ABSTRACT

EVALUATION OF THE AFRICAN ASSOCIATED STATES RESPONSE
TO TARIFF PREFERENCES GRANTED BY THE EUROPEAN
ECONOMIC COMMUNITY

By
Moise Allal

In 1958, 18 African countries were associated with
the European Economic Community (EEC), and were thus granted
preferential treatment for their exports to EEC countries.
The purpose of this study was to evaluate the way in which
14 members of the African Associated States (AAS) res-
ponded to the EEC tariff preferences during the 1962—1969
period. The evaluation of the AAS response was limited to
manufactured exports.

The study was designed to meetttmee1major objectives.

URN; first objective of the study was to evaluate
empirically the impact of tariff preferences on AAS manu—
factured exports_to the EEC. For this purpose, a methodolo-
gy was developed which takes into consideration changes in
EEC demand and AAS supply conditions over the period of ana-
lysis (1962-1969). The methodology makes use of two groups
of "control" countries: "Other LDCs" (i.e., non-beneficiary
developing countries) and "Other DMECs" (i.e., developed

market economy countries other than the EEC countries). It

Moise Allal

was applied at various degrees of commodity aggregation,
from total AAS manufactured exports to commodities at the

5—digit level of the CST Commodity Classification (i.e., the
classification system adopted by the EEC). It was shown

that, in general, the AAS failed to respond to EEC pre-
ferences by expanding manufactured exports to EEC countries
at a higher rate than would have prevailed in the absence
of such preferences. With respect to individual manu-
factured commodities, the analyses indicated a positive

AAS response for only a small fraction of the total number
of commodities investigated, these being generally raw
materials - intensive semi-manufactures. Moreover, newly
produced consumer goods (i.e., goods first produced after
the granting of EEC preferences) were exported primarily to
develOping countries in Africa.

Tfimaarond Objective of the study was to test a trade
model based on the Linder similarity of preferences theory in
order to determine whether the theory explains the geographi-
cal intensity of AAS exports. A regression equation of the

following general form was tested:

(API) f( IAPCII, D, L, P)

where:

API 2 average propensity to import.

IAPCII

ll

absolute difference between the per capita
income of the exporting country and that

of the importing country.

Moise Allal

D 2 distance between the importing and exporting
countries.

L 2 binary variable (values of O or 1). If the
official language of the importing country
is the same as that of the exporting country,
L : 1. If it is not the same, L = O.

P : binary variable (values of O or 1). If a
special trading agreement exists between
the importing and exporting countries,

P = 1. If it does not, P = O.

The Linder theory implies that the regression
coefficients for IAPCII and D should be negative while the
coefficient for L should be positive. Furthermore, the
regression coefficient for P should not be significant.
Findings from the test generally supported the theory.

The regression coefficients for IAPCII, D and L were in
most cases significant and with the predicted sign, while
the coefficient for P was generally non—significant. More-
over, as implied by the Linder theory, the model applied

better to consumer goods than to semi—manufactures.

The third and final objective of the study was to
examine, on the basis of the evaluation of the AAS response
and the results of the test of the Linder theory, two con-
trasting trade strategies which could be adopted by the AAS:
a trade strategy based on an expansion of manufactured ex-

ports to industrialized countries, and a trade strategy

Moise Allal

based on an expansion of intra-regiona] trade. It was argued
that, given the characteristics of the AAS, the latter stra-

tegy would have more chance to succeed than would the former

strategy.

EVALUATION OF THE AFRICAN
ASSOCIATED STATES RESPONSE TO
TARIFF PREFERENCES GRANTED BY

THE EUROPEAN ECONOMIC COMMUNITY

By

Moise Allal

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Economics

1977

G ’07 {ENG}?

ACKNOWLEDGMENTS

I wish to express my gratitude and indebtedness
to Professor Mordechai Kreinin, chairman of my dissertation
committee, for his guidance, suggestions, encouragement,
and patience at every stage of the preparation of this
dissertation. I am also grateful to Professor Lawrence H.
Officer and Dr. James M. Johannes for their useful comments
and suggestions.

I am greatly indebted to my wife, Linda, for her
very valuable comments and criticisms as well as for her
contribution to the editing of this text.

I wish, finally, to thank Mrs. Francesca Rosner for

her diligent and skillful typing of this manuscript.

ii

LIST OF TABLES

Chapter

TABLE OF CONTENTS

I. INTRODUCTION

1. The Association Convention between the
EEC and the African Associated States

2.

Objectives of the study

II. IMPACT OF TRADE PREFERENCES: THEORETICAL
CONSIDERATIONS . . . . .

l.

The Theory of Customs Unions and

Trade Preferences .
Determinants of the Static
Effects of a Union

. Application of the Theory of Customs
Unions to the EEC-AAS Trade Agreement

Impact of Tariff Preferences
on AAS Exports of Standardized
Manufactures . . .
Impact of Tariff Preferences

on AAS Exports of Differentiated

Commodities . . . . . .

. Methodology for the Evaluation of the
AAS Response to EEC Tariff Preferences
Comparing Hypothetical Estimates

to Actual Values . . .
Comparing Pre- -Integration to
Post- -Integration Estimates of
Various Parameters

Use of Multiple Regression
Analysis

Evaluation Method Used in this
Study . . . . . .

iii

 

Page

vii

l3

l4

l7

l9
19

2O
22
26

III.

4.

5.

Trade Theories and Tariff Preferences
Potential Constraints on the
AAS Response . . . . . .

The Linder Similarity of Preferences

Theory and Tariff Preferences . . .
The Linder Theory .
Previous tests of the Linder

Theory .

Linder Theory and Tariff

Preferences . . . . . .
LIST OF REFERENCES . . . . . . . .

EVALUATION OF THE AFRICAN ASSOCIATED STATES

RESPONSE TO TARIFF PREFERENCES GRANTED
BY THE EEC . . . . . . . . .

1.

2.

The Evaluation Framework . . . .
Period of Analysis . . . . .

Analysis of Total AAS Exports and

EEC Imports .
Changes of the Pattern of EEC
Imports and AAS Exports over
the Period of Analysis . . .
AAS Response to Tariff Preferences
for Over-A11 Exports . .

Evaluation of AAS Response for Over-All
Manufactured Exports . . . . . . .

. AAS Exports of Various Groups of

Manufactures to the EEC and Other Groups

of Countries . . . . . . . . . .
The Data . . . . .
Evaluation of AAS Export
Performance during the 1962-1969
Period . .
Summary of Findings on the AAS
Export Performance . . . .

