"R egional T ra d e A rra n g e m e n t in East A frica: E ffe c ts on K e n y a 's

exports s in c e 1977. ^

BY
PA U L O D ERO .O. O T U N G
X50/70923/200f

UKr/ror .T v
V ''I- ...
- I
W.W l

U niversity o l NAIROBI Library

lillliil
0532040 3

Research Paper submitted in partial fulfillment for the requirement of the

iegree of Masters of Arts (Economics) of the University of Nairobi.
 DECLARATION
1 declare that this paper is my original work and has not been presented for a

degree in any other university.

Date t4. f H oc-Tvgff? j o o q

Paul Odero .O. Otung.

This Research paper has been submitted for examination with our approval as
University supervisors.

*■
\
11
 DEDICATION

This work is dedicated to my family.

ACKNOWLEDGEMENTS

I have accumulated a lot of indebtness in the process of writing this research paper. First

sincere thanks to my two supervisors Mr. Jasper Okelo and Dr. Daniel Abala for their tireless

reading through my drafts and giving me positive criticisms. Your valuable comments and

guidance greatly helped to improve the quality of this paper

Secondly, I have greatly benefited from the discussion on the research proposal at the School

of Economics seminar. In this regard, am particularly grateful to Dr. Wambugu and Dr. Mary

Mbithi and my colleagues at the University of Nairobi for their excellent comments.

A lot of gratitude goes to my friends and colleagues at work. Not less deserving appreciation

are my late grandparents Reuben and Flora Otung for their wisdom. May the Lord rest your

souls in peace.

However, I am responsible for errors and views expressed in this paper.

IV
 ABSTRACT

Export-led growth strategy has been a focus for Kenya since the early 1990's. Therefore the

formation of the "new " EAC-RTA raises the question of whether this trade arrangement is of

merit to Kenya. The study reviewed the effect of various RTA's on members countries and

analysed Kenya's exports to Uganda and Tanzania since 1977, when the initial EAC

disintegrated and the eventual revival of the East Africa Community. Using a simple

regression model based on commodity exports, which are generally price focused and the

EAC-RTA dummy variable, the study estimated the factors and the effect of the EAC-RTA on

Kenya exports to the region.

The model estimation using annual data on export value from 1977 to 2008 showed that the

existence of the EAC-RTA and tarmacked road network in Kenya had significant effect on

exports. In addition the results indicated that when investment as a proportion of the GDP,

taken as a proxy for supply side constraints reduces, there is an increase of export value. The

study recommends that, to fully exploit the opportunities in the East African region, road

infrastructure costs, investment constraints in Kenya be reduced and the new East Africa

Community do not disintegrate.

ACRONYMS A N D A BBREVIATION S

ACP Africa, Caribbean, and Pacific

ADB African Development Bank
ASEAN Association of South East Asian Nations
BCI Business Climate Index
BOP Balance of Payments
CET Common External Tariff
CGE Computable General Equilibrium
COMESA Common Market for Eastern and Southern Africa
CU Customs union
EAC East African Community
EABC East Africa Business Council
EPAs Economic Partnership Agreements
ESA Eastern and Southern Africa
El/ European Union
EC European Commission
FDI Foreign Direct Investment
FTA Free Trade Agreement/Area
GATT General Agreement on Tariffs and Trade
IMF International Monetary Fund
Mercosur The Common Market of the South America
MUB Manufacture Under Bond
NTBs Non-Technical Barriers
RoO Rules of Origin
RTA Regional Trade Arrangements
SADC Southern African Development Community
WTO World Trade Organization
OECD Organization for Economic Co-operation and Development
RCA Regional Comparative Advantage
UNCTAD United Conference on Trade and development
TABLE OF CONTENTS
DECLARATION.............................................................................................................................................. ii
DEDICATION ................................................................................................................................................. iii
ACKN OW LEDGEM EN TS..........................................................................................................................iv
ABSTRACT....................................................................................................................................................... v
ACRONYMS AND A B BR EV IA T IO N S................................................................................................. vi
TABLE OF CO N TEN TS.............................................................................................................................. vii
CHAPTER O N E ............................................................................................................................................... 1
1.0 Introduction................................................................................................................................................. 1
1.1 Regional Trade Arrangements............................................................................................................... 2
1.2 Previous East African economic cooperation..................................................................................... 3
1.3 The Statement of the Problem.............................................................................................................. 10
1.4 Objectives of the Study........................................................................................................................... 11
1.5 Significance of the Study........................................................................................................................ 11
CHAPTER T W O ............................................................................................................................................13
2.0 LITERATURE REVIEW........................................................................................................................ 13
2.1 Introduction.............................................................................................................................................. 13
2.2 Theoretical Literature............................................................................................................................. 13
2.3 Empirical Literature................................................................................................................................ 18
2.4 Literature Overview................................................................................................................................ 24
CHAPTER T H R EE....................................................................................................................................... 2 6
3.0 M ETH O D O LO G Y ................................................................................................................................. 26
3.1 Theoretical Fram ework.......................................................................................................................... 26
3.2 Analytical Framework............................................................................................................................ 27
3.3 Model specification................................................................................................................................. 28
3.4 Hypothesis of the Study.........................................................................................................................28
CHAPTER FO U R ..........................................................................................................................................2 9
4.0 Data, Sources and T y p es..................................................................................................................... 29
CHAPTER FIV E ............................................................... ,WW».Ur.y. * . * , * .................... 31
5.0 Data Analysis and Results.................................................................................... 31
5.1 Descriptive Statistics............................................................................................................................... 31
5.2 Stationary an alysis.................................................................................................................................. 32
5.3 Empirical results...................................................................................................................................... 37
5.3.1GeneraI Model 1......................................................................................................................................42
5.3.2Uganda...................................................................................................................................................... 43
5.3.3Tanzania...................................................................................................................................................44
5.3.4Export tren d s.......................................................................................................................................... 45
CHAPTER S IX ............................................................................................................................................... .47
6.0 Summary, Conclusions and Policy implications............................................................................. 47
6.2 Conclusion..................................................................................................................................................48
6.3 Policy Implications.................................................................................................................................. 49
N O TES.............................................................................................................................................................50
7.0 REFEREN C ES.........................................................................................................................................51

vii
List of Tables
Table 1:1: Selected direction of Kenya's Exports and Origin of its Imports, 1992-2008.................. 8

Table 1:2: Principle commodities as a percentage of Exports, 1999-2008............................................9

Table 5:1: Descriptive Statistics.................................................................................................................. 31

Table 5:2: Unit root tests............................................................................................................................... 33

Table 5:3: Correlation of variables of exports to Uganda.....................................................................34

Table 5:4: Correlation of variables of exports to Tanzania...................................................................36

Table 5:5: General model 1 regression results.........................................................................................36

Table 5:6: Model II regression results....................................................................................................... 42

Table 5:7: Model III regression results..................................................................................................... 43

Table 5:8: Model III regression results..................................................................................................... 44

List of Figures

Figure 5:1: Value of Kenya's exports to Uganda...................................................................................45

Figure 5:2: Value of Kenya's Exports to Tanzania.................................................................................46

viii
 CH APTER ON E

1.0 Introduction
Regional trade arrangements are an integral part of international trade, operating alongside

the multilateral agreements under the World Trade Organization (WTO). The last decade has

witnessed growth in the number of regional trade arrangements (RTAs)1. In the developing

countries RTAs account for about 30-40 per cent of all trade agreements that are currently
active (WTO, 2003). Africa is home to about 18 active RTAs with the revival of defunct and

formation of new agreements occurring as part of deeper trade integration schemes. On

average an African country belongs to 4 RTAs (World Bank 2004).

The growth of regional trading blocs has been a major development in international relations

with virtually every country belonging to one or even multiple blocs (Schiff & Winters, 2003).

Where Kenya, Uganda, Rwanda and Burundi belong to EAC and COMESA, Tanzania belongs

to EAC and SADC. These regional groupings provide higher visibility to foreign investment

and possible integration into the global economy, though primarily visualized on lowering
trade barriers between members.

Regional Trade arrangements differ substantially in their treatment of issues such as, labour,

land, capital mobility and investment. Nonetheless, it is possible to characterize the growth in

regionalism in at least three broad separate, albeit related dimensions. Typically the

agreements include those that are "deepening" existing RTAs, where RTAs that originally

focused on "hard" trade restrictions like tariffs and quotas on manufactures and agriculture are

extended to "soft" restrictions such as health and environmental standards, to other product

areas like services and intellectual property, investment and capital mobility.

There is also, the "widening" of existing RTAs, as countries previously not members of any
RTA seek to join one (or more) and the accession of some to the other RTAs. Further, other

regions create new RTAs or relaunch ones that had effectively been dormant. In line with the

trend of new regionalism, a tripartite summit held in October 2008, in Uganda resolved that

three of the RTAs in Africa, namely; COMESA, SADC and EAC with a population of 600

1
million people and an average GDP per capita of U.S $ 1,184 should merge and form a single

regional economic community. The feeling was captured by the former SADC chairman, who
said that a single trading block that will span 26 countries is vital for the regions development.

"We see regional integration as a central component for our development strategy in a

globalized econom y", he said (See East Africa Standard, Financial Journal, pg 7, November 11
2008).

The search for an explanation for the proliferation of RTAs begins with a number of positives.
The intellectual case for free trade is as strong as it has always been, though dampened by the

current global economic recession. Further, the ability to model the RTAs rigorously has added
respectability to the traditional arguments linking trade liberalization and growth. All parties

seem to agree that regionalism is a useful accompaniment to the global multilateral system, as
exemplified by the push for a new EU-ACP EPAs.

The welfare and political outcomes of any RTA cannot be determined a priori (Kreinin and

Plummer, 2002). Both therefore, will depend much on wide range of country circumstances

and the domestic policy choices. Ideally, international trade should stimulate growth in a

number of ways including production and demand linkages, economies of scale due to larger

international markets, increased efficiency, competitiveness, adoption of superior technologies,

and increased productivity through specialization and creation of employment (Basu et nl,
2000, Santos-Pauline, 2000).

1.1 Regional Trade Arrangements

The most common forms of regional integration include Free Trade Areas (FTAs), Customs

unions and Common Markets. The route adopted by most regional groupings, is to integrate

by progressively liberalizing trade relations between members of the grouping. Integration

therefore is a progressive process that goes through: a Preferential Trade Area (PTA); Free

Trade Area (FTA), Customs union (CU), a Common Market and an Economic Union where
there is a common currency and a unified monetary policy. Finally, a Political Union that

represents the final stage of economic and political integration in which the legislative and

judicial process of member states are either unified or federated under consensually agreed

arrangements.

2
There is no easy conclusion on the contributions of trade arrangements to trade and economic

development, it is necessary to investigate whether the formation of EAC-RTA has been trade

enhancing. This study therefore focuses on the EAC-RTA that skipped the FTA stage and

formed a customs union under the principle of asymmetry.

1.2 Previous East African economic cooperation

The history of cooperation between the East Africa countries goes back to the early 1920's on-

account that the countries are neighbors, and their people engage in cross border trade. A
common market between the three territories came into being in stages over a number of

decades. The official formation of the East African Community was in 1967 which cemented

regional co-operation. The aim of the agreement was to "strengthen and regulate industrial,
commercial and other relations to promote harmonious and balanced dei'elopment o f economic activities
where the benefits whereof shall be equitably shared" (Treaty for East African Co-operation, 1967).

However, the life span of the EAC was short; it was dissolved in 1977, following ideological

and economic differences. Kenya undertook a capitalist economic development strategy, while

Tanzania followed a socialist approach. The industrial dominance by Kenya also increased the
tension occasioning the community's eventual disintegration.

1.2.1 Current East African Community:

The EAC trade liberalization program has not followed the traditional sequence of economic

integration from free trade area to customs union. The current cooperation commenced in 1993

with the signing of a Declaration on Closer East African Cooperation, a declaration of intent to

cooperate in virtually all economic, social and political sectors. These efforts resulted in the

formation of the "new " East African Community (EAC). EAC disposed of the intermediate

step of first establishing a free trade area. Thus on 30th November 1999, the current East

African Community (EAC) Treaty was signed by the EAC partner states of Kenya, Tanzania

and Uganda and came into force on 7th July 2000 upon ratification by the three Partner States,

the group was later expanded to include Rwanda and Burundi. The sequence of events
towards full integration of East Africa, according to Article 5-2 of the Treaty, comprises the

establishment of a customs union, followed by a common market, a monetary union and

ultimately a Political Federation. The Customs union became the entry point into the

community and integration process. The envisaged benefits include; wider market for the

3
sectors, especially the manufacturing sector to facilitate FD1, improved access to raw materials

for industrial development, increased trade among the partners and ground for sector

competitiveness. Even though Rwanda and Burundi are members of the EAC, their late entry

makes it difficult to get reliable data, hence this study did not include them due this limitation.

