling 2

scenarios (Africa-

wide and "Willing

Participants") in 3

simulations: Full

liberalization; 50%

reduction in

NTMs; and

combination of full

liberalisation and

50% reduction in

NTMs.

Tariff revenue change in

2025 in “Willing

Participants” and “Africa-

Wide” scenarios

respectively [US$ millions]:

SACU (16, 167); Kenya (-

448, -416); Tanzania (-452,

-445); Uganda (-100, -85);

Rwanda (-10, -3); Malawi (-

96, -101); Zambia (-23, -

20); Egypt (-8, -25); Nigeria

(-478, -681); Ghana (-205,

-382); Morocco (-14, -16);

Tunisia (-14, -20)___Other

countries in "Africa-Wide"

scenario [US$ millions]:

Rest of east Africa (-273);

Angola-DRC (-602);

In terms of the tariff revenues

collected at the respective

borders there is little difference

in the lost revenue between an

Africa-wide FTA and the 'willing’

countries. The main exception is

SACU where revenues increase

by $16 million in contrast to the

surprisingly large $167 million

increase from an Africa-wide

integration. This is because (a)

imports into SACU from the

Southern African Development

Community (SADC) are already

duty-free, (b) trade with other

African countries is either limited

or duty-free as with oil imports

from Angola, and (c) South

20

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

and Ghana in

west Africa,

and Morocco

and Tunisia in

north Africa

Ethiopia (-165);

Madagascar (-3); Mauritius

(-6); Mozambique (-88);

Zimbabwe (-1,123);

Senegal (-29); Rest of

Africa (-2,573) ___ Non-

fiscal effects: One-off

increase in global welfare

of $7.3 billion in 2025 and

the largest gains are

expected from South Africa

($5.7 billion), followed by

Nigeria ($2 billion) and

Kenya ($1.3 billion)

Africa imports more goods from

‘outside’ as its economy grows.

Tariff revenues into Zimbabwe

decline by over one billion

dollars, (Zimbabwe has very

high initial tariffs and surcharges

against African imports despite

being a member of the SADC

FTA). For many larger countries

(Egypt,

Morocco, Tunisia and Senegal)

tariff losses are not an issue.

2015 Elements of

Modalities for

the Continental

Free Trade

Market Access

Negotiations

on Tariffs:

Some Key

Issues for

Consideration

(UNCTAD,

2015)

CFTA Regional focus

(UMA,

COMESA,

ECCAS,

ECOWAS,

SADC)

Ex Ante; Total

liberalization

between RECs;

uses weighted

average intra-

regional tariffs to

calculate static

tariff revenue loss

COMESA, ECCAS, and

ECOWAS would

experience the most

significant tariff revenue

losses due to high intra-

regional trade with SADC,

and high corresponding

import protection

[COMESA (7.8%), ECCAS

(11.5%), and ECOWAS

(9.7%)].

Tariff elimination approaches

need to be carefully designed,

including incorporating an

adequate level of flexibilities.

Such a need for balancing act to

address adjustment challenges

is strong in Africa, and

particularly for weaker and

smaller members within the

region.

2012 Deepening

regional

integration in

Africa: A

computable

general

(Mevel &

Karingi, 2012)

CFTA Africa

aggregate; and

country/regions

: Angola &

DRC, Ethiopia,

Madagascar,

MIRAGE CGE

model; compares

simulated CFTA

scenario to a

baseline scenario

By 2022: only South Africa

(5.9%), Botswana (1.7%);

rest of SACU (1.8%)

experience tariff revenue

gains; Tariff revenue losses

are highest in Zimbabwe (-

If improvement in Trade

facilitation is realized within the

CFTA, a further US$ 85 billion

would be added to intra-African

trade. This would represent a

significant 128.4 per cent

21

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

equilibrium

assessment of

the

establishment

of a

Continental

Free Trade

Area followed

by a

Continental

Customs

Union

Malawi,

Mauritius,

Mozambique,

Tanzania,

Uganda,

Zambia,

Zimbabwe,

Rest of Eastern

Africa,

Botswana,

South Africa,

Rest of SACU,

Egypt,

Morocco,

Tunisia, Rest of

North Africa,

Nigeria,

Senegal, Rest

of Western

Africa, Central

Africa

70.5%), Malawi (-60.7%),

Zambia (-60%),

Mozambique (-54%),

Tanzania (-36.2%); Real

income gains for African

countries of 0.2 per cent or

$296m; total African trade

increase by 4% or $25.3b;

intra-African trade would

increase by 52.3% or

$34.6b.

increase over the 2022 baseline.