. Evaluation of AAS Response with Respect

to Individual Commodities at the 3-digit
Level of the SITC Classification

. Evaluation of AAS Response with Respect

to Individual Commodities at the 5-digit
Level of the CST Classification . . .

iv

Page
30
33
22;
41
47
49

51
51
52

55

56
58

61
64

65

72
82

83

87

Page

The Commodities . . . . . . 87
Evaluation Methodology . . . . . 90
Findings for Individual

Commodities . 98

Findings from Regression Analyses . 111

7. Manufactured Commodities Exported to
Groups of Countries other than the EEC . 112

8. AAS Response to EEC Tariff Preferences:
Summary of Findings and Concluding

Remarks . . . . . . . . . 117

LIST OF REFERENCES . . . . . . . . . 123

IV. DETERMINANTS OF THE GEOGRAPHICAL
INTENSITY OF AAS EXPORTS . . . . . 124

l. The Model . . 125

2. The Data . . . . . . . . 133
Grouping of Exports . . . . . . 133

3. Test of the Model for the AAS
as a Whole . . . . . . . . . 136
Test Findings . . . 136
Examples of API Values Predicted
by the Regression Equation for

the AAS as a Whole . . . . . . 142
4. Test of the Linder Model for
Individual AAS Countries . . . . . . 146
Test Findings . . . . . . . . 147
5. Concluding Remarks on the Test
of the Linder Model . . . . . . . 153
LIST OF REFERENCES . . . . . . . . . 158

V. DETERMINANTS OF THE AAS EXPORT PATTERN
AND EVALUATION OF TRADE STRATEGIES FOR
THE AFRICAN ASSOCIATED STATES . . . . . . . 159

1. Possible Explanations for the lack
of AAS Response to EEC Tariff
Preferences . . . . . . . . . 160

2. Possible explanations for the high Trade
Intensity within the AAS and between AAS
Countries and non-AAS African Countries . 164

Page

3. Trade Strategies for the AAS . . . 166
Evaluation of an Export Strategy
Based on an Expansion of
Manufactured Exports to
Industrialized Countries . . . . 168
Evaluation of a Trade Strategy
Based on Intra-African Trade . . . 177

LIST OF REFERENCES. . . . . . . . . 184
VI. SUMMARY AND CONCLUSIONS . . . . . . . . . 185
1. Summary of Findings . . . . . . . 185
2. Potential Impact of Current General
System of Preferences (GSP) Schemes
on Manufactured Exports from
Developing Countries . . . . . . . 188

APPENDIX . . . . . . . . . . . . . . . . 191

BIBLIOGRAPHY . . . . . . . . . . . . . . 251

vi

Table

10.

11.

12.

LIST OF TABLES

Tariff Reductions of the EEC on
Manufactured Imports from the AAS . . .

Growth of Value and Market Share of Total
EEC Imports and Total AAS Exports (1962-1969) .

Total AAS and "Other LDCs" Exports to EEC
and "Other DMECs" (1962-1969) . . . . . .

Imports of Manufactures and Semi—Manufactures
by EEC and "Other DMECs" from the AAS
and "Other LDCs" (1962-1969) . . . . . .

Over-All Share of Intermediate Goods and of
Finished Goods in Individual Groups of
Manufactured AAS Exports . . . . . . . .

AAS Exports of Individual Groups of
Manufactured Commodities - Growth Rate
of Value and Market Share (1962-1969) . .

EEC and "Other DMECs" Imports from.AAS
and "Other LDCs" - Mean Values and
Growth Rates for the period 1962-1969

Major S-digit Level Commodities Exported
by the AAS to the EEC . . . . . . . . .

EEC Imports of 29 Major Commodities from
the AAS, "Other LDCs", EEC, and "Other DMECS +
Soc. Countries" . . . . . . . . . . .

Summary Table of the Evaluation of AAS response
to Tariff Preferences for 29 Major Commodities.

Growth rate of AAS Exports of 29 Major
Commodities to the EEC, ROW, and Africa -
(1962-1969) . . . . . . . . . .

'Manufactured Commodities Mostly Exported to

Africa . . . . . . . . . . . . .

vii

Page

54

57

59

63

7O

73

85

88

99

103

107

113

Table

13.

14.

15.

16.

17.

18.

19.

.A-l.

A-2h

A~3

1p4a

Results of the Test of the Linder Model
for the AAS as a Whole . . . . . . .

Examples of API Values Predicted by the
Regression Equation for the AAS as a Whole .

Commodity Groups and AAS Countries
Selected for Testing . . . . . . . .

Summary of Results of the Test of the
Linder Model (Individual AAS Countries)

Inter—Regional and Intra-Regional Trade
Intensities. SITC Groups 6 + 8 - (67+68) -
1969 . . . . . . . . . . . . .

Major Exporters of Manufactures to
21 DMECs, 1962, 1972 . . . . . . . .

Major Trade Flows Between Developing
Countries and 21 DMECs in 1972

Manufactured Commodities Exported
by Each AAS Country - (1962-1969) . . .

Value of AAS Export of Individual Groups
of Commodities to Individual Groups of
Countries - (1962-1969) . . . . .

AAS Export of Individual Groups of
Commodities to Individual Groups of
Countries - Market Shares - (1962-1969)

Import of Manufactures and Semi-Manufactures
by the EEC and "Other DMECs" from the AAS
and "Other LDCs" . . . . . . *. . .

EEC Imports of 29 Major Commodities from
Various Groups of Countries - Value and
Market Share - (1962-1969) .

EEC Imports of 29 Major Commodities from
Various Groups of Countries -
Quantity and Price - (1962—1969) . . .

Value and Market Share of AAS Exports of

29 Major Commodities to the EEC, Africa,
and the Rest of the World - (1962-1969) .

viii

Page

137

145

148

149

156

172

174

191

206

210

213

215

220

225

Table Page

A-S. GNP, PCI and Language of Countries
Trading with the AAS - 1967 . . . . . . 229

A-6. Distance Between Individual AAS
Countries and Individual Importing
Countries . . . . . . . . . . . . 232

A—7. Results of the Test of the Linder Model
for Individual AAS Countries . . . . . . 237

ix

CHAPTER I

INTRODUCTION

In trade negociations over the past two decades, it
has often been proposed that industrialized nations could
greatly contribute to the economic development of the less
developed countries by granting tariff and quota preferences
to manufactured imports from these countries. Politically, if
not economically, the granting of preferences has been re—
garded by the developing countries as a superior form of aid
because the preferential system, once established, would be
less sensitive to political conditions in the preference
granting countries.

It is not certain, however, that the granting of
tariff preferences will yield the positive impact expected by
developing countries. Many of these countries may be unable
to supply the type and quality of goods demanded in industrial-
ized countries. A number of factors - such as a low technolo-
gical level, high transport costs, lack of marketing skills -
may be much more constraining than the tariffs and quotas
imposed by industrialized countries on manufactured imports.
It is therefore of interest to analyze the impact of preferen-
tial schemes which have been in existence for a period of time.
Such an analysis could help developing countries to reassess
their trade strategies, especially those based on expansion of

manufactured exports to industrialized countries.