1.2.3 Common external tariff (CET)

The new EAC-Customs union established a Common External Tariff (CET) for the partner
countries in accordance with the provisions of Article 75 of the Treaty. The protocol provides,

three tariff bands on imports of goods originating outside East Africa, zero percent (covering

principally capital goods), 10 percent (covering principally intermediate goods), and 25

percent (covering principally "sensitive sector" products and consumer goods to be protected

from import competition). Therefore the new EAC trade regimes is characterized by tariff

structure that imposes the lowest rates on raw materials and capital goods, moderate rates on

intermediate goods and the highest rates on consumer goods (McIntyre, 2005). Additional

changes under the Customs union include (Bagamuhunda, 2005);

(i) Common duty rates that will apply uniformly on all goods imported into the EAC.

(ii) Zero rates on m ost goods originating and traded within the EAC.

(iii) Reduction to zero rates on goods originating from Kenya and imported by Uganda and

Tanzania. Under the CET, Uganda will eliminate 426 tariff lines and Tanzania 906 tariff lines to

zero. The implementation will be in two phases; First, the adoption of the three-band structure,

with Uganda and Tanzania maintaining tariffs on selected Kenyan imports and then removal
of all internal tariffs by 2010 (McIntyre, 2004).

(iv) Classification of "sensitive items" that the EAC wants to protect from import competition,

these items will attract rates of more than 25 percent

(v) Harmonised commodity descriptions and codes and harmonization of customs

administration to eliminate delays and duplication

(vii) Tax incentives for exporters in the region where duties are waived including export

processing zones, manufacturing under Bond, inward processing and duty drawback for

manufactures for export

4
(viii) Computation of taxes based on a CIF value at the initial port of discharge (either at
Mombasa or Dares Salaam)

(ix) COMESA and SADC preferential treatment will continue to apply on some products for a
period of two years.

(x) The W TO Customs Valuation Agreement that aims at a fair, neutral system for valuation of

goods has been adopted. The agreement gives greater precision to the provisions of valuation
in the original GATT (McIntyre, 2004).

The implementation of the CET has raised challenges in the past as noted by the Chairman of

the EABC, "Uganda and Kenya are still maintaining this duty (on imported goods),but it was

to only affect shopping bags and not all plastic packaging material," he said." The differences

have made goods in the former countries uncompetitive and have resulted to loss of jobs and

closer of industries". In addition the Uganda and Tanzania were granted a stay of the CET for

the Commonwealth Heads of government meeting (in 2007) and the Dar es Salaam rapid port

(Dart) project for imported buses. The countries were allowed to import buses without regard

to existing local capacity." This not only goes against the policy of industrialization, but also

negates the spirit of promoting local manufacturing", he added (see Sunday Nation, July 27th,

2008, EAC Business Fret over new rules breach).This kind of sentiments clearly points out the

hurdles that the implementation of the CET must overcome to realize its main objectives.

1.2.4 Trade Liberalization episodes in Kenya

The link between trade liberalization and economic growth has been a source of both

intellectual inquisitiveness and trade policy formulations (Edwards, 1998), where it is argued
that rapid economic growth cannot be sustained without rapid trade liberalization. Kenya's

liberalization episodes were characterized by both external positive shocks, the coffee price

booms and negative shocks, the oil crises. The government's response to the negative shocks

was to increase controls so as to stem the loss of foreign exchange reserves and its inflationary
consequences. The response of the government to the positive shock was endogenous trade
liberalization (Reinnika, 1996) that precipitated a balance of payments crisis.

5
Increased economic freedom in trade involves lower trade barriers, leading to lower costs and

greater efficiency. Multilateral trade liberalization in the country started with conversion of

quantitative restrictions to tariffs equivalent. There was phased tariff reduction such that by

1992 quantitative restrictions affected only 5% of imports compared with 12% in 1987 (Swamy,
1994). Over the 1987-1992 period, the number of tariff categories and maximum tariff rates

were reduced from 25 to 11 and 170% to 70% respectively. Further, the number of tariff bands
(including duty free was reduced from 15 in 1990/91 to 4 in 1997/98 and top regular tariff
rates from 100% to 25% over the same period (Mwega, 2002). The most significant shift in

trade policy occurred in 1993 with the abolition of trade licensing requirements and foreign
exchange controls (Ndungu, 2000 and Were et al. 2001).

Over the 1993-95 periods, all current and capital account restrictions were lifted. The response

on imports and exports were immense, with exports responding more due to the combined

effect of devaluation of the Kenya shilling in 1993. In totality this raised the export earnings

dramatically in the early 1990's from 13% of the GDP in 1992 to over 20% between 1993 and
1996 (Glenday and Ndii, 2000).

The regional trade integration measures under the East Africa Cooperation and the wider

COMESA also accounted for the dominant share of increase in Kenya's exports. Recorded

exports to COMESA increased from an average of 15% for the period 1990-1992 to 34% in 1996-
98 (Glenday and Ndii, 2000).

The evidence in favour of export lead growth (ELG) and trade liberalization appears to have

influenced the country to adopt export-lead growth strategy in the early 90's, to emphasize

manufacturing for export markets through various schemes such as export compensation,

manufacturing under bond (MUB) and export processing zones (EPZ). However, the impact of

export incentive schemes especially MUB and EPZs designed to target dedicated export
processing for outside markets has not been significant .The country has not been successful in

gaining competitiveness in labour-intensive export processing (Glenday and Ndii,2000).

6
1.2.5 Kenya Trade direction

Kenya trades with both developed and developing countries, especially the Southern African

developing countries because of the country's strategic position as a regional industrial hub

with access to the sea. The geographical distribution of her trade since the onset of trade

liberalization in the early 1990's, shows that the EU has been the dominant market for Kenyan
exports followed closely by the EAC as shown in the table 1.1 below.

7
 Table 1:1: Selected direction o f Kenya's Exports and Origin of its Imports, 1992-2008

jp h ic a l
and 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
ry
c in

ts 4 44.0 39.1 34.8 33.7 35.3 34.3 31.7 31.3 29.8 27.1 27.1 28.5 26.4 24.0 26.1 26.4 25.9
IS * 36.2 36.6 35.4 40.4 37.7 322 326 328 30.5 24.8 322 23.7 23.9 21.0 22.0 20.0 17.3
ca
4.2 4.7 4.3 3.6 3.5 3.5 3.3 27 27 29 2.4 21 28 5.1 8.6 7.5 6.4
ts 10.9 7.8 8.0 6.9 7.4 9.5 13.0 9.7 6.0 15.7 7.4 6.4 6.7 11.7 6.2 9.2 5.7
roa
ts 4.2 7.0 10.2 11.8 120 122 120 11.2 8.2 9.2 8.4 4.0 8.3 7.7 7.3 8.1 8.5
ts 0.5 0.5 0.9 0.4 0.6 0.5 0.3 0.2 0.4 0.2 0.03 0.5 0.6 0.7 0.9 1.1 0.9
a

6.7 9.0 12.2 15.1 15.4 14.3 15.2 17.3 18.0 20.4 18.5 16.7 17.3 16.4 11.1 12.2 123
$ 03 0.3 0.2 0.1 0 .0 0.2 0.0 0.2 0.2 0.2 0.3 0.4 0.3 0.3 0.2 1.0 0.7
of
;s a 3.5 3.4 3.5 3.4 3.0 3.1 3.2
s 10.9 14.0 16.6 13.9 125 13.3 14.2 15.3 15.9 16.8 0.4 0.4 0.4 0.3 0.3 0.5 0.4
ts 21 1.6 1.5 0.7 0.7 29 0.9 1.4 1.5 3.5
t" A frica
ts 4.0 4.1 4.5 6.1 5.4 4.3 3.9 29 3.9 28 15.4 14.7 15.0 15.3 18.1 18.5 18.2
ts 0.3 0.1 11.2 7.5 8.3 11.6 7.5 9.0 7.1 7.0 3.7 4.1 4.0 3.5 4.6 4.0 3.5
Source: K N BS Econom ic Surveys

as a share o f total exports

as a share o f total imports

The share of Kenya's exports going to the EU over the period (1990-1999) declined by about

3.9% annually and has continued to decline since. While the share of exports to the rest of

COMESA (excluding the EAC) has shown an increasing trend over time, with the share of

exports to the rest of Africa being more or less constant. The share of exports to EAC (Tanzania

and Uganda) has increased from 7.4% in 1990 to 28.5% in 1999, representing an impressive

growth rate of 14.4% per annum. The significant increase in exports to the EAC began in 1993,

the year when Kenya made significant liberalization of her trade regime under the structural

adjustment programme and the signing of a Declaration on closer East Africa cooperation.

1.2.6 Structure and Composition of Exports

Like most Sub-Saharan Africa countries, Kenya's export structure is predominately composed
of primary commodities-mainly, tea, coffee and horticulture. The share of manufactured

exports has not only remained small but has been declining. The decline in Kenya's export

8
 performance is mainly attributed to past policies that produced an anti-export bias (Wagacha,

2000). Consequently, export growth has been erratic, based on fluctuations in earnings from a
few traditional primary exports and the tourism sector. This over reliance on primary

commodities exposes the export sector to external shocks such as fluctuations of world market

price and vagaries of weather. An export growth strategy that is based solely on the

exploitation of agricultural resources is not good for long-term development due to limited

scope for technological improvements and skill developments offered by such a strategy. Table

1.2 below shows various commodities, as a percentage of exports since the signing of the EAC
Treat)' in 1999.

Table 1:2: Principle commodities as a percentage of Exports, 1999-2008

IT E M 1 9 9 9 (% ) 2 0 0 0 (% ) 2 0 0 1 (% ) 2 0 0 2 (% ) 2 0 0 3 (% ) 2 0 0 4 (% ) 2 0 0 5 (% ) 2 0 0 6 (% ) 2 0 0 7 (% ) 2 0 0 8 (% )

Tea 2 2 .8 5 2 9 .3 9 28.41 2 6 .1 6 2 4 .1 4 2 3 .1 2 2 0 .1 9 2 0 .7 5 17.87 19.78

1 iorticulture 12.29 1 7 .7 0 1 6 .3 9 2 1 .5 6 2 6 .6 9 2 8 .3 4 2 1 .3 0 2 1 .3 9 21.71 2 2 .0 6

_________
Petroleum 6 .3 6 7 .8 5 1 0 .1 3 2 .9 7 0 .0 5 0 .7 1 3.08 1.88 2 .9 5 1.31
Products

Coffee 8.34 9 .7 7 6 .1 8 4 .9 8 4.6 4 .4 5 4.3 2 4 .0 0 3 .9 8 3 .1 4

Soda Ash 0.91 1 .1 7 1 .6 5 1 .6 2 1.75 2 .0 3 1.84 1 .7 4 2.07 4 .0 9

Cement 0 .8 7 1 .1 7 0 .8 2 1 .1 3 1.45 1 .2 6 1.36 1 .6 9 1.76 2 .2 5

Pyrethrum 0.4 6 0 .5 8 0 .8 2 0 .6 1 0 .5 9 0 .6 0 0.53 0 .4 8 0 .0 6 0

Extract

Sisal 0 .4 4 0 .5 1 0 .6 0 0 .6 0 0 .6 6 0 .7 2 0 .5 6 0 .5 5 0.53 0.4 6

fluorspar 0 .3 6 0 .5 4 0 .5 4 0 .5 6 0 .4 9 0 .5 7 0.58 0 .5 0 0.3 7 0.6 7

Hides and 0 .2 2 0 .4 1 0 .5 2 0 .3 4 0 .4 0 0 .6 1 0.41 0 .2 7 0.0 5 0.01

Skins

AD Other" 3 1 .3 7 1 2 .4 4 13.51 1 6 .9 18.53 2 0 .0 7 2 6 .1 8 2 4 .9 8 2 6 .7 2 2 5 .0 6

Total 100 100 100 10 0 100 10 0 100 100 100 100

E xpons

Source: Ministry o Trade, <EPLOTRADE sectoral bac (ground studies 2008.

f he emerging pattern is one of a highly concentrated export structure; an export destination of

a few traditional and dominant markets and an insignificant share of processed products in

the export basket. This calls for increasing the product range, moving up the ladder in agro­

9
processing and expanding exports within regional markets and other destinations. A more

diversified export structure reduces vulnerability to external shocks.

1.3 The Statement of the Problem

The East African Trade Arrangement was designed using the principle of asymmetry2, hence

retained tariffs on some products exported by Kenya into the member countries to protect

their industries. This has resulted in conflicting interpretation and breach of the CET in the

past, like levying exercise duty on plastics by some member countries. The government also
acknowledges that weak negotiating capacity impedes the country's ability to negotiate for

favorable trade agreements and therefore creates barriers against the country's exports (GoK,
Vision 2030).

Due to a series of external shocks occasioned by the oil crises in 1973 and the 1980's resulting

into balance of payments (BOP) problems and the collapse of the initial EAC in 1977, which

impacted negatively on merchandise exports, the government shifted from import-substitution

to export-promotion strategy by 1974.

Despite previous export growth strategies that were designed to promote manufacturing for

exports, the contribution of the processed exports to the GDP has remained at about 9 per cent

since the 1960's (Ministry of Industrialization, Strategic Plan 2008-2012).

But according to the Kenya's Vision 2030 economic strategy, the government plans to increase

the country's regional export share from 7 per cent to 15 per cent by 2012, through increased

capacity utilization and elimination of impediments to Kenya's competitiveness in the region.