2016 The Revenue

Implication of

Trade

Liberalisation

in Sub-

Saharan

Africa: Some

New Evidence

(Kassim,

2016)

Unilateral

trade

liberalisation

episodes

28 Sub-

Saharan

African

countries

Econometric panel

data estimation

(1981 to 2010) of

the impact of trade

liberalisation on

tax revenue (using

a dummy variable

for liberalisation)

and the direct

On average,

disaggregating total tax

revenue into domestic tax

revenue and trade tax

revenue, a 10% decrease

in the average trade tariffs

decreased trade tax

revenue by 4.3% but

increased domestic tax

Assessing the overall impact of

liberalisation on fiscal revenue

needs to consider different

aspects of trade reform. Low

income countries can struggle to

replace tariff revenue directly,

but reform (removal of NTB and

quotas) can still result in an

22

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

impact of tariff

reductions on

trade and

domestic taxes.

revenue by only 0.7%.

However, trade

liberalisation is associated

with an increase in total tax

revenues.

This means that the

replacement of quotas and

other quantitative

restrictions with tariffs

increases trade tax

revenue and ultimately,

total tax revenue.

overall increase in taxes.

2015 The Impact of

Free Trade

Agreements

between

Developed and

Developing

Countries on

Economic

Development

in Developing

Countries

(Stephens, et

al., 2015)

FTAs Involving most

OECD states

and 35

developing

countries and

regions (mostly

North-South;

South-South

FTAs)

Rapid Evidence

Assessment of

literature on (a)

impact of FTAs

between

developed and

developing

countries, and (b)

how developing

countries might

benefit (and avoid

harm) from new

FTAs - [supported

by flanking

analysis]. REA

based on a quality

assessment (QA)

of 45 FTAs

Nascent FTA: some

studies flagged potential

loss of government

revenue. Mature FTA: no

studies estimated actual

effects on government

revenue or analysed

government response. At

an aggregate level, FTAs

are neither a 'golden bullet’

to destroy impediments to

trade, nor a potent source

of the harm envisaged by

critics. "Aggregate level"

however; as there are clear

winners and losers.

Countries with a high

reliance on tariff revenues

No systematic difference

between the impact of North–

South and South–South

agreements was found in the

assessed literature. All but one

of the 19 high or moderate

quality primary studies predicted

positive effects of FTA on trade

growth (none found it to be

negative). However, there is a

wide range of estimated effects,

with significant variance of

impacts (magnitude and

direction) across partners.

Notable gaps in knowledge on

three things: impacts on the

revenue, distributional and

social/environmental effects of

23

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

(Jordan 12%; Egypt 15% of

fiscus) suffer

proportionately more from

liberalisation.

FTAs.

2015 Trade

Liberalisation

and Tax

Revenue

Performance

in Uganda

(Gaalya,

2015)

Unilateral

trade

liberalisation

episodes

Uganda Fixed and

Random effects

model used to

establish

determinants of

tax revenue

performance from

1994 to 2012.

Trade openness has a

positive effect on tax

revenue performance.

2014 Tax Revenue

and

Development

and the Fiscal

Cost of Trade

Liberalisation

1792 - 2006

(Cage &

Gadenne,

2014)

Unilateral

trade

liberalisation

episodes

HIC, MIC, LIC

(130 countries)

Used a panel

dataset of tax

revenues and

government

expenditures for

the period 1792-

2006. They

identify 140

episodes that

resulted in

decreases in tariff

revenues, 96

episodes are

associated with

tariff policy reform.

They investigate

two periods –

1792 to 1970 and

1970 to 2006.

In the period 1970-2006, at

the start of tariff

liberalisation tax revenues

fall by an average of 3.3

GDP percentage points

(58% of average trade tax

revenue.

The time taken to recover

tax revenue typically takes

10 years for MICs and 12

years for LICs. However,

this average masks

important variations in

experience. Over 50% of

MICs affected recover their

tax revenue within 5 years,

60% within 10 years and

Nearly 25% of developing

countries never fully recover lost

revenue, this is higher for LICs.

Of those countries that do

recover, the majority do so within

5 years.

24

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

70% within 20 years. Over

40% of LICs affected

recover their tax revenue

within 5 years, 56% within

10 years and 60% within

20 years.

Other revenue (including

aid flows) is important for

maintaining government

expenditure.

2014 Key Issues in

Regional

Integration

Volume III

(Mangeni, et

al., 2014)

COMESA

EAC FTA

COMESA, EAC Tariff Reform

Impact Simulation

Tool (TRIST)

analysis on

revenue impact of

adopting a

Customs Union

Common External

Tariff (using 2013

figures)

Tariff revenue increases on

CET implementation

[Zimbabwe (0.3%),

Madagascar (1.7%),

Zambia (3.0%), Ethiopia

(3.4%), Djibouti (21.8%),

Malawi (22.2%), Eritrea

(38.9%), Egypt (59.1%),

Comoros (64.9%),

Mauritius (64.5%), and

Seychelles (129.2%)],

except Sudan, with

revenue fall of -4.8%

largely due to the loss of oil

revenues to South Sudan.

COMESA: arrangements for

providing adjustment support.

However, the Member State can

be expected to complement

such support with domestic fiscal

reforms to deepen and widen the

tax base, as well as improve

revenue collection.