1

In 1958, 18 African countries were associated with the
European Economic Community (EEC), and were thus granted pre-
ferential treatment for their exports to EEC countries. The
purpose of this study is to evaluate, in the light of existing
trade theories, the way in which 14 members of the African
Associated States (AAS)l responded to EEC tariff preferences
during the 1962-1969 period.

Although the AAS countries export mostly agricultural
and raw materials, this study focuses on manufactured exports
for the following reason. The General System of Preferences
(GSP) schemes implemented by various industrialized countries
since 1971 apply to manufactured exports from developing
countries.2 Findings from this study may therefore be useful
in predicting the impact of these schemes on manufactured ex—
ports from countries at a similar level of development as the

AAS countries.

 

1The 14 countries are Cameroon, Togo, Senegal,
Madagascar, Ivory Coast, Congo, Dahomey, Niger, Upper Volta,
Chad, Mali, Mauretania, Gabon, and Central African Republic.
Due to lack of data, 4 AAS countries (Zaire, Rwanda, Burundi,
Somaliland) are not included in the study.

2On July I, 1971, the EEC exempted from custom duties
(within the limits of a tariff-quota system) all imports of
manufactures and semi—manufactures from 91 developing countries.
In the following months, other industrialized countries (Japan,
Norway, U.K., Denmark, Finland, Ireland, New Zealand, Sweden,
Switzerland and Austria) followed the EEC move and implemented
their own General System of Preferences (GSP). The U.S. GSP
scheme was implemented in 1976.

The tariff preferences granted by the EEC to the AAS
are much less restrictive than those included in the various
GSP schemes. Furthermore, during the 1962—1969 period (i.e.,
the period covered by this study), the AAS could still benefit
from a potentially high level of trade diversion since most of
the other developing countries did not yet benefit from similar
trade preferences. Thus, the AAS countries response to EEC
preferences should be greater than.that of similar countries
to the GSP schemes. A failure of the AAS to respond positively
and significantly to the EEC preferences would raise doubts as
to the usefulness of the GSP schemes in expanding manufactured
exports from highly underdeveIOped countries to industrialized
countries.

1. The Association Convention between the EEC
and the African Associated States

During the negotiations leading up to the 1957 Treaty
of Rome, France campaigned vigorously for the association of
its overseas territories with the European Community. As a
result, 18 African countries, colonies of France, Belgium and
Italy, became "associate members" of the EEC on January 1,1958.

The goals of the association convention were to per-
petuate the existing preferential treatment granted by some EEC
members to their colonies. As associate members, the AAS
countries cannot legally influence the policies of the EEC, do
not have to contribute to the budget of the community, and are

not expected to carry out the obligations of the Treaty of

4

Rome, except for the special provisions of the Treaty which
apply to them.

The Treaty of Rome, which became effective on
January 1, 1958, granted the AAS substantial trade preferences,
as well as important financial aid and technical assistance.

In return, the AAS countries granted the EEC countries reverse
preferences in trade and investment. The association conven-
tion was extended in 1963 for a five-year period (First Yaoundé
Convention), and on January 1, 1970, was extended for another
five-year period (Second Yaounde Convention). The Yaounde
Conventions introduced certain distinctions between the treat-
ment of the AAS and-that of the French Departments and Terri-
tories. The provisions of these conventions have recently
been superceded by the Lomé Convention of February 28, 1975,
which covers trade and aid relations between the present
members of the EEC and 46 African, Caribbean and Pacific
countries (ACPs), including the AAS. The Lomé Convention
terms represent a change in policy by the EEC on reverse
preferences. The ACP countries are not required to grant
reverse preferences, but must give the EEC the most favored
nation treatment and must avoid discrimination between EEC
member states.

In many instances, the tariff preferences granted by
the EEC to the AAS were less favorable than those which existed
prior to 1958 under bilateral arrangements between individual
EEC members and their colonies. However, since preferences

apply to a much larger market than that of the metropolitan

country, the AAS countries should benefit from the associatiOn.
The Treaty of Rome and the two Yaounde Conventions
contain a number of special provisions with respect to tropical
products of great interest to the AAS. With respect to manu—
factured exports, the AAS enjoys the same tariff preferences
as individual EEC members. From 1958 to 1968, tariff duties
imposed on manufactured imports by EEC countries from the AAS
were gradually reduced, reaching a zero value in 1968. During
the same period, the EEC countries imposed a common external

tariff on manufactured imports from non—associated countries.

2. Objectives of the study

 

This study is designed to meet four major objectives.

First, trade preferences granted by the EEC to the AAS
are examined in the light of existing trade theories in order
to identify and compare the various types of response to pre-
ferences suggested by these theories.

Second, the effect of tariff preferences on the AAS
export of manufactures to the EEC is evaluated empirically
both from the view point of AAS exports and from that of EEO
imports. In the case of AAS exports, the growth of the value
of exports of various manufactured goods to the EEC and other
regions of the world is estimated for the period 1962-1969.
The growth of the market share of exports to these various
regions is also estimated for the same period. A similar
analysis is undertaken for the same commodities and the same

period with respect to EEC imports. In this analysis the

growth of EEO imports from the AAS and from other regions of
the world are estimated in terms of both value and market
share. The AAS response to tariff preferences is subsequently
evaluated on the basis of a comparison of import and export
growths using the method described in chapter II.

Third, a trade model based on the Linder similarity of
preferences theory is tested with respect to AAS exports. The
test is performed for each AAS country and for the AAS as a
whole. In each case, the test is repeated for 13 groups of
manufactured exports, seven groups of consumer goods, and six
groups of intermediate manufactured goods. Exports are divided
into consumer goods and intermediate goods in order to determine
whether, as implied by the Linder theory, the proposed trade
model applies better to the former goods than to the latter
ones. Test results are subsequently evaluated in the light of
the previous findings regarding the AAS response to tariff
preferences.

Finally, on the basis of the evaluation of the AAS
response and the results of the test of the trade model, a
number of foreign trade strategies which could be adopted by
the AAS countries are considered. These strategies are dis-
cussed with respect to various economic development models of

potential interest for the AAS.

CHAPTER II

IMPACT OF TRADE PREFERENCES: THEORETICAL CONSIDERATIONS

This chapter will present the implications of current
trade theories with respect to the potential impact of tariff
preferences granted by the EEC to the AAS. First, the theory
of customs unions will be briefly described, and then applied
to the EEC-AAS trading arrangement in order to identify the
potential effects of tariff preferences on AAS exports to the
EEC. Second, a methodology will be developed in order to
evaluate, in chapter III, the AAS response to EEC tariff pre-
ferences. Third, various trade theories will be reviewed, and
theh‘implications for AAS-EEC trade discussed. Particular
attention will be given to the Linder similarity of preferences
theory which provides the basis for the trade model tested in

chapter IV.