However the East African market is currently dominated by imports from outside the region,

while the overall business climate index declined in the year 2008 (EABC 2008).

This research paper sought to address the problem of how the EAC-RTA has affected Kenya's

exports since 1977.

10
1.4 Objectives of the Study
The main objective was to analyse the effect of the East Africa Community trade arrangement

on the country's exports and propose policies based on the findings.

The specific objectives included:

1) To analyse the effect of the EAC-RTA on Kenya's exports to the region from 1977 to
2008.

2) To identify and quantify the factors which determine Kenya's exports to the region.

3) To recommend policies that promote trade based on the study findings.

1.5 Significance of the Study

Through forming the EAC, the expectation is that the RTA should facilitate trade, labour and

capital movements, reduce the cost of doing business, increase investment and thereby
increase the aggregate economic activity of its members.

The EAC regional integration appears to be strong and moving towards deeper integration

with the implementation of the customs union and a desire for political integration come 2010.

However, the EAC-Customs union presents uncertain economic impacts on member countries.

For instance, there have concerns by the trading partners of the adverse affects under the

customs union on their industries by import competition from Kenya which is viewed to have

a relatively advanced industrial sector.

There are also past incidences where different countries have interpreted the CET differently,

like levying exercise duty on plastics and failure to reduce transportation costs may have

resulted into a decline in the business climate index (BCI)3 in the region (EABC 2008).The BCI

is an initiative of the East Africa Business Council with the objective of giving a platform for

the business community to provide necessary inputs leading to the elimination of Non Tariff

Barriers (NTB's) and improvements of other business climate. The decline in BCI is significant

as it indicates that there was a drop in the business climate. Therefore whereas intra-regional

trade should be promoted, the tariff reduction alone under the customs union may not
generate high interregional trade without elimination of the NTBs. Some studies show that the

11
implementation of the EAC-CU with maximum tariff of 25 per cent may cause trade diversion

Lucio Castro, et al. (2005).

The total trade (exports and imports) is predicted to have marginal gains of about 2 and 3 per

cent respectively Okelo (2006). These creates doubt about the merit of the new trade

arrangement and its possible contribution the country's economic development

With the current EAC in its 9th year of operation, it is an opportune time to examine what

effect, if any, it has brought about thus far. It is with this in mind that the study seeks to
evaluate the contribution of the RTA on Kenya's exports, resulting from the expanded regional

market. This was informed by the fact that export lead growth brings in technology transfer,

and efficient allocation of resources imposed by international competition. The study is

therefore significant as it will inform Kenya's trading policy within the East African region.

12
 CH APTER TW O

2.0 LITERATURE REVIEW

2.1 Introduction
This chapter attempted a review of the literature which have attempted to explain the

formation of RTAs and their effect on member countries. According to various authors,

economic groupings that represent varying degrees of integration tend to reduce the trade
barriers between them resulting in easier flow of factors of production. The reasons advanced

for these regional groupings include proximity of countries, economic size of the countries, per

capita income and the international terms of trade among other factors. Although these are

some of the reasons for regional groupings, within the East African region Kenya has been

seen to have a competitive advantage, in part due to earlier industrial impetus and export

growth strategies. The commodity composition of East African intraregional trade reveals that

unlike trade with the rest of the world, manufactures play an important role. McIntyre (2005)

indicates that for Kenya 11.5 percent and 43.4 percent of its imports from Uganda and

Tanzania, respectively, are manufactures. For Uganda, 33.8 percent and 71.3 percent of its

imports are manufactures from Kenya and Tanzania, and for Tanzania 56.8 percent and 16.6

percent of its imports are manufactures from Kenya and Uganda. In short, the expansion of

intraregional trade has provided a market for Kenya's manufacturing sectors in the EAC
member states, particularly Kenya.

2.2 Theoretical Literature

Ihe assumption in trade theory, though not necessarily true is that a customs union, for

example would be welfare improving since tariffs, which are in general welfare reducing

would fall under such arrangement. However, in the static influencing concept of trade

creation and trade diversion, Viner (1950) showed that a customs union effect on welfare is

ambiguous. The basic Viner model was hinged primarily on international trade theory which

assumed that, import-competing goods may be produced under increasing marginal cost
conditions, while exportable goods are produced under constant cost conditions in each
country.

13
Since Viner's seminal analysis, trade creation and diversion have been treated as synonymous

with the impact of customs unions and other regional integration arrangements. If the effect of

increased trade shifts production from low-cost producers outside the trading bloc to high-cost

producers within the bloc, this would result into trade diversion. But if the increased regional

trade leads to the shifting of production from less efficient, high-cost producers to more

efficient, low-cost producers within the union, the agreement would be trade creating. Thus a

trade union will be beneficial if on balance it is "trade creating" and harmful if it is "trade

diverting". Dollar, (1992) noted that a well crafted trade bloc can increase competition in
domestic industries and spur productive efficiency gains which improve the quality and

quantity of inputs and goods available to the economy.

Bhagwati (1993) first questioned the validity of this assertion, pointing to an earlier important

contribution by Lipsey (1957) that had spelt out the welfare improvement criteria in a specific

model that differed from the ones defining the "natural trading partners." Subsequently,

Bhagwati and Panagariya (1996) offered a systematic critique of the natural trading partners
hypothesis. Regarding proximity as the basis of welfare-improving unions, they demonstrated

that ceteris paribus, a union with a proximate partner could be more harmful than with a

distant one. Regarding the volume-of-trade criterion, following Panagariya (1996), they

pointed out that there was a presumption that the more a small country imported from its

union partner, the more it would lose from liberalizing preferentially. As long as the country

continued to import from the outside world, the price facing its consumers and producers

would not change. Therefore, it would fail to reap any efficiency benefits that accrue via the
decline in the internal price when liberalization is non-discriminatory. Instead, the country

would lose tariff revenue on good imported from the partner country with the lost revenue

transferred to the latter's exporters.

In assessing the effects of forming an effective RTA, the efficiency gains of economic

Integration depends on whether the products produced by members of the RTA are in direct

competition with, or complementary to, each other. But to be competitive or have efficiency

gains in an RTA, there must be a considerable overlap in the range of commodities that

14
members of the RTA produce. In this case, complementarities' exists when members of RTAs

produce commodities or products that do not compete much with the local production of

other RTA members. Traditional integration theory contends that, in the case of

complementary economies, economic integration will have the usual trade diversion and trade

creation effects; the higher the barriers to trade with non-members, the higher the risk of trade

diversion.
Khandelwal (2004) developed estimates of bilateral product complementary indices4 in

COMESA and SADC. The results indicate that, within COMESA, product complementarities
between Kenya's exports and the imports of the other member countries average 38.6. For all

other countries, except Egypt, average product complementarilv for exports is far lower, EAC

member Uganda, had an average of 19.8. The trading patterns of the EAC members however,

indicate that trade linkages may be relatively weak hence there is no easy way to characterize

these economies as either complementary or competitive.

The partial equilibrium nature of the Vinerian framework is not wholesome in explaining the
impact of RTAs. In comparative advantage theory neighbouring countries are unlikely to be

internationally competitive producers of the large number of products consumed by residents

of most countries. This consideration would limit the number of cases in which regional

integration arrangements among small countries to be trade creating owing to member

countries' international competitiveness in various products. Also, the assumption of

unlimited capability to produce exportables at constant unit costs is extreme and cannot hold

under various production conditions.

Other studies have used gravity models to assess the impact of regional agreements on trade

flows. Based on the law of gravity, the gravity model equations predict that the volume of

trade between two economies should increase with their size (proxied by real GDP) and

decrease in transaction costs (measured as bilateral distance). The standard gravity model
predicts that the volume of trade between any two countries i and j, (v. exporting country j:

importing country) is a function of the country's GDP (Mi,Mj), distance between i and j(D) and

gravitational constant(G). Such that

!5
That is if [3,u=l and 6=2 to give the Newton's Law which the gravity model is anchored on.

Applevard and Files (2001) notes that the gravity model5 provides a multivariate framework

for assessing the impact of RTAs on the level in terms of volume rather than composition and

what drives one country to export to another. The model predicts that a country's absolute

trade potential depends on its total economic size as well as other factors such as land area,

population, geographical distance, cultural similarities, policy and political ties (Kirkpatrick &
Watanabe, 2005). These studies showed that the volume of trade is positively related to the
national incomes of trading partners, and a decreasing function of the distance between them.

The assumption being that, in the absence of a regional trade agreement, members' trade will

be proportional to the gross domestic product (GDP) and the distance between them.

Kristjansdottir, (2005) indicates that GDP reflects the capacity to supply exporting goods and

demand for exports by the importing country, which is assumed to increase as its GDP

increases. Other studies show that gravity model works best for countries that are similar, like
the East African region and have considerable intra-industry trade with one another (Helpman

1990). In Yarmrick and Gosh (2001), using gross domestic product and distance as the

variables to capture the bilateral trade between two countries, the study concludes that RTAs

are trade creating.

Even though the gravity model has gained wide application, a number of fundamental issues

have been raised on its usage. The model appears to exist in a vacuum due to its lack of solid

theoretical micro-economic underpinnings. The postulation that bilateral trade between

countries is dependent of the size of their GDP and distance between them is not embedded on

any theoretical literature of international trade. Equally, the model ignores the role of

technology, factor endowments and demand functions or structural differences that are known
to influence international trade. In addition some econometric issues still remain unresolved.

The use of total trade (imports and exports) as dependent variable and the other standard
variable with proxies to test the effects of membership of an RTA is in dispute. The use of total

16
trade as dependent variable imposes equality of coefficients between exports and imports. It is

suggested that these should be estimated separately.

Polak (1996) points out that using constant distance between trading partners is problematic,

because the absolute distance of some countries from their trading partners is much lower

than others. He gives the example of the EU (and their dependent on adjacent trading

partners) as more favorably located in these terms in contrast to the South-East Asia countries

(with higher dependence on the more distance European countries and other OECD markets),
in addition the use of dummy variables to capture additional effects of an RTA may lead to

incorrect inference, especially where there is multiplicity of membership to various trade

arrangements.

Meade (1955) outlined the static theory of regional integration arrangements by abandoning

the Vinerian assumption of constant costs of production in trading countries and recognized

the necessity of ensuring equilibrium in international payments balances. Thus, he brought to

the fore adjustment in both international and domestic relative prices to achieve general

equilibrium under regional integration arrangements. The framework highlights the central

role of prices and international terms of trade for achieving and maintaining equilibrium in

international trade and payments under preferential trading arrangements. The analysis

focused on the economic welfare of the world economy, not simply the countries forming a

regional integration arrangement. In going beyond the "small country" perspective, Meade

recognized the potential for significant secondary effects of regional integration arrangements

on third-countries and the world economy at large owing to adjustment in world markets for

traded and non-traded goods.

The use of CGE models has raised questions about the appropriate specification of behavioural

and technical relationships and the choice of parameter values. Baldwin and Venables (1995)

suggests that the uncertainties surrounding simulation results found by CGE models should
be tested through sensitivity exercises designed to probe the implications of key assumed
economic relationships, functional forms and parameter values.

17
Lipsey and Lancaster (1956/57) developed the theory of "second-best" which predicts that for

distorted economic systems, the elimination of one set of distortions does not guarantee an

improvement in overall economic welfare so long as other economic distortions remain

unchanged. The theory of second best therefore implies that reducing tariffs on a

discriminatory basis under a regional integration arrangement does not guarantee an

improvement in welfare for individual countries.

2.3 Empirical Literature

The early empirical studies of different regional integration arrangements like EU,

Nafta,Mercosur,Asean focused on explaining past trends in trade flows and related variables,

such as prices and national income. By their nature empirical studies involve ex post analysis

involving not only formally explaining past trends in trade flows and related variables, such as

prices and national income, but also specifying what course trade and other variables would

have taken had an extant regional integration arrangement not been established.

2.3.1 Studies on specific RTAs

The European Union being an early effort on integration has been the subject of many

quantitative ex post analysis particularly its impact of trade in manufactures, after the Rome

Treaty of 1957, using a variety of ex post approaches.

Studies by Truman (1969) and Prewo (1974) using trade share measures, Balassa (1967, 1975)
using income elasticities of demand for imports and the assumption that higher (lower)

income elasticity values imply trade creation (diversion), and Aitken (1973) using the gravity

model method to explain bilateral trade flows between the trading partners, sought to estimate

trade diversion as well as trade creation. Despite their different theoretical approaches, these

studies concluded that, following creation of the European Community (EC), trade creation in

manufactures significantly outweighed trade diversion in manufactures. They also found that
formation of the European Community led to significant expansion of manufacturing exports
to EC countries by non-member countries.

Balassa (1975) went beyond estimating the impacts of the European Community on trade, to

examine the impact of the new regional integration arrangement on EC welfare. Using the

18
average EC tariff rate for manufactures and his own estimate of trade creation for 1970, he

calculated that EC welfare was improved by about 0.15 percent of EC GDP per annum.