EAC: Revenue collections from

trade taxes have increased,

especially on VAT and excise,

resulting from increased trade,

although revenue loss was

highly feared during

the negotiating of the customs

union.

2013 An analysis of

the SADC

(Sandrey,

2013)

SADC FTA SADC Review of trade

and tariff data to

Phased tariff reduction

programme commenced in

Data taken from ITC most recent

2010 trade data, augmented by

25

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

Free Trade

Area

measure progress

of tariff reduction

programme

2001 - minimum FTA

conditions met at 85%;

maximum FTA conditions

met in 2012 when tariff

phase-down for sensitive

products was achieved

(excluding Mozambique

[extended phase-down to

2015]; and Malawi,

Zambia, Tanzania [25%

allowances on Sugar and

Paper Products until 2015])

WTO and SACU overall trade

figures. [Even this data is

incomplete and not necessarily

consistent with other data]

2012 Trade

Liberalisations

and Fiscal

Adjustments:

The Case of

EPAs in Africa

(Bilal, et al.,

2012)

Africa/ EPAs Focus on

African regional

and country

levels -

associated with

negotiated

EPAs

Meta-Analysis of

cross-country

studies; for

comparisons,

recalculated

“revenue loss” as

a percentage of

total tax revenues

- "absolute

revenue loss"

calculations taken

from studies, "total

tax revenue" taken

from African

Economic Outlook

database [external

to study].

West Africa: Absolute

losses (Nigeria USD682m),

Relative Losses (Cape

Verde USD24-34m [>60%

of trade revenue, >25% tax

revenue]; Mauritania and

Senegal [>40% of trade

revenue]; Guinea-Bissau

and Gambia [>25% total

tax revenue]). Central

Africa: Tariff revenue loss

significant for all countries

in region, especially (Chad

[>15% of total tax revenue];

Cameroon USD150m

[8.5% total tax revenue];

CAR USD6m [7% total tax

revenue]). Eastern and

Southern Africa: (Comoros

Comparing regions: the worst

affected countries, and therefore

those most likely to face

important domestic revenue

mobilisation issues in the near

future, are concentrated in West

and Central Africa.

26

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

USD8.6m [43% total tax

revenue], Djibouti

Seychelles and Mauritius

[>10% total tax revenue]).

SADC: losses not

significant (>10% total tax)

- Namibia, Mozambique

and Angola most affected.

2012 The Fiscal

Cost of Trade

Liberalization

(see also 2014

paper)

(Cage &

Gadenne,

2012)

Unilateral

trade

liberalisation

episodes

HIC, MIC, LIC Used a panel

dataset of tax

revenues and

government

expenditures in

developing

countries for the

period 1945-

2006; they identify

110 episodes of

decreases in tariff

revenues and

consider whether

countries can

recover those lost

revenues through

other tax sources.

Trade taxes fall by close to

4 GDP percentage points

on average during

episodes of tariff

reductions. Less than half

of the countries recover the

lost tax revenues 5 years

after the start of the

episode.

Tariff revenues are a major

source of public resources in

countries at lower levels of

development. They are a third of

total tax revenues (nearly 5% of

GDP) in LICs, a fifth in MICs and

less than 2% in HICs.

2011 What is the

evidence of

the impact of

tariff

reductions on

employment

(Cirera, et al.,

2011)

Unilateral

trade

liberalisation

episodes

Meta-Analysis of

109 studies (67

econometric; 42

GCE) on impact of

tariff reductions on

Employment and

Majority of CGE simulation

studies (24 of 28): negative

total tax revenue impacts

[or need for increases in

other tax rates to

compensate for lost

On average, a 1 percent

increase in the volume of trade

due to trade reform raises

aggregate employment in the

reforming country by 0.34

27

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

and fiscal

revenue in

developing

countries? A

Systematic

Review

Fiscal Revenue. 8

CGE studies

address both

employment and

revenue effects;

12 CGE studies

exclusively

revenue effects.

revenue].

SR - Tariff reductions likely

to reduce trade tax

revenue; CGE - only very

high import demand and

substitution elasticities

would generate increase in

imports sufficient to

compensate for tariff cut,

Econometric - positive

impact of larger shares of

trade to GDP on total tax

revenue. MR - potentially

positive, especially if

complementary tax policies

and increases in customs

tax collection efficiency are

implemented.

percent (31 studies).

2010 Tax Revenue

and (or?)

Trade

Liberalsiation.

(Baunssgaard

& Keen,

2010)

Unilateral

trade

liberalisation

episodes

Uses panel

data for 117

countries over

32 years,

cleaned for a

variety of

problems in

standard data

sources

Econometric panel

data estimation.

The paper addresses the

question of whether

countries have recovered

from domestic taxes the

revenues lost from trade

liberalization. The findings

strongly suggest that high

income countries have. For

middle income countries,

there are robust signs of

recovery in the long run.

Low income countries

have, on average, only

Low income countries can

struggle to directly replace lost

tariff revenue.