1. The theory of customs unions and trade preferences
Customs unions theory encompasses trading arrangements

from the simple preferential area to total economic integration.

 

The theory as first proposed by Viner (l),dealt mostly with the
effects of tariff preferences — or discriminatory tariffs - on
world welfare. It was not concerned with the effects of the
free flow of factors of production associated with a common
market such as that of the EEC, or the effects of common

monetary, fiscal, and social policies associated with total

7

economic integration.

Viner analyzed the effects of a customs union on the
production patterns of member and non-member countries. He
assumed that the establishment of a customs union does not
affect consumption in the member countries. The two main

effects of a union, according to Viner, are trade creation

 

and trade diversion. Trade creation occurs when tariff pre-
ferences lead to an increase in imports from lower—cost
member countries at the expense of higher-cost domestic pro—
duction. Trade diversion occurs when preferences lead to a
decrease in imports from lower-cost non-member countries, and
their displacement by imports from member countries.

In Viner's model, it is implicitly assumed that the
price elasticity of demand for any imported commodity is zero.
It is probable, however, that tariff preferences alter relative
prices, and lead to changes in consumption patterns. An
elastic demand means that the creation of a union will have
positive and negative consumption effects similar to the pro-
duction effects described above. Meade (2) analyzed these
consumption effects in a model which assumed a fixed pattern of
production (i.e., zero elasticities of supply), but allowed
the pattern of consumption to change with the creation of the
union. By comparing the ratios of the marginal utility of
products within individual countries - as indicated by their
domestic price ratios-Meade was able to evaluate the effects

of changes in the exports and imports of these countries.

There is actually no reason to treat production effects
and consumption effects separately since shifts in production
will affect consumption and vice versa. Lipsey (3) introduces

the concepts of inter—country substitution and inter-commodity

 

substitution in order to differentiate the effects of a union.
Inter—country substitution consists of trade creation and trade
diversion due to shifts in the locus of production, while
inter—commodity substitution consists of the substitution of
one commodity for another as a result of a change in relative
prices. Either substitution yields both production and con-
sumption effects.

The preceding review of the theory of customs union
was concerned only with the static effects of a union (i.e.,
consumption, production and terms of trade effects). A union
could also yield substantial dynamic effects in the form of
increasing competitiveness, increasing investments, and an
increasing rate of technological change. Since dynamic
effects take place over a relatively long period of time, and
the period of analysis used in this study (1962-1969) is

relatively short, only static effects will be considered.

Determinants of the static effects of a union

The static effects of a union are a function of a
number of variables whose impact may be evaluated within the
framework of a partial equilibrium analysis (1,2,5), or a
general equilibrium analysis. No attempt will be made to go

into the details of these analyses since they are fairly

IO

3

SI.

3

:dard, and may be found in a large number of books dealing
with the welfare impact of customs unions. Rather, a summary
of the results of these analyses will be provided.

The static effects of a union are mainly a function
of domestic supply and demand elasticities, the elasticity of
supply of exports from member and non—member countries, the
level of the preferential margin enjoyed by member countries,
and the production cost differential between member and non—
member countries.

The higher the elasticities of demand and supply for
goods traded between member countries within the union, the
greater will be the scope for trade creation. On the other
hand, the higher the elasticity of supply of exports from
non-member countries to the union, and the higher the elasti-
city of demand by member countries for these same exports, the
greater will be the scope for trade diversion. Trade diversion
will be larger than trade creation if the proportion of pre-
union trade with non-member countries is relatively high and
vice versa.

The level of tariffs will also affect trade creation
and trade diversion. The higher the level of pre—union tariffs
imposed on imports from member countries, the greater will be
the scope for trade creation. Similarly, the higher the level
of tariffs imposed by member countries on imports from non—
member countries, the greater will be the likelihood of a

shift in production from the latter countries to the former,

11

and therefore the scope for trade diversion. Finally, the
greater the efficiency of non—member countries over member
countries in production, the greater will be the welfare loss
per unit of trade diversion.

The determinants of static effects may also be analyzed
in the two special cases where the elasticity of demand and the
elasticity of supply are assumed to be, respectively, equal to
zero. If demand is assumed to be completely inelastic, trade
creation will be larger the greater the elasticity of supply
in member countries. Similarly, the larger the elasticity of
domestic supply in non-member countries, the greater will be
trade diversion. On the other hand, if the elasticity of
supply is assumed to be equal to zero, it may be shown that
the higher the original tariff of union members relative to
that of non-member countries, the greater will be the scope
for trade creation. Furthermore, the lower the degree of sub-
stitutability among products of member countries, and the
higher the degree of substitutabilitybehNeen products of
member countries and those of non-members, the smaller will be
the scope for trade creation and the larger the scope for trade
diversion.

One important question not raised above is whether the
net impact of trade creation and diversion is a function, all
else being equal, of the mix of countries entering into a union.
Prior to the publication of Viner's book, it was thought that a

union of complementary economies should yield a higher increase

12

in net welfare than a union of rival economies. Viner took an
opposite view arguing that a union of rival economies should
be more beneficial because there is greater scope for trade
creation than for trade diversion (l, p.51). Actually, it
can be shown that, depending on circumstances, either type of
union may be more beneficial to world welfare. Let us define:

QC

volume of trade resulting from trade

creation (in tons)

(AP)C = difference between the domestic price
of the higher—cost member country and
the C.I.F. export price from the lower-
cost member country (in dollars)

QD = volume of trade resulting from trade
diversion (in tons)

(AP)D

difference between the C.I.F. export price
from the lower-cost non—member country and
the C.I.F. export price from the higher—
cost member country.

Thus ( AP)C((.AP)D) measures the welfare gain (loss)
per unit of volume of trade, while QC(QD) measures the volume
of trade associated with trade creation (diversion).

If a union gives rise to both trade creation and trade
diversion, the net impact on world welfare, W, will be equal
to:

w = QC-(AE)C - QD-( AP)D

The difference (QC — OD) should be greater for a

l3

union formed of rival economies than for a union formed of
complementary economies. On the other hand, the difference
((AP)C - (AP)D) should be smaller for the former union than
for the latter one. Thus, it cannot be concluded that one
type of union is necessarely more beneficial to world welfare
than the other. '

Having defined, in broad terms, the static effects of
a union and reviewed the factors which determine the size of
these effects, the general conclusions derived above will now
be applied to the particular case of the EEC—AAS trade arrange-
ment.