Additionally, he considered the economic cost of trade diversion under the EU Common

Agricultural Policy (CAP), which he calculated at $0.3-to-0.4 billion per annum. The analysis

arrived at a net welfare gain for the EC customs union of $0.4 billion per annum, or less than

one tenth of one percent of EC GDP per annum.

The studies by Gasiorek, Smith, and Venables (1992), and Haaland and Norman (1992) are
both based on an analytical model of imperfect competition in the European Community

developed by Smith and Venables (1988). These studies examined the implications of reducing

intra-union trade costs for manufactures by 2.5 percent, in principle enabling EU firms

producing differentiated products under imperfect competition to expand output and reduce

costs along declining average cost schedules. Harrison, Rutherford, and Tarr (1994) employed

a more extensive model that does not impose uniform pricing by firms across EU markets.

The quantitative results indicated that deepening of economic integration in the European

Union should be expected to achieve substantial economic gains, of about 1 percent of GDP

per annum in several EU countries, owing predominantly to pro-competitive effects of

product standardization. The results also suggested the occurrence of appreciable trade

diversion under the CAP of 1992, possibly limiting gains in welfare in the European Union and

contributing to losses in welfare in other parts of the world. The CGE simulation results
indicated that rationalization of production occurred within the European Union, with large

numbers of EU firms forced to shut down in the face of declining terms of trade and profit
margins.

Studies by Brown, Deardorff, and Stern (1992) and Roland-Horst, Reinert, and Shiells (1992) on

North Atlantic Free Trade area (NAFTA) which used CGE model incorporating imperfect

competition and increasing returns to scale. The Brown, Deardorff, and Stern model assumed
that firms set prices above average cost (though monopolistic profits are eventually bid away
by entry of new firms), while the Roland-Horst, Reinert, and Shiells model assumed the firms

set prices at average cost following the so-called contestable market theory (with no entry or

exit by firms). The third study by Bachrach and Mizrahi (1992) involved a simple specified

19
model, that assumes perfect competition and constant returns to scale in production. All the

three studies found out that NAFTA provides positive gains to member countries. However,

the variation in simulated economic gains was wide, with the smallest gains found by the

Bachrach and Mizrahi model (gains range from insignificant for Canada and the United States

to 0.32 percent of GDP per annum for Mexico) and the largest gains found by the Roland-

Horst, Reinert, and Shiells model (gains range from 2-to-3 percent of GDP per annum for the

United States and Mexico, respectively, to 10.57 percent of GDP per annum for Canada).

Studies on Asean Free trade Area (FTA) by DeRosa (1995), and Lewis and Robinson (1996),

found that AFTA is trade creating. The larger gains in trade and marginally larger gains in

welfare obtained by DeRosa are attributed in part to the differences in base periods and

simulation scenarios. Nonetheless, both studies conclude that Asean-FTA contributes

comparatively little to higher economic welfare in Asean countries, except possibly the two

highest-incomes and particularly open Asean countries, Malaysia and Singapore. This seems

counterintuitive in light of the static economic theory which indicates that relatively open

economies should be expected to experience limited if not negative gains from regional

integration arrangements. However, as DeRosa (1995), explains, both Malaysia and Singapore

benefit principally from the diversion of trade by other Asean countries. The two countries

supply the largest proportion of the increased intraregional demand for manufactures

previously supplied by advanced countries outside the region.

Yeats (1997) in an ex post study of the regional integration arrangement of Mecrosur-

Argentina, Brazil, Paraguay and Uruguay, using CGE modelling, to investigate whether recent

commodity patterns of exports by Mercosur countries to different destinations conform to the

past revealed comparative advantage of Mercosur countries in natural resource-intensive and

labor-intensive goods. He found that the fastest growing products in intra-Mercosur trade are

capital-intensive goods in which Mercosur countries have not previously displayed strong

export performance. Thus, the Yeats study suggests that the new patterns of trade are at odds

with what their historical comparative advantage would predict is indicative of possible
adverse effects of Mercosur on member countries.

20
Amjadi and Winters (1997) explored whether the regional integration arrangements offer

member countries opportunities for enjoying economic gains from avoiding trade-costs with

third countries. Specifically, they investigate whether transportation costs between Mercosur

countries and the rest of the world (represented by the United States) are sufficiently high to

afford significant gains to Mercosur countries from adopting preferential trade with one

another. The study concluded that extra-regional transportation costs are appreciably higher

than intra-regional transportation costs but that the margin between the two costs is not large

enough per se to yield substantial gains with the introduction of trade preferences among
Mercosur countries.

CGE models have become a widely used tool for evaluating the effects of trade policy reforms

in both regional and multinational initiatives (U.S. International Trade Commission, 1992;

Francois and Shiells, 1994; Martin and Winters, 1995). This is because regionalism currently

involves not only "small" blocs of countries but also "large" blocs of countries that count for an

appreciable share of world trade. While economic theory identifies how policy changes will

affect economic variables, it does not define the size of the impact and in the case of RTA's

leaves the effect ambiguous. The CGE models provide an empirical foundation for policy

analysis that can quantify the magnitudes of the effects identified by theory and suggest the

likely net effect, whether trade creating or trade diverting of an RTA. CGE models generally

include sectoral structure, factor markets, macro data and any innovative features of the
model, such as dynamic behavior and international labour, migration. The main advantage of

CGE models is that it takes into account linkages between markets, both product and factor

markets. The CGE framework accepts substitution of goods in demand and supply,

simultaneous adjustment of interrelated markets for goods and factors of production in

trading countries and is relevant in policy debates.

In contrast the Partial Equilibrium (PE) models focus on one sector at time without analysing

the interactions between markets. The partial equilibrium analysis of regional integration

arrangements appear incompatible with the modern neoclassical trade theory defined more

by general equilibrium theory and an emphasis on inter-relationships among markets for

goods and factors of production throughout an economy. But the main advantages of PE

21
models is that they are simple models, transparent (as rely on few key parameters) and more

useful when the policy effects are small in nature.

2.3.2 Studies on the East African region

Lucio Castro, et al (2005) in their study on regional trade integration in East Africa: trade and

revenue impacts of the planned east African community customs union, concluded that there

will be modest increase in regional trade flows as a result of the CU implementation. In the

first phase of CU implementation, there is almost no expansion of regional trade because of the
temporary tariffs on selected imports from Kenya. If the CU was implemented without the

temporary tariffs and a top rate of 20 percent, regional imports would increase by just less than

6 percent for Uganda, by about 2.4 percent for Tanzania, and about 1.4 percent for Kenya

relative to the pre-CU situation. Almost all of this increase in regional trade would be trade

diversion, which is more pervasive at the higher top tariff rate of 25 percent. He also noted that

in all three countries, nontariff barriers such as discriminatory surcharges, standards, and

import procedures hinder trade.

Okello (2006) in his study on the impact of East Africa Community customs union on Uganda

economy: a computable general equilibrium (CGE) analysis. He used a single-country multi­

sector CGE model to demonstrate the possible outcomes of East Africa customs union. The

study found that the customs union reduction of tariff rates has only had a small effect on the

macroeconomic variable in Uganda of less than 0.2 percent growth in real GDP, less than 0.2 of

absorption, 2.1 percent of net income tax and 2 and 3 percent for exports and imports

respectively. However, the removal of tariffs had mixed results on the industrial sector as

some industries gained and while others lost.

McIntyre (2005) examined the impact of the EAC-CU on Kenya, noted that the trade linkages

among the three EAC member states are not strong. However, the establishment of the EAC-

Customs union and the introduction of the EAC- CET had potential positive benefits for the

country. The results from the SMART trade simulation model6 suggested that the EAC-CET,
bv lowering tariffs, has a positive impact on trade largely from trade creation. The lower tariffs

result in lower import prices and increased flows of cheaper imports that improve consumer

22
welfare. However, the study warns that there are transitional costs that must be addressed to

minimize economic dislocation, including revenue losses.

N'g'ang'a (2005) examined the effect on trade, welfare and productive activities in East Africa

due to the new EAC. She found that the formation of the new EAC has not led to a large

increase in trade volumes among the EAC countries. There was no sudden break in the overall

trend, confirming that the EAC RTA has not had a major impact on the exports in the region.

The study also concluded that the pattern of trade in the EAC is being driven by the process of
development, rather than by trade pressures. Productive activities in the region show more of

a change following the formation of the EAC. The level of intra-industry trade was also

observed to increase by almost 175 per cent in the years following regional integration.

Estimates from the gravity model revealed that trade linkages between the EAC members are

quite dense.

Ngeno et al. (1999) in the study titled "Regional integration in East Africa, the case for Kenya",

noted that East Africa countries join one regional group or another with the desire to address

common problems in a collective and coordinated manner the objective is for enhanced

growth and development through joint efforts. This is consistent with the overall objective of

the customs union as a means of promoting increased intra-regional trade and economies of

scale through pooling of fragmented and small markets to support industrialization. The latest

EAC development strategy highlights this goal, even if actualization of the strategic objectives

may be wanting.

Maria Nassali (2005) indicated in her study titled "The East Africa community and the struggle

for constitutionalism; challenges and prospects" that countries of the East Africa have enjoyed

close historical, commercial, and cultural parameters overtime. They share similarities in

educational background and common law jurisdiction. Therefore out of commonness in these

areas it is viable to collaborate than act individually for economic development achieve the

aspirations of their citizens. But political differences between member countries can impede
the successful implementation of an RTA.

23
Elbadawi and Mwega (1997) in their paper titled “Regional integration, trade, FDI in Sub-

Saharan Africa", indicates that the geographical coverage of the regional integration schemes

was of great importance to the EAC countries industrialization. As such these schemes should

be made of large number of countries to allow development of sufficiently large internal

markets to support their industrialization. This line of thought may have informed the

inclusion of Rwanda and Burundi into the EAC even though they were not members at the

time the EAC treaty was ratified.

Ademole Ovejide (1997) titled "Regional Integration and trade liberalization in sub-Saharan

Africa" shows that these regional integration efforts have generally achieved little success.

However, little achievement could be attributed to failure to reduce trade barriers due to

inconsistent regional agreements with national aspirations and policies, such as loss of tariff

revenues and land resources.

2.4 Literature Overview

The reviewed literature on various regional integration arrangements concludes in general

that regional trading arrangements are trade-creating and welfare-improving for member

countries and trading blocs as a whole. These studies incorporated the use of both general

equilibrium and partial equilibrium methods, the results are near universal that regional trade

arrangements are beneficial to member countries.

Studies such as Aitken (1973) using the gravity model showed that the formation of the EU

was trade creating for the manufacturing sector in member countries, while studies by Reinert,

and Shiells (1992) using a CGE model concluded that NAFTA had positive gains to member

countries. Yeats (1997) in his study of Mercosur using CGE modelling, to investigate whether

recent commodity patterns of exports by Mercosur countries to different destinations conform

to the past revealed comparative advantage, found that the fastest growing products in intra-
Mercosur trade are capital-intensive goods, where the countries did not have previous

comparative advantage. Okello (2006) using the CGE to examine the impact of the EAC-CU

concluded that the EAC-CU will have a small impact on macro-economic variables, with less

24
that 3 per cent growth in imports and exports, while Lucio Castro, et al (2005) found that the

increased regional trade within the EAC would actually be trade diversion. The evidence from

these studies suggests that the static gains of regional integration arrangements especially for

developing countries RTAs are modest at best (less that 0.3 percent of GDP per annum).

The literature reviewed revealed many factors that influence the effect of RTAs on the member

countries, some of the identified factors include, income of RTA member countries as

measured by their GDP, membership to particular RTAs, applied tariff rates within the RTA,
transport costs measured through distance, income elasticity of demand, revealed comparative

advantage and closer ties (boundary, culture, history etc) among the trading countries.

However, from these literature there is no uniform approach on the methodology. The

researchers have employed different methods depending on the study objectives and resource

requirements. The study by Ng'ang'a (2005) for example, employed the gravity model, which

makes it possible to challenge the outcome given the inherent weakness of the gravity model,

while McIntyre (2005) used the SMART model which is static in nature hence does no take into

account the dynamism of international trade, the model also employs borrowed elasticities,

which heavily influences the final results. These two studies employed different methods, to

arrive at divergent results on trade linkages within the EAC member countries, Ngang'a (2005)

using trade intensity index finds that there are strong linkages within the member states, while

McIntyre (2005) applying the complementarity index concludes that the linkages are weak.

This study contributes to existing literature by using a partial equilibrium approach, applying

the conventional export model that incorporates the dynamic effect of real exchange rate as a

proxy for relative prices and income per capita of the trading partners, (Ogun 1998, Klaassen

1999, Alemayehu 1999), tarmacked road network, investment as a proportion of the GDP and

the existence of the EAC-RTA as explanatory variables to estimate Kenya's exports to the

region from 1977 to 2008. It is expected that the results from the study will add to the empirical

literature on RTAs, specifically on the East African region, after the formation of the "new"
EAC.