28

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

been able to recover $25c

per dollar of lost trade tax

revenues through domestic

revenues. However,

experiences have varied

widely.

2009 Revenue

Mobilisation in

Sub Saharan

Africa:

challanges

from

Globalisation

(Keen &

Mansour,

2009)

Unilateral

trade

liberalisation

episodes

41 Countries in

Sub Saharan

Africa

Uses a tax dataset

for the period

1980 to 2006,

disentangling tariff

from commodity

tax revenue and

distinguishing

between resource

related and other

revenues.

Most countries that lost

trade tax revenues recover

the loss through indirect

taxes. 21 of the 30 LICs

had higher tax revenues

than before liberalisation

while 9 did not replace the

trade tax loss. 9 of the

MICs increased tax

revenue, while 2 did not.

Most countries that lost trade tax

revenues recovered the loss,

and in many cases were able to

increase total taxes through

indirect taxes. The introduction

of VAT in the 1990s accounted

for over 60% of tax increases.

Though not explicitly addressed,

tax revenue recovered, on

average, in a three to five year

period.

2008 Effect of

Import

Liberalization

on Tariff

Revenue in

Ghana

(Brafu-

Insaidoo &

Obeng, 2008)

Unilateral

trade

liberalisation

episodes

Ghana Econometric

regression: SR

and LR

quantitative effect

analysis using

estimated

aggregate real

imports equation;

buoyancy and

elasticity of tax

system; relative

effects of changes

LR estimation: liberalisation

has both direct and indirect

effects on tariff revenue. A

1% reduction in average

tariff rate directly caused

revenue loss of 1%, but

improves tariff revenue by

causing an upsurge in

imports by 0.26%. The total

net effect of a 1% reduction

in the average tariff rate is

a revenue loss of 0.74%.

Import tax is neither buoyant nor

elastic in Ghana over the entire

period, suggesting the continued

existence of administrative

inefficiencies in the country’s

customs duty collection (and

became less elastic and buoyant

over liberalisation period). Long

run findings on tariff liberalisation

effects suggest that SR duty

revenue losses will be sustained,

as direct revenue loss from tariff

29

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

to import tax

policy, and tax

evasion, using

decomposition

analysis.

SR estimation: effects do

not differ substantially from

LR solution.

rate reductions outweighs the

revenue enhancing effect. The

main contributors to overall

import tax change for the

liberalization period were real

currency depreciation and

reductions in average official

tariff rate.

2007 Evaluating the

Revenue

Effects of

Trade Policy

Options for

COMESA

Countries: The

Impacts of a

Customs

Union and an

EPA with the

European

Union

(Brenton, et

al., 2007)

COMESA/EPA COMESA

(Ethiopia,

Madagascar,

Malawi,

Zambia)

Simple partial

equilibrium

approach;

simulations of full

liberalization of

COMESA-

COMESA imports

and EU-COMESA

imports - at tariff

line level. Three

Scenarios: (1)

EPA - Importance

of Tariff

exemptions on

Revenue impact;

(2) EPA - changes

in total Tax

Revenue (incl.

VAT and excise

tax on imports);

and (3) cumulative

revenue impacts

of COMESA FTA,

Scenario (1) - (Applied

Tariffs; Statutory Tariffs;

Applied tariffs + removal of

tariff exemptions): Ethiopia

(-17.2%; -32.6%; -4.6%),

Madagascar (-29.9%; -

43.9%; -12.1%), Malawi (-

6.5%; -8.5%; +26.2%),

Zambia (-17.7%; -23.2%;

+24.6%). Scenario (2) -

(Applied Tariffs; Statutory

Tariffs; Applied tariffs +

removal of tariff

exemptions): Ethiopia (-

3.4%; -6.4%; -0.8%),

Madagascar (-4.1%; -5.6%;

-1.7%), Malawi (-0.8%, -

1.1%, +3.3%), Zambia (-

1.7%; -2.3%; +2.3%).

Scenario (3) - (FTA; CU;

EPA; Total Effect): Ethiopia

(-1.0%; -3.6%; -2.4%; -

7.0%), Madagascar (-0.8%;

It is important to control for tariff

exemptions in simulations, also

the predicted losses are

substantially higher for all

countries when statutory rates

are used in simulations.

If tariff exemptions are

withdrawn from the analysis, the

impact of tariff reductions on

revenue loss is much lower (and

revenues grew in Zambia and

Malawi because of reductions in

extensive tariff exemptions).

30

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

COMESA CU, and

EPA.

+4.3%; -4.9%; -1.4%),

Malawi (-0.8%; +1.2%; -

2.2%; -1.8%), Zambia (-

0.2%; -2.1%; -0.8%; -

3.1%).