2. Application of the theory of customs unions to the
EEC-AAS trade agreement

 

_The EEC-AAS trade arrangement constitutes a preferential
area between a group of developed countries and a group of

deve10ping countries. It is far from constituting a free trade

 

EEEE because the AAS did not abolish tariffs on imports from
the EEC. Rather, it imposed lower tariffs on EEC imports than
on imports from other countries, and also abolished quantitative
restrictions on the former imports while maintaining them on the
latter.

The potential impact of tariff preferences on AAS ex-
ports to the EEC will be considered only for manufactured ex-
ports. Two cases will be discussed: (1) exports of standardized

commodities, and (ii) exports of differentiated commodities.

14

Impact of tariff preferences on AAS exports of

standardized manufactures

As will be shown in chapter III, standardized commodi-
ties exported by the AAS to the EEC consist largely of pro-
cessed agricultural products and processed raw materials which
are not produced in the EEC countries. Furthermore, few sub-
stitutes for these commodities are produced in the EEC. Under
these circumstances, increased AAS exports to the EEC due to
tariff preferences should yield relatively greater trade diver-
sion than trade creation. The same outcome should also apply
to EEC exports to the AAS. Thus, for standardized commodities,
the main impact of tariff preferences should be trade diversion.
In particular, this impact should be caused by a shift of EEC
imports from Latin American and British Commonwealth Countries
to the AAS. The size of the impact will be a function of the
elasticity of supply of exports from the AAS and non—beneficiary
countries, the elasticity of demand of imports by the EEC, the
height of the tariff imposed on imports from non-beneficiary
countries, and the difference in productive efficiency between
the AAS and the latter countries.

First, let us consider the case where the world supply
of a commodity is perfectly elastic, and the AAS supplies a
small fraction of the total EEC imports - an assumption which
applies to many commodities exported by the AAS to the EEC.
In this case, various developments could take place.

In the short—term, the AAS may not be able to increase

supply, eSpecially if an increase in supply means a larger

15

agricultural production. It will therefore simply increase
its export prices by a percentage equal to the height of the
tariff. In other words, the AAS will improve its terms of
trade vis-a-vis the EEC. This outcome will not take place,
however, if importers in the EEC countries have a monopoly

1, and thus appropriate most of

power on imports from the AAS
the value of the tariff in place of the AAS producers.

In the longer term, and if the AAS producers do bene—
fit from the tariff preferences, supply of exports from the
AAS could be increased as a result of the higher prices AAS
producers receive for their exports. If increase in produc-
tion takes place at constant cost, increased diversion will
occur in favour of the AAS. The only constraint to this
diversion will be the productive capacity of the AAS. However,
if the increase in production takes place under increasing
costs, the competitive edge afforded the AAS by the tariff
preference will be lowered. AAS exporters will increase ex-
ports up to the point where profits are maximized. In general,
this will occur before the tariff preference is fully absorbed
by the increase in production costs.

Another case treated by Johnson (4, pp. 189—190) may

also apply to the AAS. This is the case where the supply

 

firms in the AAS are subsidiaries of importing firms in the—
EEC.

16

capacity of the AAS is large relative to the import demand of
the EEC. As will be shown in chapter III, this case may apply
to four or five commodities exported by the AAS to the EEC.
Under conditions of perfect competition the AAS must, in this
situation, sell its products to the EEC at the world market
price. The only impact of trade preferences will be an in-
crease of the quantity of AAS exports to the EEC. The AAS
terms of trade with the EEC will not, in this case, improve.
However, if the condition of perfect competition is dropped,
the AAS may institute a system of price discrimination whereby
export prices to the EEC may be set higher than export prices
to the rest of the world, but lower or equal to the price tha
EEC importers would have to pay in the absence of tariff pre—
ferences. Being in a monopoly position, the AAS could adjust
the export price to the EEC so as to maximize profits. In
this case, exports to the EEC could be increased, and the AAS
may in addition benefit from an improvement in its terms of
trade with the EEC.

A third situation which may be considered is one
whereby EEC imports originate from a very small number of
major exporters, some of these being AAS countries. In this
case, AAS exporters could adjust their export prices to the
EEC so as to maximize profits. This price adjustment would
normally be accompanied by a shift of EEC imports from non-
beneficiary countries to the AAS. However, this outcome will

take place only under the condition where the non-beneficiary

l7

suppliers to the EEC are unable to lower their export prices in
order to maintain their market share in the EEC. These suppli-
ers' export prices would have to be lowered by the full amount
of the tariff imposed on their exports to the EEC, or by a
fraction of the tariff if the AAS were to increase its export
prices to the EEC. The capacity of non-beneficiary suppliers
to offset the advantage afforded the AAS by tariff preferences
will be a function of two factors: the height of the tariff
and the production cost differential between AAS suppliers and
non-beneficiary suppliers. In general, the lower the tariff
and the higher the difference in production costs between AAS
suppliers and non-beneficiary suppliers, the easier it will
be for these latter suppliers to offset the AAS advantage. In
any case, total exports to the EEC should increase. Depending
on the circumstances, this increase will be supplied solely by
the AAS or, proportionately, by all EEC suppliers.

Impact of tariff preferences on AAS exports

Bfidifferentiated commodities

Differentiated commodities produced by countries
within the AAS include a variety of goods from SITC Groups 0,
l, 5, and 8 (see chapter III). While the EEC countries pro-
duce few of the standardized commodities described earlier, or
few substitutes for these commodities, they do produce a large
number of the above differentiated commodities. Thus, the
granting of tariff preferences could lead, in the case of these

commodities, to both trade creation and trade diversion. For

18

example, an increase of EEO imports of shoes from the AAS due

to the tariff preferences could lead to a decrease of domestic
production in the EEC countries, as well as a decrease of im-

ports from non-beneficiary countries.

The analysis used for standardized commodities can
also be applied to differentiated commodities, and would
yield similar conclusions. Two variables should, however,
play a greater role in the case of differentiated commodities.
These variables are (i) the elasticity of substitution of
commodities exported by the AAS for those produced in the EEC,
and (ii) the elasticity of substitution of commodities ex—
ported to the EEC by the AAS for those exported to the EEC by
non-beneficiary countries. In general, the higher the value
of these elasticities of substitution, the greater will be the
scope for trade diversion and trade creation.

It should also be noted that differentiated commodities
include a larger proportion of consumer goods than do standard-
ized commodities. Since tariff duties tend to be relatively
high for consumer goods, the potential impact of tariff pre-
ferences should be relatively greater for differentiated
commodities than for standardized ones. It is not, however,
certain that the AAS will actually take advantage of these
preferences by substantially increasing its export of the
differentiated commodities to the EEC. A number of factors
which may constrain such an expansion will be reviewed in

section 4 of this chapter.

l9
3. Methodology for the evaluation of the AAS
response to EEC tariff—preferences

There exists a large number of methods for evaluating
the impact of various forms of economic integration. Some of
these methods were reviewed by Kreinin (5) who concluded that
"each approach... is fraught with dangers arising from its
own heroic assumptions", and "the only hope of arriving at
approximate orders of magnitude lies in utilising a variety of
methods and comparing the results" (5, p.900).