25
 CHAPTER THREE
3.0 METHODOLOGY
3.1 Theoretical Framework
The theory of international trade that as developed from the seminal writings of Heckscher

and Ohlin is fundamentally based on the twin observations that countries differ from each

other in the composition of their factor endowments and that productive activities are

distinguished by the different relative intensities of factors required. The Heckscher-Ohlin

Theorem states that countries export those commodities which require, for their production,

relatively intensive use of those productive factors found locally in relative abundance. The

twin concepts of relative factor intensity and relative factor abundance are most easily defined

in the small dimensional context in which the basic theory is usually developed. Two countries

engaged in free trade with each producing the same pair of commodities in a purely

competitive setting, supported by constant returns to scale technology that is shared by both

countries.

Following the partial equilibrium model outlined in Goldstein and Kahn (1985) and discussed

further in Edwards and Wilcox (2003)8 the specification of the export model differs between

studies, particularly with respect to the variables included. But the core of the underlying

(long-run) framework is usually a system of equations for export supply (Xs) and export

demand (X^), for country's exports in a given commodity or commodities, which

simultaneously determine the export price and the export quantity.

The export demand is positively affected by foreign income (Y*) and the price of competing

foreign goods (P*), but is negatively affected by the foreign price of domestic exports

(Px*=Px/e). The quantity of exports supplied is specified as a positive function of its own price

and a negative function of the domestic price index and variable costs. As export sales become

profitably relative to domestic sales (Px/P rises) firms shift production towards the export

market. Other supply side variables include tariff rates, infrastructure costs and capacity

26
 utilization. It is assumed that in a small country case like member countries of the East African

region, the supply of exports from one country to the other does not influence the prices.

3.2 Analytical Framework

Although literature on commodity export starts from structural equations, which

accommodate a number of factors, the estimated reduced form equations are generally price-

focused; they include either current or lagged (relative) prices. Therefore conventional

commodity export models usually incorporates the real foreign income of trading partners and
real exchange rate as a proxy for relative prices as explanatory variables in estimation of

! export functions in general (Ogun 1998, Klaassen 2004, Ndung'u and Ngugi 1999).The general

reduced form of export equation i s ;

MJr;) = A0 + * , £ - ) + * , In GDP,.......................................................................................................... (1)

Where:

A” is the export supply value from country i to j;

P is the export price in domestic currency of country i;

P‘ means the consumer price index in country i;

GDP, represents GDP of country i;

This study adopted a similar approach, but introduced investment as portion of the GDP to

take into account capital formation, as a proxy for structural supply issues that may affect the

supply of exports (Gotur, 1985; IMF, 1984) and the GDP per capita of the destination countries

(Uganda and Tanzania) to capture the purchasing ability of the residents, while tarmacked

roads surface in kilometers as a measure of infrastructure conditions. Further, to examine the

effect of the EAC-RTA the study introduced the RTA proxy where a dummy value of (1)

represents existence of the East African Community and (0) when EAC was not in existence

(i.e. 1978-1993).

27
I The study assumed;
i. Small country case: Where the new international trade is not big enough to influence

international prices.

ii Perfect competition: Domestic prices automatically move to converge to international

prices.

3.3 Model specification

An empirical model along the standard export trade model is thus specified as;

* ; = /( RER ’ ' GDP GDP Z-RTA ’ '-Trdnet ) ......................(2)

Where:

£ the export supply value from country i to j

RER, real exchange rate of exporting country i

GDEapj income per capita of the importing country;

GDR L,
Investments as a proportion of the GDP of exporting country i

T f'd n et = Tarmacked road network of exporting country i

JA , = the proxy for membership of i and j to the E AC-RTA

i = is the exporting country (Kenya)

= export destination country (Uganda and Tanzania)

The above equation (2) will be transformed into a logarithmic form for estimation purposes;

InX, =/?„+/?, In RER, + 0 2 In GDPcap + /?, In GDP,nvl + fi<RTAy + fi5 In Trdnet + e y .................... (3)

3.4 HYPOTHESIS OF THE SUDY

The study hypothesized that;

i. There is a positive relationship between the real exchange rate and the value of exports.

ii. There is a negative relationship between income per capita and the value of exports.

iii. There is a negative relationship between tarmacked road network and value of exports.

28
 CHAPTER FOUR

4.0 Data, Sources and Types

For estimation purpose, the study used annual secondary data covering the period from 1977-

2008. The year 1977 was chosen to captures the period when the original EAC collapsed until

the formation of the new EAC and the Kenya embarked on phased rationalization of the tariff

bands due to external shocks, trade liberalization induced by the structural Adjustment

Programmes (SAP) and an outward oriented growth strategy. The main sources of data on
Real Exchange Rate, GDP, investment and the export values were the economic surveys by the

Kenva National Bureau of Statistics, Kenya Revenue Authority, data from the East Africa

Community secretariat, available in the official website and other relevant sources such as the

World Bank, COMESA, UNCTAD and WTO. The selection of the countries (Uganda and

Tanzania) was due to the fact that they are the main export destination of Kenya's export

products within the EAC. Due to data considerations and the late entry of Rwanda and

Burundi into the regional trade arrangement this study did not include them.

The macro-economic variables are non-stationary overtime, failure to distinguish between

stationarity and non-stationary may lead to spurious regression problem. The study applied
the Ordinary Least Squares (OLS) when estimating the log-linear model specification referred

in 3.2. Augmented Dickey-Fuller (ADF) tests was used to test for stationary of the data.The

model was estimated in natural logarithms to make it less sensitive to extreme observations

when applying OLS estimation and estimated parameters interpreted as elasticises. The RTA

dummy variable was not transformed into a logarithmic form, but to show its effect on exports

a graphical trend analysis was performed showing the years when the RTA was in existence

and when the EAC was not in existence.

It has been acknowledged that the application of simple OLS using time series data may

produce spurious, even biased regression results (Cheremza and Deadman, 1992 and

Alemayehu, 1999). Modern time series modelling techniques provide better ways of

addressing these concerns. In order to avoid spurious regression, stationarity and

cointergartion analysis of the series is conducted, this validates the usual test statistics (t- and

29
1 F-statistics, and R2). In this study stationarity was achieved by appropriate differencing using

A ugm ented Dickey Fuller (ADF) (Dickey and Fuller 1979).

Assume the equation;

m
\y=P i+ P 2+ a , T Jy +s
i-i

I where y = is the variable of interest = {RERi,GDPCap),GDPinvi,RTAij,Trdneti,} and £ is the white

noise residual of zero mean and constant variance

I 1(31,(32,01— a m} is a set of parameters to be estimated,

i Both of the null and alternative hypotheses in unit root tests are:
HO: 6 = 0 (y is non-stationary/a unit root process)

HI: 6 ? 0 ( y is stationary)

The unit root hypothesis of the Dickey-Fuller can be rejected if the t-test statistic from these

j tests is negatively less than the critical value tabulated. In other words, by the Augmented

I Dickey Fuller (ADF) test, a unit root exists in the series y (implies nonstationary) if the null

|hypothesis of 6 equals zero is not rejected (Gujarati 1995).

30
 CHAPTER FIVE
5.0 Data Analysis and Results
5.1 Descriptive Statistics
This chapter presents the descriptive and empirical analysis of variables estimated in the

model. The descriptive statistics gives the mean, the standard deviations and graphical

analysis of the observed variables while the empirical analysis gives the regression results of

the estimated model.

5.1.1 Standard Deviations and Means

In this section a summary of the main variables that have been used in estimation of the model

are shown in the table 5.1 below.

Table 5:1: Descriptive Statistics

U ganda T a n z a n ia K enya
Var O bsv M e d ia n M ean Std M ed ian M ean S td M e d ia n M ean S td
d e v ia tio n d e v ia tio n d e v ia tio n
Log 32 4 .5 1 5 .3 3 1 .4 0 3 .3 1 4 .3 2 .1 7
Exports
Log 32 281 300 5 .6 6 .3 1
GDP-per 5 .6 2 .3 9 2
capita
Log Real 32 61 4 .1 0 .1 9
exchange
RateSK
Log 32 19 2 .0 .1 3
Investee
ntto gdp
Kenya
Log 32 4 .0 8 .2 3 .1 6
tarmacke
d road-
network
in Kenya
Source: Authors computation

From the descriptive statistics all the variables have all the observations indicating that there

no missing observations in the primary variables. The mean and median of exports to Uganda

is 15.33 and 4.5 respectively with a standard deviation of 1.40 while the exports to Tanzania

had a mean of 14.3 and median of 3.3 with standard deviation of 2.17 over the years. This

indicates that Kenya exports a near equal share to the regional countries with Uganda slightly

a head of Tanzania as an export destination of Kenyan goods, the low standard deviation of in

31
1both cases shows the robustness of the data and points to low fluctuation of Kenya's exports to

the countries.
The GDP per capita of the two countries Uganda and Tanzania have a mean of 5.66 and

median 281 for Uganda and 300 for Tanzania with low standard deviation of 0.39. When the

GDP per capita is taken as a reflection of the level of development it shows that there is little

vanation between the two economies though Tanzania is slightly ahead. It can be concluded

that there is potential for expanding exports by Kenya into either Uganda or Tanzania due to

near similar levels of demand based on income per capita.

The tarmacked roads in kilometers which is a proxy for the infrastructure in Kenya's has the

highest mean of 8.23 and a standard deviation of 0.15. The study attempted to use tarmacked

road per capita, however this resulted into extreme low figures as the population growth rate

was consistently much higher than the rate of tarmacked road surface. The high mean shows

that infrastructure costs are generally high for exporters in Kenya with the low standard

deviation indicating that infrastructure costs has remained constantly high over the years.

5.2 Stationarity analysis

Recent studies have drawn attention to the fact that most time series data might be non-

stationary which is likely to result in 'spurious regressions' and the concomitant incorrect

statistical inferences. Though first differencing can be used to overcome this problem,

potentially useful information about long-run equilibrium relationships between economic

variables might be lost. The level information may be of significance particularly when a group

of variables are cointegrated. The Augmented Dickey Fuller (ADF) test was utilized to test for

the presence of unit roots. The test was performed in levels and in first difference including

both a constant and a deterministic trend. The results are given in table 5.2.

32
Table 5:2: Unit root tests

Variable Level First difference 1(d)

Log Exports to Uganda -1.44 -5.38 1(2)

Log gdp par capita- -2.07 -3.69 1(1)

Uganda
Log exports to -0.82 -4.53 1(2)
Tanzania
Log gdp per capita - -0.66 -2.98 1(1)
Tanzania
Log real exchange Rate -2.46 -3.70 1(1)
$K
Log investment to -2.46 -3.25 1(1)
GDP-Kenya
Log tarmacked road -4.32 -2.80 1(1)
network-Kenya
Critical Values at 5% -2.96 -2.97
1% -3.66 -3.67
Source: Authors computation

Critical Values for the test are -3.66,-2.96 at 1% and 5 % respectively. 1(d) refer to the order of

integration. From the results the variables are non-stationary in the first difference except for

exports to Tanzania and Uganda that were integrated of order two.

33
Table 5.3 Stationary Test for residual

ADF Test Statistic -4.288299 1% Critical Value* -3.6959

5% Critical Value -2.9750
10% Critical Value -2.6265
•MacKinnon critical values for rejection of hypothesis of a unit root.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(RESIDl)
Method: Least Squares
Sample(adjusted): 1982 2008
Included observations: 27 after adjusting endpoints
Variable Coefficient Std. Error t-Statistic Prob.
RESIDl(-l) -1.105346 0.257759 -4.288299 0.0003
D(RESID1(-1)) 0.026371 0.180054 0.146459 0.8848
C -11.02458 16.00673 -0.688747 0.4976
R-squared 0.595781 Mean dependent var -5.314684
Adjusted R-squared 0.562096 S.D. dependent var 125.3201
S.E. of regression 82.92967 Akaike info criterion 11.77830
Sum squared resid 165055.9 Schwarz criterion 11.92228
Log likelihood -156.0071 F-statistic 17.68689
Durbin-Watson stat 1.462894 Prob(F-statistic) 0.000019

Source: Authors computation

From the results the ADF statistic is less than the critical values at 1%, 5% and 10% level of

significant. The residuals are therefore stationary at 1%, 5% and 10%. If the residuals are non-

stationary they cannot become the Error Correction Term (ECT) and consequently an error

correction model is not adopted. The residuals were found to be stationary at 1%, 5% and 10%

levels of significance by the ADF test. The Durbin Watson statistic of 1.4 shows that there is no

serious serial correlation between the dependent variable and the residual of the estimated

equations. Therefore the residual becomes the Error Correction Term (ECT). The residuals is

therefore taken to be independent and identically distributed as N (0, 5).

5.2.1 Correlation of variables

Modelling using variables in the first difference to achieve stationary leads to loss of long-run

information. The concept of correlation implies that if there is long-rung relationship between
two or more non- stationary variables, deviation from this long run-path are stationary.

34
Under specific assumptions as to the properties of the random error term 8, the OLS estimator

has some useful properties. Specifically, the OLS is consistent and unbiased as an estimator of

|3 if the following conditions hold: None of the dependent variables are perfectly correlated

imulticollinearity); 8 is an independently distributed normal error with mean zero and with

constant variance. 8 is uncorrelated with any of the independent variables. The underlying

model relating the dependent and independent variables is linear. If these conditions hold, the

estimates are reliable and that hypothesis tests are informative.