2005 Impact of

Changes in

Tariffs on

Developing

Countries’

Government

Revenue

(Kowalski,

2005)

Unilateral

trade

liberalisation

episodes

Selected

developing

countries

(African

Countries:

Zambia,

Malawi,

Zimbabwe,

Tanzania,

Uganda,

Madagascar,

Mozambique,

Morocco)

Simulations of

reduction of tariffs

- partial

equilibrium model

(changes in

consumer surplus

and tariff

revenue); general

equilibrium

approach (capture

substitution effects

and welfare

effects) using

GTAP model

(revenue, trade,

welfare) for 3

Swiss formula

scenarios [15, 10,

5]

All countries have negative

revenue impact (Swiss 5,

10, 15). On Swiss 10,

comparison of impact on

revenue and welfare

estimates (GE simulation;

PE simulation): Revenue

(%) - Zambia(-17.5; -39.2),

Malawi (-16.6; -46.1),

Zimbabwe(-43.6; -49.5),

Tanzania(-37;-41.8),

Uganda(-10.8;-26),

Madagascar(-4.5;-13.4),

Mozambique(-11.8;-25.3),

Morocco(-55.4;-65.2);

Welfare (US$m) -

Zambia(21.8;4.4),

Malawi(10.2;2.7),

Zimbabwe(52.8;12.2),

Tanzania(-26.7;9.6),

Uganda(-10.5;1.2),

Madagascar(4.5;0.2),

Mozambique(6.8;3.5),

Morocco(240.7;326.7).

While the removal of quantitative

restrictions, tariffication of quotas

or reduction of non-tariff barriers

can preserve or increase

government revenues without

major tax system reform, a

complete removal of tariffs will

inevitably lead to a loss of tariff

revenue. Many developing

countries’ applied tariff

schedules are characterised by

dispersed tariff rates in low

import demand elasticity sectors

and high tariff rates in high

import demand elasticity sectors

- this is favourable when high

rates are applied to low elasticity

products; this scenario will

lessen negative revenue

impacts. The need for co-

ordination of tariff reforms with

other tax policies is particularly

evident in developing countries

where trade taxes account for

significant shares of public

revenues and GDPs. Failure to

31

Year Title Ref (Author,

date)

CFTA / Other

FTA /

Unilateral

Country

Coverage

Methodology Fiscal Impacts (Other

Impacts)

Comments

(Recommendations,

Observations)

take fiscal constraints into

consideration can be one of the

principal causes for

unsuccessful trade reforms.

Broader tax reforms are needed

to replace potentially lost tariff

revenue (e.g. income, sales and

VA tax - widespread adoption of

VAT globally linked to this).

2001 Regional

Integration:

Observed

Trade and

Other

Economic

Effects

(Grosso, et

al., 2001)

SADC FTA General; not

country-

specific; one

study (World

Bank, 2000) on

SADC

Literature review

of the main

findings of

empirical work

already

undertaken on the

effects of RTAs on

flows of trade and

investment, and

their impact on

economic welfare.

Changes in tariffs and

trade volumes will lead to a

loss of government tariff

revenue, which can be a

significant cost especially

for non-OECD countries

where trade taxes are an

important source of

government revenue. In

SADC, tariff revenue

amounted to 9.8 percent

and 5.6 percent of

government revenue for

Zimbabwe and Zambia.

.

32

33

6.2 Tabulation of Findings: Characteristics of Fiscal Loss Compensation

Mechanisms in Regional Trade Agreements

This table is adapted from Walkenhorst (2006), and updated with information from the studies reviewed. The table breaks down the

characteristics of the compensation mechanisms employed in different RTAs for cross-comparisons.

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

CARICOM Common

Market;

15(8 as

defined in

the

Treaty)

Imposition of

intra-regional

tariffs collected

by the affected

Member State.

Overseen by the

Council for

Trade and

Econ.

Development

(COTED)

On application

by less

developed

members to

COTED.

Retaining of

revenues from

intra-regional

tariffs.

(Article 160 of

the Treaty.)

Temporary,

according to

decision by

Council for

Trade and

Economic

Development;

members must

apply case by

case for

(Walkenhor

st, 2006)

COMESA

and EAC

FTA;

20(13)

Resources

mobilized from

donors (EU

[Euro111m],

World Bank,

AfDB). Over

same period,

member state

Two facilities set

up under

COMESA fund

also available

for EAC:

Infrastructure

Fund and

Adjustment

Compensation

based on

estimates of

fiscal losses

from

implementation

of commitments

and obligations.

Sustainability

and ownership

of fund

threatened by

overreliance on

donor funds and

very low

member-state

No timeframe,

but

compensation is

based on

country’s

demand and

each must justify

the value of

Rwanda’s taxes

on international

trade actually

rose following

the adoption of

the CET. The

growth came

from VAT on

(Walkenhor

st, 2006)

(Bilal, et al.,

2012)

(Marape,

2006)

(Olympio &

Wangwe,

34

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

contributions

(to base fund)

total Euro7.6m.

Facility.

Assessments of

revenue loss to

be

compensated

under

programme are

made ex post.

COMAid forms

management

arm of RISM.