In this section, various evaluation methods will be
reviewed, including the one selected for this study. These
methods were developed in order either to predict the impact
of a potential or recently establiShed union, or to determine
whether a union has had the expected impact. Given the
purpose of this study, we will be concerned primarily with

the latter type of evaluation methods.

Comparing hypothetical estimates to actual values

One method of determining whether tariff preferences
have had the expected impact is to compare hypothetical
estimates of what trade flows, growth rates, market shares,
etc. would have been in the absence of preferences to the
actual values of these parameters. If the hypothetical
estimates are approximately equal to the actual values, it
may be concluded that the tariff preferences have had the
expected impact. Various hypothetical estimates have been

used to determine whether a union has had the expected impact.

20

One approach, reviewed by Kreinin (5) and used by the
EFTA secretariat (6), among others, compared the hypothetical
share of imports in total consumption to the actual share for
the year 1965. The hypothetical share was estimated on the
basis of a projection of pre-integration trends during the
1954-59 period. The EFTA secretariat thus obtained estimates
of trade creation and trade diversion for major commodity
groups. In this approach, it is assumed that the share of
imports in total consumption would have developed over the
1959—65 period in the same manner as during the 1954-59 period.

Similar approaches have been used, ranging from simple
extrapolations of pre-integration growth rates of export values
or market shares to extrapolation of a world trade matrix.
These approaches suffer from many weaknesses. First, in most
cases they do not take into consideration new developments in
world trade, such as other trade agreements, or trade negocia-
tions yielding global tariff cuts. Second, income and price
movements are not usually taken into consideration. Third,
changes in the countries' competitive position which affect
the ratio of import to domestic prices are usually neglected.
Despite these weaknesses, the above approaches can be useful
if applied in conjunction with other approaches.

Comparing pre-integration to ppst-integration

estimates of various parameters

One evaluation method adopted by Balassa (7) compares

the post-integration income elasticity of demand for imports

21

to the pre-integration income elasticity. Let us define:

MI = intra-union imports

I3
I!

E extra—union imports
Y = sum of the GNPs of member countries.

The income elasticity of demand for intra-union
imports, a , and that for extra-union imports, B , may then
be estimated from the following equations:

Log MI = a + aLog Y
and Log ME = b + BLog Y

Trade creation is indicated by an increase in the
value of a.from a period preceeding the union to a period
following the union. On the other hand, trade diversion is
indicated by a decrease in the value of B between the same
two periods.

‘ There are two main criticisms of this evaluation
method. First, it assumes that income elasticities of import
demand would have remained unchanged in the absence of the
union. Second, it neglects changes in supply conditions
(e.g., changes in competitiveness).

A second evaluation method, used by Kreinin (5) to
evaluate trade creation and trade diversion resulting from
the establishment of the EEC, compares the pre-integration
import/consumption ratio to the post-integration ratio. The
change in the "total imports/consumption" ratio measures
trade creation while the change in the "extra-EEC imports/

consumption" ratio measures trade diversion. Kreinin

22

recqgnzeS‘mxn;factors other than economic integration could
affect the import/consumption ratios (e.g., income and price
movements). In order to isolate the integration effect, it

is necessary to know what changes in the ratios would have
occured in the absence of integration. Kreinin uses for this
purpose a "control group" or "normaliser" approach. He assumes
that "the factors affecting the import/consumption ratios moved
over the period of the 19608 in an identical fashion in both
the EEC and in the 'control' countries, and that the reaction
of the economy to these changes was the same in both markets".
(5, p.902). Consequently, import/consumption ratios may be
adjusted for non-integration effects.

Kreinin's method is an improvement over other methods
in the sense that it isolates the integration effects from
other effects. Its application may, however, present several
problems. First, it may be difficult to find the proper
"control group" or "normaliser". Second, the statistical data
which are required (e.g., output, wholesale prices) are not

usually available for many developing countries.

Use of multiple regression analysis

Another evaluation method makes use of multiple re-
gression'analysis with trade preferences being one of the in-
dependent variables. Aitken and Obutelewicz (8) applied such
a method to evaluate the respective impacts of EEC tariff pre-
ferences on AAS exports, and of AAS tariff preferences on BBC

exports. Two different regression equations were fitted to

23

estimate these two impacts. Given the relevance of the Aitken—
Obutelewicz paper to this study, their evaluation method is
reviewed in some detail.

The two regression equations are shown below:

LXad = b'O + blLDad + b2LYa + bBLYd
BC EEC FTM
+ b4LPad + b5LPad + b6LPad
FA ,
+ b7LPad + e ad ....................... (l)
LXda = b'O + blLDda + b2LYa + bBLYd
CB AAC TMF
+ b4LPda + b5LPda + b6LPda
+ b LP AF + b LA - + e' (2)
7 da 8 da da ..............
where:

2 Log

L
- X = value of exports, in dollars

I
m
H

African country

— d 2 developed country
a and d stand for the exporting country
when used as the first subscript, and for
the importing country when used as the
second subscript.

- D 2 distance between the importing and

exporting country

- Y = GNP

BC P EEC
ad ’ ad ’

variables corresponding, respectively,

FTM FA

- P Pad , and Pad are dummy

to the British preference for the exports

24

of African Commonwealth countries, the
preference of the EEC countries other
than France for the exports of the AAS,
the French preference for Tunisian-
Moroccan exports, and the French pre-
ference for the export of AAS countries
which are former colonies of France.

- In equation (2) the dummy preference variables re-
present the African country's preference for the
exports of the respective developed countries with
which the African country has a trade preference
agreement.

- A dollar value of aid from d to a.

da
- e'ad’ e'da 2 error terms.

The authors estimated the two regression equations for each
year of the l958-l97l period. They considered that tariff
preferences had an impact on AAS exports to the EEC if the

PEEC

coefficient increased in value from 1958 to 1971, and
was statistically non-significant at the beginning of the
period but significant in the latter part of the period. The
same type of results for the coefficient PAAC was considered
to show that the AAS preferences had an impact on EEC exports
to the AAS.

The above evaluation method should, theorwatically,

isolate the effect of tariff preferences on exports from the

preference—receiving country. However, its application by

25

Aitken and Obutelewicz calls for two main criticisms. First,
with respect to equation 1 (impact of EEC preferences on AAS
exports), the authors should not have restricted the analysis
to African exporters only. It is possible that AAS exports

to the EEC compete primarily with exports from non-African
countries, either developed or developing. If this were the
case, PEEC would reflect both the impact of tariff preferences
and the increased competitiveness of the AAS. It may not,
therefore, be‘concluded that a large and significant PEEC
coefficient at the end of the period proves that the tariff
preferences had an impact.