Pearson correlation multivariate procedure was used to establish whether the variables are
correlated in the long-run. According Gujarati (2004), rnulticolineality becomes a serious
problem if the pairwise or zero-order correlation coefficient between two regressors is in

excess of 0.8. The results given below in tables 5.4 and table 5.5 were run to test both the

existence of correlation between the variables at 5% level of significance for the exports to

Uganda and Tanzania respectively. The results showed that most of the variables are not

correlated, except tarmacked road network in kilometers in Kenya and exports to Uganda.

However, the correlation is not strong to result into serious endogenity problem and unbiased

estimates, hence no need to perform a two stage least square, which may not produce any

better results than a simple OLS.

W''/p> ? f r y *r- ,,,

"'*n i 't'1' •*»„
east 1CAA-a

35
Table 5:4: Correlation of variables of exports to Uganda

Log Log real Log

exports to Log GDP per exchange Log investment to tarmcked
Variable Uganda capita -Uganda RateSK GDP in Kenya road Kenya
Pearson Log exports to Uganda
Correlati 1 .0 0 0 -.1 2 3 .4 9 9 - .3 7 4 -.8 7 5
on
Log GDP per capita-
-.123 1.000 -.424 .440 .153
Uganda
Log Real exchange -.319
.499 -.424 1.000 -.645
RateSK
Log investment to GDP
-.374 .440 -.645 1.000 .258
in Kenya
Log tarmacked road
-.875 .153 -.319 .258 1.000
network -Kenya
Source: Authors computation

Table 5:5: C orrelation o f variables of exp orts to Tanzania

Log of Log Log of GDP

exports to Log Real Investment to Log tarmacked percapita in
Tanzania Exchange GDP in road in Km in Tanzania
Variable rateSK Kenya Kenya
Pearson Log of value of .029
1 .0 0 0 .720 -.598 -.720
Correlation exports to Tanzania
Log real exchange -.070
.720 1 .0 0 0 -.645 -.319
rateSK
Log Investment to
-.598 -.645 1 .0 0 0 .258 -.039
GDP in Kenya
Log tarmacked road
00
VI
hU

-.720 -.319 .258 1 .0 0 0

Kenya
Log of GDP percapita
.029 -.070 -.039 -.187 1 .0 0 0
in Tanzania
Source: Authors computation

36
5.3 Empirical results
The empirical results from equations developed in section three are presented here. The study

used three different models to establish the relationships between the variables in a general

model that combined both export values and used the country as a variable to asses whether

country of export destinations was significant. The second set of models used individual
country (Uganda and Tanzania) to run the regressions. The Ordinary Least Squares (OLS) was

used in the regressions, with 32 observations representing export values since 1977 to 2008.

The estimation results are shown in tables 5.6, 5.7 and 5.8, and figures 5.1, 5.2 at the end of the
discussion. The broad conclusion from the models is that the existence of the regional trade

arrangement and tarmacked road network in Kenya are significant determinants of export

values.

In the general model I, the country of destination was found to be significant and negatively

related to exports, the results indicated that as exports value increase for Uganda there is a

marginal drop in exports to Tanzania. This result is consistent with the observed trend where

Uganda is the leading export destination of goods from Kenya within the EAC. This can be

attributed to historical factors that followed the collapse of the then EAC in 1977, Tanzania

closed its border with Kenya resulting into lag in exports to the country.

Analysis concerning the impact of infrastructure on economic expansion tends to investigate

the linkages between infrastructure and economic growth as framed by the theoretical

underpinnings of growth theory (Straub 2007). The empirical analysis usually establishes a

positive and significant relationship between infrastructure and output growth or productivity

(Straub 2008). The channel from infrastructure to growth involves the productivity enhancing

effects of improvements in infrastructure through expansion of the capacity in a specific factor

of production. Further, increasing infrastructure stock indirectly raises the productivity of

other factors of production. Therefore improved infrastructure has the potential to reduce the

costs of doing business and encourage investment in productive assets (Manuel 2006). The

results of this study showed that a 1% increase in the tarmacked road surface leads to a 4.3%

increase in exports confirming previous observations. Noting that tarmacked road network

"as used as a proxy for infrastructure conditions, it can be concluded that an improvement of

37
infrastructure in Kenya is desirable for increased exports. However, Kenya's export growth is

being constrained mainly by inadequate transport infrastructure. This has resulted in

considerable wear and tear, unnecessary road checks, delays in port clearing, high accident

rates and inflated cost of transportation.

Kenya has been losing its competitiveness in attracting investment and retaining the stock of

investment over the last decade. According to Kenya's Vision 2030, Kenya's performance in

attracting FDI has been marginally better at nearly US$6 per US$1,000 of GDP (US $82 million

in total) since 2003. This is partly explained by factors such as negative perception by investors
about corruption, inadequate infrastructure, political instability, among other reasons. The net

stock of foreign direct investment was 15% during the first half of the 1990s decade but

plunged to less than 6% compared with Uganda's meteoric rise from 8% to 50% (Mullei 2003).

This translates to an annual average of US$ 59 million in FDI or 25.7% of what was received by

Uganda or 18.7% of the FDI that went to Tanzania between 1997 and 2002. The results showed

that investment as a proportion of the GDP, a proxy for export supply constraints in Kenya is

negatively related to exports in both regressions; the study found that a 1% decrease in

mvestment as proportion of the GDP results into 0.15% increase in exports value. This

confirms that a decrease in the supply constraints would result into an increase in exports

value, implying therefore that Kenya should improve its image to attract more investments by
reducing the current investment constraints like high power costs, licensing procedures,

corruption, political stability among other investment issues in order to increase its export

share in the EAC regional markets.

The exchange rate is an instrument used to explain the international competitiveness of the
country' exports. This variable is negatively related to export growth. An appreciation of the

exchange rate helps promote price stability but reduces the country's international

competiveness and vice versa (Ndung'u and Ngugi 1999). A depreciated real exchange rate

'educes relative prices of domestic goods and services enhancing production of tradable goods

and discouraging the production of non-tradable goods. On the other hand an appreciated

exchange rate increases the relative prices of domestic goods and discourages the production

38
 t the tradable goods, while encouraging production of non-tradable goods. I n economic

Itheorv the depreciation a country's exchange rate leads to an increase in the exports. This
studv showed that a 1% depreciation in real exchange rate in Kenya results into 0.95% increase

in exports, though not a significant determinant of exports from Kenya into the trading

partners. This is consistence with past Kenya's policy choices where real exchange rate has

been used to promote price stability and not as an instrument to promote Kenya's trade

competitiveness.

Past studies have tried to investigate trade openness and its effect on the per-capita income

levels of a country. By examining per capita income on total trade, the studies embody an

underlying assumption that exports and imports contribute equally to income growth.

Imports, however affect growth negatively, such that increasing the import trade share by 1

percentage point is associated with a reduction of the per capita GDP by approximately 0.1

percent (Zhang et al 2003). The combined general model I of this study showed that a 1%

increase in GDP per capita results into 0.049% increase into exports. While the empirical

results on individual country models for both Uganda and Tanzania showed that income per

capita of the trading partners is negatively related to exports from Kenya, where a 1% decrease
in GDP per capita of the trading partners resulted into a 0.05% increase in exports, confirming

the linkage between imports and income per capita.

The explanation to this negative relationship is that countries which import primarily

consumption goods grow slower than countries with a large proportion of their imports

composed of intermediate goods. The general argument is that importing from established

foreign firms may discourage the development of domestic infant industries, leading to less

technological improvement (Keller 2000). Therefore, reducing the imports of infant industry

products may encourage the internalization of costs and help increase the factor productivity,

hence countries' overall per capita income level.

Economic arrangements that represent varying degrees of integration tend to reduce the trade

harriers between them resulting in easy flow of factors of production. The reasons advanced
for the regional trade arrangements include proximity of countries, economic size of the

39
countries, per capita income and the international terms of trade among other factors. The

imam objective for formation of East Africa Community was to develop policies and
program m es aimed at widening and deepening co-operation among partner states in political,

economic, social and cultural fields, research and technology, defense, security and legal and

[judicial affairs for their mutual benefit (Treaty for the Establishment of the East African

Com m unity 1999).The empirical results reinforced this objective by finding that RTA as a

dummy for the existence of the EAC was significant in determining Kenya's export to the

region. To analyse the effect of the RTA dummy, the study conducted a trend analysis,which
showed that Kenya's exports to the trading partners decreased following the years after the
disintegration of the initial EAC in 1977.

I Therefore does the study track the effect of the EAC-RTA on Kenya's export performance?

I This is the question that the study was set to answer. Figure 5.land 5.2 shows how Kenya's

exports performance during the periods when the EAC was in existence and when the initial

EAC disintegrated. The graphical analysis shows the difference in the two periods following
the disintegration and revival of the EAC-RTA and its effect on Kenya's export trends.

The export trends for both Uganda and Tanzania showed that between the periods of 1977 to

1992, when there was no EAC-RTA there was near stagnant growth in the exports value. But

following the revival of the current EAC and signing of a declaration on the need to corporate

in all areas by the three countries (Kenya Uganda and Tanzania) in 1993, export trade

improved as shown by the increasing trend in export values. The designing of the EAC treaty

customs union protocol using the Principle of Asymmetry may explain the slacked export

growth between 1994 and 2005. The customs union was the entry point into the Regional

Integration Arrangement (RTA).The gradual elimination of the various tariff lines that were

initially maintained by both Uganda and Tanzania on goods originating from Kenya to protect

their domestic industries under the Principle of asymmetry, resulted in the increase in the

value of Kenya's exports after 2006.

40
IGenerally, the high adjusted R2 of 0.89 and F-statistic for the goodness of fit of 93.87 in the

general model, with the specific country model having 0.89 as the adjusted R2 and F-statistics
lo t 57.51. This shows that the independent variables in the model can jointly explain Kenya's

export performance for the period of 1977-2008, indicating that other factors outside the model

I explain about 20% of Kenya's exports to the trading partners. The negative intercept of -17.1,

both in the general and specific country models shows that all other variables being constant,

Kenva will have a high negative trade balance with Uganda and Tanzania, with Kenya

importing more from these countries than it exports.

All the significant variables are similar in the three models, expect income per capita of the
trading partners being positively related with exports from Kenya in the general model.

Subsequently, the general model I was picked as the best model in this study given the high

values of the adjusted R and F-statistic.

■»
v
U
M

*»
w
41
lr

41
/
5.3.1 General Model I.

llogEXP = logGDPCAP + logRER$K + loglNVGDP + logTRNKM + RTA1 + countryl

I The Regression results are shown in table 5.5 below.

Table 5:6: General m od el I regression resu lts

^ efficien ts:
E s t i m a t e S t d . E r r o r t v a l u e P r ( > 111)
Intercept) - 1 7 . 1 0 0 8 5 7.23388 - 2 . 3 6 4 0.0215 ★
log gdpeap 0.04911 0.23313 0.211 0.8339
log ReRSK -0.85943 0.70322 - 1 . 2 2 2 0.2267
Log InvgdpK - 0 . 1 9 2 0 4 0.76758 - 0 .2 5 0 0.8033
Log Trnkin 4.24759 0.92955 4 . 5 6 9 2 . 6 7 e- 0 5 ★ **
RTA11 2.56209 0.20466 1 2 .5 1 9 < 2e-16 *★ *
count ryl, 2 -1.03357 0.15060 - 6 . 8 6 3 5 . 38e-09 ★ *★

Significant codes: 0 •***’ 0.001 0.01 •*’ 0.05 V 0.1 ‘ ’ 1

Residual standard error: 0.6012 on 57 degrees of freedom

Multiple R-squared: 0.9081, Adjusted R-squared: 0.8984
F-statistic: 93.87 on 6 and 57 DF, p-value: < 2.2e-16

Source: Authors computation

Model I

EXPSij=-l7.1 +0.049GDPCAPj-0.86RER$K-0.19 INVGDPi+4.24TRNKMi+2.56RTAij -1.033 country+£

t-value (-2.364) (0.211) (-1.222) (-0.250) (4.569) (12.519) (-6.863)
Adjusted R2: 0.8984

42
5.32 UGANDA

ogEXPUg = IogGDPCAPUg + IogRERSK + logINVGDP + logTRNKM + RTA2+E

Table 5:7: M odel II regression results

leef f t c i e n t s :
E s t i m a t e S t d . E r r o r t v a l u e P r (>11 1 )
I nt er cept ) -- 1 7 . 6 2 3 6 8 9.69084 - 1 . 8 1 9 0.074138
Leg gdpcapUg - 0 . 0 5 2 2 8 0.31170 - 0 . 1 6 8 0.867377
Log ReR$K -0.94750 0.94197 - 1 . 0 0 6 0.318654
Log Invgdpk - 0 . 1 5 2 3 4 1.02831 - 0 . 1 4 8 0.882742
Log Trnkin 4.34789 1.24519 3 . 4 92 0 . 0 0 0 9 2 5 * * *
?"A21 2.55529 0.27418 9 . 3 2 0 3 . 95e-13 ★ * *