Under calls for

proposals,

Burundi received

Euro12.7m and

Rwanda

Euro22.6m to

compensate for

potential fiscal

losses as a

result of applying

the EAC

Customs Union.

contributions. revenue loss

over time.

more imports

and an excise

levy on petrol

which replaced

tariffs.

A Mid Term

Evaluation of

RISM called for

a shift away

from revenue

loss

compensation to

an adjustment

support facility.

2010)

(Healy &

Baker,

2014)

ECCAS FTA; 11(8) Community

levy of 0.4% on

imports from

third countries.

Levy is paid into

ECCAS

accounts at

Central Banks

as a Community

Integration

Contribution

(CCI) for the

Compensation

Fund.

50% of the CCI

is for lost tariff

revenue on intra-

regional imports,

50% for the

operation of the

block. No

payments to

date as FTA is

yet to be fully

implemented.

To ensure actual

payments,

Central Banks

are authorised to

make payments

to ECCAS

automatically.

(Walkenhor

st, 2006)

(Bilal, et al.,

2012)

(AFDB,

2005)

(WTO,

2013)

35

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

CEMAC 6(2) Community

levy of 1% on

imports from

third countries

– the Taxe

Communautair

e d’Integration

(TCI). There

are also

transfers from

oil revenues

and support

from the World

Bank.

30% of the TCI

goes to support

CEMAC

institutions. 70%

goes to the

Fonds de

Développement

de la

Communauté

(FODEC). The

FODEC is

managed by the

Banque de

Développement

des États

d’Afrique

Centralé.

40% of FODEC

money is for

fiscal

compensation.

60% for regional

integration

projects.

TCI payments

have been in

arrears.

Political buy in is

needed to

ensure financial

support.

(AFDB,

2005)

(WTO,

2013)

ECOWAS FTA;

15(11)

Community

Levy of 0.5%

on imports

from third

countries.

Member states

must apply to

Executive

Secretariat for

compensation.

ERDF manages

community levy

reserves, and

responsible for

granting

concessionary

Lost tariff

revenue from

intra-regional

imports as a

result of

liberalization

(Article 48),

according to

customs

declaration.

Compensation

Some members

slow to

implement

commitments;

pay contributions

to the

compensation

fund. This

inability to

mobilize

resources from

Four years (from

Jan. 2002),

degressive

compensation

(100% of loss in

2002, 80% in

2003, 60% in

2004, 30% in

2005, 0% of loss

from 1 January

2006.)

To better fund

infrastructural

development,

ECOWAS Fund

restructured into

holding

company

ECOWAS Bank

for Investment

and

Development

(Walkenhor

st, 2006)

(ECOWAS,

2003)

(Gathii,

2009)

(Marape,

2006)

36

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

and non-

concessionary

loans.

determined by

Protocol Relating

To The

Application Of

Compensation

Procedures For

Loss Of

Revenue….

Rev1(b) 2002.

In 2002, Benin

received over

373m CFA

Francs as

compensation

for revenue lost

in 1998-2000.

within and

outside

ECOWAS

resulted in

limited

infrastructural

development

activities.

(EBID) (with two

subsidiaries:

ECOWAS

Regional

Development

Fund (ERDF);

ECOWAS

Regional

Investment Bank

(ERIB))

WAEMU /

UEMOA

Economic

and

Monetary

Union;

8(7)

Community

Solidarity

Surcharge of

1% charged on

imports from

outside region.

Efforts also to

mobilize

resources

through grants,

Managed by

UEMOA Fiscal

Transition

Programme. (47

measures

aimed at

compensating

for the loss of

fiscal revenues

and at

Lost tariff

revenue from

intra-regional

imports (on

industrial goods),

according to

customs

declaration.

Fund is

allocated: 50%

Crisis in Cote

D’Ivoire resulted

in a shortfall of

CFA 6 billion in

2003 & 2004.

Compensation in

place for 9,5

years (1 July

1996 to 31

December

2005). UEMOA

RISF was

operational for 6

years. It

progressively

Parts of funds

raised are used

for other

purposes than

revenue loss

compensation.

FLCM was

Implemented

efficiently.

(Walkenhor

st, 2006)

(Bilal, et al.,

2012)

(Marape,

2006)

(UNECA/

AU, 2006)

37

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

donations and

loans (donors

and financial

institutions).

effectively

mobilizing

domestic

resources).

Various support

structures to aid

project

implementation.

UEMOA and

ECOWAS

harmonised

their process of

application for

compensation.

for economic

infrastructure,

30% for social

infrastructure,

20% for

adjustment

support – limited

to poor

countries.

Niger received

CFA 20 billion

for losses

between 1995

and 2002.

reduced tariffs

and paid

compensation

on actual loss of

revenue: in the

first period

(1996-1997)

UEMOA tariff

rate reduced

from 20% to

14% so

compensation of

6% made to

states with loss

of revenue

[second period:

tariff 8%

compensation

12%; third

period: tariff 4%

compensation

16%]. This

continued to

2000 (customs

union achieved).