Second, the authors should have applied their evaluation
method to individual commodities rather than to global exports.
Let us suppose that a large proportiOn of commoditiesexported
by the AAS to the EEC differ in type and quality from those
exported by non-AAS African countries (e.g., Tunisia and
Morocco). Let us further suppose that a shift in EEC demand
in favor of AAS products occurs for reasons unrelated to tariff
preferences (e.g., as the result of a change in taste, a sub-
stitution for other products, etc.). Under these circumstances
the EEC would increase its imports relatively more from the AAS
than from non-AAS African countries. Thus, it could not be

concluded that a high and significant PEEC

coefficient proves
that tariff preferences had the expected impact. If the evalua-
tion method were applied to individual commodities, especially

standardized commodities, the possibility of changes in demand

26
due to factors other than tariff preferences would be greatly

reduced.

Evaluation method used in this study

The evaluation methodadopted in this study is a
modified version of that used by Young (9). The method can
be reliably used in order to determine whether the AAS res-
ponded positively to the EEC tariff preferences.‘ It may also
be used, with less reliability, in order to estimate the net
impact of tariff preferences. It may not, however, be used to
obtain separate estimates of trade diversion and trade creation
for the purpose of estimating the net welfare impact of these”
preferences. The method will now be described in general terms,
and it will then be shown how it could be applied to the EEC-
AAS trade arrangement. I

(The rate of growth of exports from one country to any
other country is a function of the following four factors:
(1) the growth of effective demand for imports in theimporting
country, (ii) the ability of the exporting country to increase
its exports so as to meet this growth of demand, and its ability
to produce these exports at competitive prices, (iii) the ‘
granting of tariff preferences in favor of the exporting country,
and (iv) the establishment of a special relationship between
the exporting and importing country, giving rise to special
privileges in favor of the two countries (e.g., monetary
arrangement, foreign investment privileges). V

The purpose of the evaluation is to isolate the impact

of factor (iii). This may be achieved by using two groups of

27

"control" countries in addition to the groups of preference-
receiving and preferenceegiving countries.

Let us define X1 and X2 as, respectively, the group
of exporters which was granted tariff preferences and that
which was not. Let us also define I1 and 12 as, respectively,
the group of importing countries which granted the tariff pre-
ferences to X

l’ and that whiCh did not. Let us finally define

G as the rate of growth of exports from country i to country

13
j during the period since the preferences were granted. We

may then set out the following table of growth rates:

3.1. 3.2.
3_ G11 G12
X2 G21 G22

A high value cf Gll may be due to factor (i) - i.e., a high
rate of growth of imports by Il due to increased demand, or to
factors (iii) - tariff preferences - and (iv) - special privi-
leges. Let us assume, for the time being, that no special
privileges were established during the period of analysis.
Then, in order to isolate the effect of tariff preferences,
we first consider the difference between G11 and G21. A dif-
ference greater than zero could be an indication that the
tariff preferences had the expected positive impact on Xl
exports to I1 since changes in Il demand should affect exports
from X1 and X2 equally.

A difference of growth rates greater than zero does

not, however, indicate by itself a positive impact of tariff

28

preferences. A higher growth rate of exports from Xl could
also be an indication of an increased over-all competitiveness
of X1 exporters (i.e., a higher growth rate could be attributed
to factor (ii)). An indication of this increased competitive-
ness may be provided by a comparison of 012 to G22. If the
difference between G12 and 022 is greater than zero, it may be
concluded that the X1 countries have become, over-all, more
competitive.

If the higher growth rate of exports from X1 countries
to I1 countries were due only to the greater competitiveness
of the former countries, then the following relationship should
hold:

(G

- G - (G

11 21) 12 ‘ G22) = 0
If the above difference is greater than zero, it can be con-
cluded that the tariff preferences had an impact on X1 exports
to 11'

If it is known that special privileges were established
during the period of analysis, it is not possible to conclude
that tariff preferences had an impact (i.e., the difference
between G11 and G21 may be due to special privileges only).
Under these circumstances, the analysis becomes extremely
complicated since it is difficult to quantify the effect of
special privileges on X1 exports to 11'

Let us now show how the above evaluation method could

be applied to the EEC-AAS trade arrangement.

First, it may be noted that special privileges and ties

29

do exist between the AAS and France. However, these privileges
were already established prior to the association arrangement,
and have not changed substantially since then. The effect of
special privileges may therefore be considered to have remained
constant over the period of analysis.1

Next, this method of evaluation requires that a choice
be made of "control" countries. Three alternative choices of
X2 countries could be of potential interest: (1) the world, ex-
cluding the EEC2 and AAS countries, (ii) all other developing
countries, and (iii) a group of developing countries similar in
many characteristics (e.g., per capita income, size) to the AAS
countries.

The choice among the above alternatives will depend on
the purpose of the analysis. In the present study, the purpose
is to determine whether the AAS has displaced its principal
competitors on the EEC market as a result of tariff preferences.
AAS exports to the EEC compete primarily with exports from other
African countries and from countries in Latin America and Asia.
The group of X2 countries adopted for this study therefore in-

cludes all developing countries, with the exclusion of the AAS.

 

1It is possible that de-colonisation has reduced the
importance of these privileges. However, since the period of
analysis (1962-69) directly follows the granting of independence,
the impact of decolonisation may be considered constant over the
period of analysis.

2The EEC countries are excluded because they benefit
from the same tariff preferences.

30

The choice of I2 countries is much easier to make. As
will be shown in chapter III, the bulk of AAS exports goes to
AAS countries and to industrialized countries outside the
socialist bloc. The group of Izcountries adopted for this
study thus includes industrialized countries other than the
EEC and socialist countries.

Having chosen the "control" countries, it must be de-
cided whether the evaluation method should be applied to indi-
vidual commodities or to over-all exports or groups of exports.
The application of the method to individual commodities (e.g.,
commodities at the 5 digit SITC level) seems preferable for
two reasons. First, as stated earlier with respect to the
Aitken-Obutelewicz study (8), application of the method to
over-all exports could yield invalid conclusions. Second, it
is of interest to find out if the size of the impact is function

of the height of the tariff imposed on non-beneficiary countries.

4. Trade theories and tariff preferences

As indicated in section 1, the impact of tariff pre-
ferences on exports from the preference-receiving country de-
pends in part on the height of the tariff prevailing prior to
the establishment of the union and on production cost dif-
ferentials between the preference—receiving country and both
the preference-giving and the non-beneficiary countries. Cost
differentials reflect the comparative advantage or disadvantage
that the preference-receiving country has vis-a-vis the other

countries in the production and export of particular commodities.