Significant codes: 0 '***' 0.001 '** ' 0.01 0.05 7 0 .1 " 1

Residual standard error: 0.8054 on 58 degrees of freedom

Multiple R-squared: 0.8322, Adjusted R-squared: 0.8177
F-statistic: 57.51 on 5 and 58 DF, p-value: < 2.2e-16

Model II
EXPUg = -17.6-0.05 GDPCAPUg-0.9 RER$K-0.2 INVGDP + 4.3 TRNKM +2.6RTA2

Running best fit Model resulted into

Coefficients:
Intercept) log tarmacked in KM log Regional Trade Arrangement
-15.305 3.501 2.588

EXPUg=-15.3+3.5TRNKM + 2.6 RTA+£

t-value (-1.810) (3.492) (9.320) (0.8054)

Adjusted R2=81

Source: Authors computation

5.3.3 TANZANIA
Log EXPTZ = log GDPCAPTz + logRHRSK + logINVGDP + logTRNKM + RTA3

Table 5:8: Model III regression results

Tcefficients:
Estimate Std. Error t value P r O l t l )
Intercept) -17.62368 9 . 6 9 0 84 - 1 . 8 1 9 0 . 0 7 4 1 3 8
log gdpTz -0.05228 0.31170 - 0 . 1 6 8 0.867377
Log ReR$K -0.94750 0 . 9 4 1 9 7 - 1 . 0 0 6 0 . 3 1 8 6 54
Log Invgdp -0.15234 1 . 0 2 8 31 - 0 . 1 4 8 0 . 8 8 2 7 4 2
Log Trnkm 4.34789 1.24519 3 . 4 92 0 . 0 0 0 9 2 5
RTA31 2.55529 0.27418 9 . 3 2 0 3 . 95e-13

Ngmf. codes: 0 ***** 0.001 ***’ 0.01 **’ 0.05 0.1 ‘ ’ 1

Residual standard error: 0.8054 on 58 degrees of freedom

Multiple R-squared: 0.8322, Adjusted R-squared: 0.8177
F-statistic: 57.51 on 5 and 58 DF, p-value: < 2.2e-16

Model III

EXPTZ = -17.63-0.05GDPCAPTz -0.95RER$K-0.151NVGDP + 4.35TRNCAPK + 2.55RTA3

Running the best fit m od el resulted into;

Coefficients:

llntercept) log tarmacked Road in KM log Regional Trade Arrangement

-15.305 3.501 2.588

EXPTz= -15.301+3.50TRNM + 2.59 RTA+£

Se (9.69084) (3.492) (9.320) (0.8054)

Adjusted R2=81

Source: Authors computation.

44
5.3.4 Export trends
I Graphical analysis was used to show the effect of the existence or non-existence of the RTA

have on Kenya's exports in absolute values, to Uganda and Tanzania in the periods under
studv between 1977 to 1992 when the EAC was not existing, and 1993 to 2008 when the EAC

was revived.

VALUE OF EXPORTS FROM K B W A TO UGANDA FROM 1977-2008

45.000,000

YEAR

Figure 5:1: Value of Kenya's exports to Uganda

Source: Authors computation.

45
 VALUEOFEXPORTSFROM KBIYA TOTANZANIA FROM 1977-2008

35,000000
VALUE OF EXPORTS TO TANZANIA

YEAR

Figure 5:2: Value of Kenya's Exports to Tanzania

Source: Authors computation.

46
 C H A P T E R S IX

l 6 .0 Summary, Conclusion and Policy implications

6.1 Summary
Recent academic literature has contributed immensely to the debate on the importance of
RTAs and their contribution to regional trade. This study has explored a broad range of
theories and empirical literature on RTAs. These theories focused on the relationship between

a country's membership to an RTA and a set of economic variables. The theoretical

development of RTAs has evolved from the classical to the neo-classical trade theories and
provided a foundation for empirical analysis of the EAC-RTA.

On empirical grounds, a large literature supports the view that there exist a functional

relationship between RTAs and expansion of trade. The extensive theoretical and empirical
iterature discussed showed that modeling and estimating exports largely depend on the

objective of the study. The observation drawn from the studies reviewed was that the price
effect variables are included in explaining a country's exports in many of such kind of studies
and Kenya was found not to be an exception.

The objective of the study was to investigate the effect of the EAC-RTA and the factors that

determine Kenya's exports to Uganda and Tanzania. It conducted an analysis of the various

ractors which helped in identifying the following variables as having possible influence on the

country's exports to the regional trading partners'; GDP per capita of trading partners,

investment as a proportion of the GDP in Kenya as a proxy for supply constraints, Real

exchange Rate of Kenya shilling and the US dollar, tarmacked road network in Kenya, and the

existence of the EAC-RTA to reflect the regional trade arrangement following the collapse of
T«e initial EAC in 1977.

47
6.2 Conclusion

The results indicated that the variables included in the equation had the expected signs. The

-tudy concluded that integration in East African region and tarmacked road network as a

rroxy for infrastructure are significant determiners of Kenya's exports to Uganda and

Tanzania.

From both models, the study infers that real exchange rate is not a significant factor in

\plaining exports though negatively related to exports. In this paper inconsistency was noted
in the general model I where GDP per capita had a positive sign but turned negative on

individual country models. Investment as proportion of the GDP used as proxy for supply

constraints, had a negative coefficient, hence the study concluded that reducing supply

constraints would result into increased exports, though not a significant determinant of trade

between the trading partners. However, it is important to keep in mind that investments in the

country as proportion of the GDP have been declining overtime. It is also crucial to note that

even though Kenya exports to the two countries have increased, it has also expanded to other

countries in the wider COMESA region.

Over the years investors in Kenya have complained of the high costs in conducting business in

the country. The results vindicate this position as tarmacked road network as a proxy for

nfrastructure conditions, was found to be positively significant in explaining exports.

Therefore the study concluded that by improving the infrastructure in the country exports

Tade can be increased.

<ith increasing levels of multilateral trade liberalization, export-led growth strategy is

rtevitable. Even though Kenya has reduced its dependence on traditional major export

destinations by pursuing a regional approach in her trade policy. The empirical estimations of

3is study concluded that there exists a relationship between the East Africa Community and

dheperformance of exports. The positive coefficient of the RTA dummy shows the importance

df regional trade arrangements in enhancing Kenya's export growth strategy. That

■®twithstanding, the share of commodity exports to the region is still relatively low, therefore
^country stands to gain by pursuing export opportunities into the regional markets through

48
 promotion of valued added m erchandise in line with the aspirations of the economic

blueprint, Vision 2030.

6.3 Policy Implications

Akey question for Kenyan policy makers is 'why has substantive trade policy reform in Kenya

produced to date a limited response in terms of the growth and diversification of exports? The

study identified the existence of the RTA as a significant determinant in explaining export

values from Kenya into the trading partners underscoring the importance of Regional Trade

Arrangements in trade policy orientation. Specifically, there is need for careful consideration

for the development of an external trade policy focused on regional trade facilitation and trade

expansion.

The trend analysis showed that Kenya benefits when the EAC was in existence due to
increased exports. In order to consolidate the gains there is need to conclude the formation of a

common market and provide political goodwill to prevent a second disintegration of the new

EAC. A common market is preferable as it will allow in addition to goods and services free

movement of factors of production like capital, labour, and entrepreneurship. Kenya should

therefore undertake domestic policy reforms to accommodate common market issues like land

ownership, citizenship and the use of identity cards as a traveling document.

To enhance trade facilitation, efficiency in Mombassa port and border controls should be

increased while the EAC partner states develop a common infrastructure development policy

to reduce the infrastructure costs. Currently, the EAC is in the process of developing an EAC
Transport strategy and Road Sector development programme to identify regional strategic

priorities and transport sector development. In addition EAC partner states in collaboration

with the business associations need to establish one stop investment authority to market the

EAC region as single investment destination to increase the share of both domestic and

Foreign Direct Investment (FDI) in the partner states. This will help offset the negative
perception of the region as being unstable and provide a more secure and predictable

investment climate in the East Africa region.

1 49
NOTES

1. RTAs is a process where two or more countries in a particular region/area

voluntarily join together to pursue common objectives for the mutual benefit of
members mainly economic in nature but with inherent social and political
implications.

2. Principle of Asymmetry integration is where some member countries are allowed to

retain tariffs than the agreed CET on other members; within the EAC this is to
temporarily protect producers in Tanzania and Uganda from the increased
competition from some Kenyan products.

3. BCI is calculated on a scale of 0-100 where 0 represents poor condition and 100
excellent condition, through a sample survey of 240 business leaders, Uganda (25%),
Kenya (21%), Tanzania (20%), Rwanda (18%) and Burundi (15%), 140 truck drivers,
and 187 clearing and forwarding agents, on a set of 13 predetermined business
factors are considered.

4. Complementary Index is a measure of similarities between the export basket of one

country and the import basket of another country. The value of the complementarity
index can range from 0, which represents no complementarity between the exports
and imports of two countries, to 100, which implies a perfect match. The higher the
index between two countries, the greater the product complementarity.

5. The gravity model has a long history in the social sciences. It has been used to
explain social flows, primarily migration, in terms of the "gravitational forces of
human interaction." Its name is derived from its similarity to Newtonian law of
gravitation in that large economic entity such as countries or cities are said to exert
pulling power on people or their products.

6. SMART is a static partial equilibrium model operable under strict ceteris paribus
conditions. It provides a snapshot of the projected impact of tariff reductions while
disregarding any adjustment process accompanying this change. Thus, the
dynamics that affect the change are not explicitly modeled, nor can complex
variations in the setup be considered.

7. Trade intensity indices provide additional insights into the nature and importance of
secular changes in bilateral trade flows. These indices can highlight the relative
importance of (seemingly minor) changes in trade between countries that have
relatively small global trade shares.

8. The model is an imperfect substitute model where imperfect substitutability

between domestic and export products enables domestic and export prices to differ
from one another (Goldstein and Kahn, 1985).

50
7.0 REFERENCES

Aitken, N.D. (1973). 'The Effect of EEC and EFTA on European Trade: A temporal Cross-

section Analysis', America Economic Review, Vol 63, No 5.pp. 881-892.

Alemayehu, G and Haile K. (2002). "Regional Economic Integrations in Africa. A review of

problems and prospects with a case study of COMESA" A Seminar paper presented in

Addis Ababa. Ethopia.

Amjadi, A., and L.A. Winters. (1997). Transport costs and "natural" integration in Mercosur.

Policy Research Working Paper 1742. International Trade Division. The World Bank.

Washington, D.C.

Appleyard, D and Field, A. (2001). " International Economics", 4th Edition, McGraw

Hill/Irwin. Singapore.

Aryeetey, E. (1997). "Sub-Saharan Africa Experience with regional Integration" in Zubair

Iqbal and Mohsin S.Khan, eds. Trade reforms and regional integration. International

Monetary Fund: Washington, D.C.

Bachrach, C., and L. Mizrahi. (1992). The economic impact of a free trade agreement

between the United States and Mexico: ACGE analysis. In Economy-wide modelling of the

economic implications of a FTA with Mexico and a NAFTA with Mexico and Canada.

Publication No. 2508. U.S. International Trade Commission. Washington, D.C.

Bagamuhunda, K., 2005. "Update on EAC-Customs union," East Africa Community

Secretariat. Arusha.Tanzania.

Bhagwati, J (1993). "Regionalism and Multilateralism: An Overview," in New Dimensions in

Regional Integration. Jaime de Melo and Arvind Panagariya, eds. Cambridge University
Press, U .K .

Balassa, B. (1967). "Trade Liberalization and "Revealed" Comparative Advantage". The

Manchester School, Vol 33, No 2.pp. 99-123.

51
Basu,A, E.A Calamitis and D. Ghura. (2000). "Promoting growth in Sub-Saharan Africa:

Learning what works". Economic Issues No.23. International Monetary Fund: Washington,

DC.

Baunsgaard, Thomas and Michael Keen. (2004). "Trade Revenue and (or?) Trade

Liberalization", September International Monetary Fund: Washington, DC.

Bender, S. and Li, K. (2002). "The Changing Trade and Revealed Comparative Advantages

of Asian and Latin American Manufacture Exports". Yale Economic Growth Center

Discussion Paper No. 843.

Charemza, W.W and D.F Deadman. (1992). New directions in econometric practice: general

to specific modeling, cointergration and vector auto regression. Edward-Elgar Publishing.

Clausing, K. (2001). "Trade creation and Trade diversion in the Canada-United States Free

Trade Agreement". Canadian journal o f Economics,V ol.34, No 3. pp. 677-696.

Clarete, R. Edmonds, C. and Wallack, J. (2002). Asian Regionalism and its Effects on Trade

in the 1980s and 1990s. Asian Development Bank. ERD Working paper series No. 30.

Deardorff, Alan V., 1984. "Testing trade theories and predicting trade flows," Handbook of

International Economics, in: R. W. Jones & P. B. Kenen (ed.), Handbook o f International

Economics, Volume 1, No 1. pp. 467-517.

DeRosa, D.A. (1995). Regional trading arrangements among developing countries: The

ASEAN example. Research Report 103. International Food Policy Research Institute:

Washington, D.C.

Dickey, D.A. and Fuller, D.A. 1979. Distributions of the estimators for autoregressive time

series with a unit root, journal o f American Statistical Association, Vol 74, No. 366 pp. 427-
431.