Compensation

changed:

compensation

then reduced

38

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

year on year:

100% in 2000-

2002, 80% in

2003, 60% in

2004, 30% in

2005, 0% 2006.

IE

SACU Customs

Union;

5(1)

Common

Revenue Pool

of all customs

and excise

duties –

member states

make annual

budgetary

contributions.

Revenue

sharing

formula (RSF)

provided for in

Article 34 of

the SACU

Agreement of

2002. RSF

revised to

prioritize

development.

Funds held in

Common

Revenue Pool;

categorized into

3 components

for distribution

(customs,

excise and

development).

The costs of

SACU are

covered via

deductions from

the “gross

amounts of

customs, excise

and additional

duties collected,

before

distribution”.

Revenue

Sharing

Formula:

Customs pool -

revenue sharing

based on share

of intra-SACU

imports; Excise

pool - each

member gets

20% of the

development

fund. Remaining

excise revenue

is distributed

proportionately

to members’

GDPs.

Indefinite

duration

Parts of funds

raised are used

for other

purposes than

revenue loss

compensation.

Revenue-

sharing

arrangements

dictate a

considerable

amount of

implicit

redistribution of

fiscal revenues

from South

Africa to the

BLNS.

(Walkenhor

st, 2006)

(Bilal, et al.,

2012)

(Gathii,

2009)

(Flatters &

Stern,

2005)

39

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

Ghana EU

EPA

FTA EU to provide

supportive

measures on

net fiscal

losses as a

result of the

EPA.

EDA

Development

Programme.

Contribution in

significant

proportions to

net fiscal impact

of fiscal reforms;

support to fiscal

reforms.

Contribution to

take the form of

“compensation

transfers”.

EU will provide a

specific amount

of compensatory

finance to cover

fiscal losses

based on results

of a CGE model.

Unclear how this

clause will be

implemented, or

specific

interpretation of

‘supporting

measures’ and

‘net fiscal

losses’.

Undefined

period; principle

is that

compensation

transfers from

the EU will be

available to the

point that net

fiscal impact of

EPA is

neutralized.

(Bilal, et al.,

2012)

Adjustment/Development Funds

The

African

Developm

ent Fund

ADF of the

African

Developm

ent Bank.

Funds

projects of

40

regional

low-

income

African

countries.

Shares

held: 50%

by

regional

members

Has been

successful in

mobilizing

resources from

the non-

regional

members who

replenish its

resources

every three

years

ADF provides

budget support

for financing

country-driven

economic

development.

Recommended

that the bulk of

funds mobilised

should finance

priority

infrastructural

projects which

Based on

Country

Performance

Assessment

(CPA): channels

resources to

countries where

the policy and

institutional

environments

are most

conducive for

sustainable and

Loan projects or

policy based

loans: repayable

over fifty (50),

grace period of

ten (10) years.

Credit lines

repayable over

twenty-five (25)

years, grace

period of five (5)

years. No

interest charged

In 2006, ADF

had successfully

financed

projects in forty

(40) low-income

African

countries. Focus

of ADF on

providing

support for

poverty

reduction and

economic

(Marape,

2006)

40

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

(40), 50%

non-

regional

members

(26).

will benefit more

than one

country.

broad-based

growth

on loans (except

0.75% annual

service charge

on amounts

outstanding, and

annual

commitment fee

of 0.5% on

undisbursed

balances.)

growth efforts in

member

countries.

SADC

Trade

(TRF)

12

Member

States (7

LDCs) 3

countries

(Angola,

DRC,

South

Africa) not

included.

A contribution

agreement

between the

EU and SADC.

A Steering

Committee is

made up of

three Member

States (in

rotation),

representatives

of SADC and

Apex

organisations of

the private

sector and civil

society, the EU

and other

donors. The

programme has

a Facility

Support Unit

Each of the

participating MS

received an

allocation of €1.4

million under the

Trade Protocol

window (total

€16.8million) and

the five SADC

EPA signatories

(Botswana,

Lesotho,

Namibia,

Mozambique

and Swaziland)

a further €1.2

million (total

€6million).

The programme

runs for a total of

five years from

September

2014-September

2019. The total

contract value is:

€31,600,000.

Because of

delays, TRF

implementation

was officially

launched 3

years late at the

SADC Ministers

responsible for

Trade Meeting,

A key challenge

for the Fund was

to elicit

proposals from

Member States.

This was

hampered by

delays in

establishing the

FSU.

MS could

request

technical

assistance in the

preparation of

their project

Interview

with FSU

and SADC

Secretariat.

http://

www.sadc.in

t/files/

4614/3532/8

380/

TRF_Briefin

g_Paper_No

_1.pdf

41

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

(FSU) at the

Secretariat.

14 August 2017. applications.

CARICOM

Developm

ent Fund

(CDF)

15 (8 as

defined in

the

Treaty)

All Member

States

contribute to

the fund.

Funds are also

raised from

donors and

other

institutions (EU

and Caribbean

Development

Bank)

The CDF has its

own legal

identity with a

CEO and CDF

Board. Approval

by the Board for

assistance

programmes is

confirmed by

the Community

Council and the

Council for

Finance and

Planning.

Access is limited

to LDCs and

Guyana. The

Guidelines and

Procedures

focus

expenditure on

competitiveness

promotion.

The First

Contribution

Cycle was 2008

to 2012. It is to

continue

indefinitely.

(Walkenhor

st, 2006)

http://www.

caricomdev

elopmentfu

nd.org

MERCOU

R Fondo

para la

Converge

ncia

Estructural

del

(FOCEM)

4(2 as

defined in

MERCOS

UR)

Argentina

provides 27%

of the budget

for the fund,

Brazil 70%

Paraguay 1%

and Uruguay

2%.

In operation

since 2007, the

Fund has

financed 40

projects.

Paraguay has

received 48% of

the funds and

Uruguay 32%.

Member States

must finance at

least 15% of the

The Fund targets

“internal

asymmetries” in

the endowments

and economic

structures of

member states

to further

competitiveness

and social

cohesion.

Projects have

It has been in

operation since

2007 (approved

in 2005). In 2015

the CMA

renewed the

Fund for a

further 10 years.

(Correa,

2010)

42

Agreeme

nt

Integratio

n; #

states (of

which

LDCs)

Financing

Structure

Management,

governance

mechanisms

Fund Allocation

Criteria

Risks

(financial,

developmental,

political

economy)

Operational

Period

Comments

(Recommendati

ons,

Observations,

Lessons

Learned)

Sources

total project

cost. Project

proposals are

put forward for

approval by the

Common

Market Council

(CMC)

been in housing,

transport, energy

and value chain

integration.

Argentina and

Brazil are limited

to accessing

10% each of the

Fund

43

Extract from ECOWAS: Protocol Relating To The Application Of Compensation Procedures

For Loss Of Revenue Incurred By ECOWAS Member States As A Result Of The Trade

Liberalisation Scheme (Rev 1(b))

Article 2

Under the terms of article 48 of the Treaty, compensation shall be paid to Member

States incurring loss of Customs revenue from imports as a result of the

implementation of the ECOWAS trade liberalisation scheme.

II. EVALUATION OF REVENUE LOST

Article 3

Loss of customs revenue incurred by a Member State is defined as total shortfall in

receipts recorded by the Member State as a result of its importation of approved

originating industrial products.

Article 4

Total loss of Customs revenue incurred by an ECOWAS Member State importing

industrial products originating from another ECOWAS Member State, as a result of

the application of the preferential intra-Community regime shall be assessed

according to the following formula:

where

CRL = (Cde + Str), Vcif

CRL = Customs revenue lost

Cde = Customs duty and taxes of equivalent effect

Str = statistical taxation rate

Vcif = CIF value of product.

Article 5

Customs revenue lostCustoms duty and taxes of equivalent effect statistical taxation

rateCIF value of product.

Shall be compensated, anyimport duties and taxes assessed on the basis of the

44

formula stipulated in article 4 above.

This provisions shall not be applicable to domestic taxes levied on locally produced

goods or goods imported from within the Community.

III. DURATION

Article 6

The duration of the compensation mechanism is fixed at four (4) years, dating from 1

January 2002.

Amounts payable as compensation shall be calculated on a decreasing scale in the

following manner:

Article 7

Customs revenue lost as a result of the imposition of

100% of loss incurred,

in 2002 80% of loss incurred,

in 2003 60% of loss incurred,

in 2004 30% of loss incurred,

in 2005 0% of loss incurred,

with effect from 1 January 2006.

Member States shall transmit to the Executive Secretariat, within a time-limit not

exceeding six months from the date of reference, a list of Customs declarations

processed under the ECOWAS trade liberalisation scheme, accompanied by the

originals of the certificates of origin of each product, and copies of the declarations of

release for consumption.

V. COMPENSATION PROCEDURE Article 8

The Executive Secretariat shall be given a time-limit of 90 days from the date of

45

receipt of applications for compensation, to screen the applications, determine the

amounts payable as compensation and effect payment.

Article 9

Payment of compensation shall be effected by the Executive Secretary who shall

render account thereof to the Council of Ministers.

VI. FINANCING

Article 10

Compensation for loss of revenue incurred by Member States shall be financed from

the proceeds of the Community levy.

46

Continental Free Trade Area Compensation Mechanisms (Draft)

Abstract

Recommended publications

The Demise of WTO Dispute Settlement: Are Dispute Settlement Mechanisms Under Free Trade Agreements...

US trade with Africa: African growth & opportunity?

Making Rules in the WTO: Free or Managed Trade

Free Trade Agreements and the WTO (PowerPoint)

Balancng Basic Human Needs and Free Trade in the WTO

Optimism for the WTO Doha Round: The Bickering Atlantic Bigemony and New Pro Free Trade Coalitions