52
Dimelis, S. and Gatsios, K. (1995). "Trade with Central and Eastern Europe: The Case of

Greece, in R. Faini and Portes, R. (eds.) EU Trade with Eastern Europe: Adjustment and

Opportunities". London. C.E.P.R.

Dollar, David and Aart Kraay (2001). "Trade, Growth, and Poverty". Washington, D.C.

Drvsdale, P. and Garnaut, R. (1993). 'The Pacific: an application of a general theory of

economic integration', in Pacific Dynamism and the International Economic System, ed.

Bergensten, C.F. and Noland, M. Institute for International Economics: Washington, DC.

East Africa Community (EAC) Secretariat, (1999). Treaty for the Establishment of the East

African Community. Arusha.Tanzania.

East Africa Community (EAC) Secretariat, (2004). Protocol on the Establishment of the East

African Customs Union. Arusha.Tanzania.

East Africa Community (EAC) Secretariat, (2006). Development Strategy 2006-2010 and

EAC Facts and Figures Socio-economic. Arusha.Tanzania

Edwards, S. (1998.). Openness, Productivity and Growth: What do We Really Know? The

Economic Journal, Vol. 108, No.147. pp. 383-398.

Edwards, L. and Wilcox, O. (2003). Exchange rate depreciation and the trade balance in

South Africa. Paper prepared for the National Treasury, 2003.

Egger, H., Egger, P. and Greenaway, D. (2004). Intra-Industry Trade with Multinational

Fimas: Theory, Measurement and Determinants. Leverhulme Center Research Paper.

Elbadawi, Ibrahim A. and Francis M. Mwega. (1998). "Regional Integration, Trade, and

Foreign Direct Investment in Sub-Saharan Africa", in Zubair Iqbal and Moshin S. Khan (eds.),

Trade Reform and Regional Integration in Africa. International Monetary Fund:

Washington, D.C.

53
Foroutan, F. (1993). 'Regional Integration in Sub-Saharan Africa: Past Experience and Future

Prospects', in J. De Melo and A. Panagariya (eds), New Dimensions in Regional Integration.

Cambridge University Press, Cambridge,U.K.

Francois, Joseph, and Clint Shiells, eds (1994). Modeling Trade Policy: Applied General

Equilibrium Assessments of North American Free Trade. Cambridge University Press.

Gasiorek, M., A. Smith, and A.J. Venables. (1992). Trade and welfare: A general equilibrium

model. In Trade flows and trade policy after "1992", ed., L. A. Winters. Cambridge
University Press Cambridge, U.K.

Glenday, G. and Ndii, D. (2000). Export Platforms in Kenya. Discussion paper No.752.

Harvard Institute for International Development. Harvard University .Cambrige.

Goldstein, M. and Khan, M.S. (1985). Income and price effects in foreign trade, in R. Jones

and P. Kenen (eds), Handbook o f International Economics, Vol. 2, North-Holland, Amsterdam.

Gotur, P. (1985). Effects of Exchange Rate Volatility on Trade: Some Further Evidence."
IMF Staff Papers. Vol. 32 (3). pp. 475-512. International Monetary Fund: Washington, DC.

Grubel, H and Lloyd, P.J (1975). "Intra-industry trade: The theory and measurement of

international trade in differentiated products". New York, Wiley.

Gujarati, D. N. (1995). Basic Econometrics, 3rd ed., New York: McGraw-Hill Press.

Harrison, G.W., T.F. Rutherford, and D.G. Tarr. (1994). Production standards, imperfect
competition, and completion of the market in the European Union. Policy Research
Working Paper 1293. International Economics Department, The World Bank. Washington,
DC.

Haaland, J., and V. Norman. (1992). Global production effects of European integration. In

Trade flow's and trade policy after "1992", ed., L. A. Winters. Cambridge University Press.

Cambridge, U.K.:

54
Helpman, E. (1999). "The Structure of Foreign Trade", journal o f Economic Perspectives

VoLl3 No 2, pp. 121-144

Kalbasi Hassan. (1998). "The Gravity Model and Global Flows". A paper presented at

Department of Economics, University of Isfahan, Iran.

Keller, Wolfgang. (2000). "Do Trade Patterns and Technology Flows Affect Productivity

Growth?" The International Bank for Reconstruction and Development. The World Bank.

Washington D.C.

Khandelwal, P. (2004). "COMESA and SADC: Prospects and Challenges for Regional

Integration", Draft Working Paper, International Monetary Fund: Washington D.C.

Kirkpatrick, Colin and Watanabe, Matsuo. (2005). "Regional trade in Sub-Saharan Africa:

An analysis of East African Trade Cooperation, 1970-2001," The Manchester School, Vol. 73,

No. 2. pp. 141-164.

Klaassen, Franc, 2004. "Why is it so difficult to find an effect of exchange rate risk on

trade?," journal o f International Money and Finance, Vol. 23, No 5. pp. 817-839.

Kreinin Mordechai, and Plummer Michael, (2000). Effects of Regional integration on FDI:

an empirical approach, journal o f Asian Economics, Vol 19. No. 5-6 pp 447-454..

Krueger, Anne O. (1999). "Trade Creation and Trade Diversion under NAFTA", NBER

Working Paper No. 7429, National Bureau of Economic Research, Inc, Cambridge.

Kristjansdottir, Helga (2005). "A Gravity Model for Exports from Iceland". University of

Iceland and the Centre for Applied Micro Econometrics.

Krugman, Paul (1981). "Intra-industry specialization and the gains from trade), journal o f

Political Economy, Vol.89, No.S.pp. 959-973.

55
Lewis, J.D., and S. Robinson. (1996). Partners or predators? The impact of regional trade
liberalization on Indonesia. Policy Research Working Paper 1626. Country Operations
Division. The World Bank. Washington, D.C.

Lipsey, R.G. (1957). The theory of customs unions: Trade diversion and welfare. Economica
Journal, Vol. 24 No 93. pp 40-46.

___. 1960. The theory of customs unions: A general survey. Economic Journal,Vol. 70 No 279
pp. 496-513.

___. 1970. The theory of customs unions: A general equilibrium analysis. London:
Weidenfeld and Nicolson.

____, and K. J. Lancaster. (1956/57). The general theory of second best. Reiriew o f Economic
Studies,Vol. 24 Nol. pp. 11-32.

Lucio Castro, (2005), "Regional Trade Integration in East Africa: Trade and Revenue

Impacts of the Planned East African Community Customs union," International Trade

No.0509005, EconWPA.

Manuel, T. A. (2006). Ministerial Roundtable Discussions on Infrastructure Development and

Regional Integration: Issues, Opportunities and Challenges. National Treasury, Republic of South
.Africa.

.Martin, Will, and L. Alan Winters, eds (1995). The Uruguay Round and the Developing

Economies. Discussion Paper No. 307: World Bank Washington, DC.

McIntyre, Meredith (2005). Trade Integration in the East African Community: An

assessment for Kenya. IMF Working Paper WP/05/143.

Meade, J.E. (1955), The theory of customs unions. Amsterdam: North-HoIIand.

MWega, F.M. (2002).Promoting non-traditional exports in Kenya: 1980-1996. In Helleiner,

G.K (ed). Non-traditional export promotion in Africa: Experience and Issues. Palgrave.

New York.

56
M ontenegro, C. and Soto, R. (1996). "How Distorted is Cuba's Trade? Evidence and

Predictions from a Gravity Model", Journal o f International Trade and Economic Development,

Vol. 5, No. 1, pp. 45-68.

Nassali, M. (2005). "The East Africa Community and the struggle for constitutionalism;

challenges and prospects.

Ndulo, B.J and N. S. Ndung'u (1998). "Trade and growth in Sub-Saharan Africa". In Iqbal,

Z and M.S Khan (eds).Trade reform and regional integration in Africa. International

Monetary Institute. Washington DC.

Ndung'u, N. S. and R.W. Ngugi (1999). Adjustment and liberalization in Kenya: the

financial and foreign exchange markets. Journal o f International Development II Vol 11, No 3,

pp. 465-491

Ng'ang'a, W. (2006). "The "N ew " East African Community: Effects on Trade, Welfare and

Productive Activities in East Africa". University of Saskatchewan. Saskatoon.

Ng’eno, N em ehiah K., Herzron. O. Nyangito, Moses M. Ikiara, Eric E. Ronge and Justus

Nyamunga (2001). "Regional Integration in East Africa: The Case of Kenya". Paper

Submitted to the African Centre for Economic Growth (ACEG), Nairobi, Kenya.

Ng'eno, N. (2002). "The Status of Regional Trade Liberalization in East Africa". African

Centre for Econom ic Growth; Nairobi.

Ogunkola, E. O. (1998). 'An Empirical Evaluation of Trade Potential in the Economic

Community of West African States', (AERC). Research Paper 84, African Economic

Research Consortium Nairobi.

Ogun, O.M (1991). Structural adjustment and agricultural supply response in Sub-Saharan

Africa: synthesis and limits of current analysis. Research fo r Development, Vol.6, No.2, pp. 22-

23.

57
Okello A. Charles. (2006). The impact of East Africa community customs union on Uganda

economy: a Computable General Equilibrium (CGE) analysis. Mimeo. Department of

Economics. Gulu University. Uganda.

Ovejide, A., I. Elbadawi and P. Collier, (eds) (1997). "Regional Integration and trade

liberalization in Sub-Saharan Africa". Volume 1: Framework, issues, and methodological

perspectives. Macmillan Press Ltd. London.

Panagariya, A. (1996). 'The Free Trade Area of the Americas: Good for Latin America?'

World Economy Vol 19No.5, pp. 485-515.

Panagariya, A. (2004). "Miracles and Debacles: In Defense of Trade Openness". The World
Economy, Vol.27, pp. 1149-1171.The World Bank. Washington, D.C.

Prewo, W.E. (1974). Integration effects in the EEC: An attempt at quantification in a general
equilibrium framework. European Economic Review,Vo\ 5, No 3 pp. 379-405.

Republic of Kenya. (2008), Vision 2030, Government Printer. Nairobi.

Republic of Kenya. Ministry of Industrialization, Strategic Plan 2008-2012, Nairobi.

Reinnika, R. (1996). The credibility problem in trade liberalization: empirical evidence from

Kenya. Journal o f African economies, Vol 5 No 3. pp. 444-468.

Rothchild, D. (1968), "Politics of Integration: An East African Documentary". Nairobi: East

African Publishing House.

Roland-Horst, D.W., K.A. Reinert, and C.R. Shiells. (1992). North American trade

liberalization and the role of Non-tariff barriers. In Economy-wide modelling of the

economic implications of a FTA with Mexico and a NAFTA with Mexico and Canada.

Publication No. 2508. U.S. International Trade Commission. Washington, D.C.

Santos-Paulino, A.U. (2000). "Trade liberalization and export performance in selected

developing countries". Keynes College, University of Kent. U.K.

"•w ry
EiSf AfliHi,*;' 5g
Shuo Zhang, Jan Ondrich,J, David Richardson,(2003).Mimeo, Department of Economics,

Syracuse University.

Schiff, M., and L.A. Winters. (1996). Regionalism and development: International

Economics Department, The World Bank. Washington, D.C.

Soloaga, Isidro and Winters, Allan (2000). Regionalism in the Nineties: What Effect on

Trade and Development Economics Group, World Bank.

Straub, S. (2007). In frastru ctu re and developm ent: a critical a p p raisal of th e m acro level
literature. U niversity o f Edinburgh. Uk.

Straub, S. (2008). In frastru ctu re and grow th in developing co u n tries: recent ad v an ces and research
challenges. T h e W o rld Bank D evelopm ent Research D ep artm ent. W orld Bank. W ashin gton D.C.

Treaty for East African Co-operation. 6th, June 1967. East African Community. Arusha.

Tanzania.

Truman, E.M. (1969). The European Economic Community: Trade creation and trade

diversion. Yale Economic Essays, Vol 9. No 2, pp. 201-257. Yale University.USA.

United States International Trade Commission, (1992). "Economy-wide Modeling of the

Economic Implications of a FTA With Mexico and a NAFTA with Canada and Mexico."

USITC Publication No. 2508.

Viner, J. (1950). "T h e Customs union issue". New York: Carnegie Endowment for
International Peace.

Vollrath, Thomas L., (1991). "A Theoretical Evaluation of Alternative Trade Intensity

Measures of Revealed Comparative Advantage," Weltxvirtschaftliches Archiv-Review o f World

economics, Vol. 127 No 2. pp. 265-80.

World Bank (2004). World Development Indicators. Washington, D.C.

59
1
WTO (W orld Trade Organization) (2003). World Trade Report 2003, Geneva, WIO.

Yarmrick, S and Gosh (2002). "Are Regional Trade Arrangements Trade Creating? An

Application of Extreme Bound Analysis" .A seminar paper presented at Midwest Economic

Association.

Yeats, A. (1997). Does Mercosur's trade performance raise concerns about the effects of
regional trade arrangements? Policy Research Working Paper No.1729. International Trade
Division, The World Bank. Washington, D.C.

60