31

For trade creation to take place with respect to a given
commodity, the preference-receiving country must enjoy a com-
parative advantage vis-a-vis the preference-giving country.
On the other hand, for trade diVersion to take place, the
preference-receiving country must have both a comparative
advantage vis-a-vis the preference-giving country and a
comparative disadvantage vis-a-vis non-beneficiary countries.

In order to demonstrate that a country has responded
as fully as possible to tariff preferences, it must be shown
that: (1) its exports of manufactures to the preference-giving
country have grown at the expected rate, and (ii) all commodi-
ties for which it has a comparative advantage vis-a—vis the
preference-giving country were in fact exported to this country.
The evaluation method described in section 3 of this chapter
fulfills only condition (1). In order to fulfill condition
(ii), a list of commodities for which the preference-receiving
country enjoys a comparative advantage should be established
and compared to the list of commodities actually exported to
the preference-giving country. Such a comparison will indicate
the extent to which the preference-receiving country has taken
advantage of tariff preferences.

To establish the above list, one must apply current
comparative advantage theories with respect to the preference-
receiving and preference-giving countries.

There are currently six main comparative advantage

52

theoriesl, ranging from the path-breaking Heckscher-Ohlin (H-O)
theory of international trade to more recent theories such as
the technological gap and the product cycle theories. Theore-
tically, the application of these different theories to a given
country would yield different lists of commodities for which
the country enjoys a comparative advantage. However, as shown
by Hufbauer (10), there is a great deal of overlap among these
theories, and there should therefore be considerable overlap
among the commodity lists resulting from the application of

the theories.

It is outside the scope of this study to identify the
commodities for which the AAS enjoys a comparative advantage
vis-a-vis the EEC. We must therefore rely on completed studies
relating to developing countries, the assumption being that
conclusions derived from these studies apply at least in part
to the AAS. A study by Lary (11) adopted the approach advocated
by Kenen (12) who combines the H-0 factor proportions theory
and the human skills theory into a "new" factor proportions
theory of trade. In this approach, "value added per employee"
is used as a guide to factor intensity in manufacturing. The,
higher the value added per employee, the more capital—intensive
is the industry; the lower the value added, the more labour-
intensive it is. Using this approach, Lary established a long

list of labour-intensive commodities of potential interest to

g

V

1For a review of these theories, see Hufbauer (10).

33

less developed countries. This list was used in this study to
identify the commodities for which the AAS is likely to enjoy
a comparative advantage vis-a-vis the EEC.

Obviously, it is not necessary that the AAS export all
commodities in the above list to the EEC in order to conclude
that there was a positive response to tariff preferences. Given
the economic size of the AAS countries, they can specialize in
the production and export of only some of these commodities.

It was therefore decided to select from the list only those
commodities produced in the AAS during the period under review.
This shortened list was compared to the list of commodities
actually exported to the EEC in order to assess the extent to

which the AAS responded to the tariff preferences.

Potential constraints on the AAS response

The fact that a country enjoys a comparative advantage
in the production and export of a number of commodities does
not imply that these commodities will actually be exported. A
potential comparative advantage may not be exploited for a
number of reasons. One important reason, which does not apply
to AAS exports to the EEC, is the existence of various barriers
to trade (e.g., quotas, prohibitive tariffs) imposed by the
importing country. Other reasons include high transport costs,
adoption of industrial policies focusing on import substitution
rather than export expansion, etc.

Two main factors may constrain AAS attempts to respond

to EEC tariff preferences by producing and exporting to the EEC

34

commodities for which n enjoysa potential comparative
advantage.

First, although the AAS may enjoy a potential compara-
tive advantage in the production and export of labour-intensive,
low-skill intensive Commodities, it may not have the capacity
to produce and export these commodities to the EEC. Industrial
production, however labour-intensive or low-skill intensive,
requires a minimal level of technological know-how, quality
control, and skilled labour in order to yield goods of the
type and quality required by EEC consumers. Furthermore, the
marketing of these goods requires good managerial skills and
the existence of good marketing channels. (The lack of indi-
genous technical and managerial skills, or the lack of foreign
investors who provide such skills, may impede the growth of
AAS exports to the EEC.

Second, with respect to finished goods, tariff pre-
ferences may have little impact on trade expansion from the
AAS if, as hypothesized by Brown (13),.product differentiation
is more important than price differentiation as a determinant
of the pattern of trade. Brown's hypothesis is that the com-
petitiveness of an exporter is less a function of the price of
the exported commodity than of the particular characteristics
of the commodity which differentiate it from other commodities
of the same type. Thus, product differentiation provides the
exporter with a competitive edge which is not offset by existing

jprice differentials. Brown's hypothesis is supported by several

35

instances of empirical evidence. For example, the creation

of the EEC did not make the export structure of individual

EEC countries more skewed. Instead, each country is exporting
and importing more, but trade is more heavily intra-industry
than inter-industry (14). In a similar vein, the Swedish
textile industry has been rejuvenated by specializing in
luxury textile products instead of competing against exports
from LDCs.

An implication of comparative advantage theories is
that the more dissimilar the economic structures of two
countries, the greater should be the scope for mutually bene-
ficial trade. Thus, the lowering or removal of trade barriers
between a group of developed countries and a_group of developing
countries should result in a substantial expansion of trade be-
tween these two groups. This expansion may not, however, take
place if, as hypothesized by Linder (15), trade is likely to
be more intense between countries with similar rather than dis—
similar economic structures. If the Linder theory is valid,
the impact of trade preferences granted to developing countries
should be minimal. It is therefore of interest to determine
whether the AAS export pattern supports the Linder theory. The

following section will describe this theory in some detail.

5. The Linder similarity of preferences theory and
tariffpreferences1

In his Essay on Trade and Transformation (15), Linder

distinguishes between trade in primary products and trade in

-. 1The material in this section refers to chapter III of
iJlnder's book (15).

36

manufactures. While he accepts the orthodox theories of the
determinants of trade patterns (e.g., the H-0 theory, the
product cycle theory) for the former products, he rejects

these theories for the latter products. Linder's theory
contrasts sharply with other theories of the commodity composi-
tion of trade because it predicts that trade in manufactures
should be most intense between countries with similar, rather

than dissimilar, economic structures.

The Linder theory

According to Linder, the qualities and types of manu-
factured commodities consumed by a country are characteristic
of its level of development and industrial structure. Thus,
manufactures must be of a type and quality which cater to the
needs and tastes of the local p0pulation. This proposition
regarding the effects of demand is fairly obvious and does not
need further elaboration.

. The Linder approach differs from other approaches
primarily with respect to the supply side. According to the
orthodox trade theories, comparative advantage, however defined,
determines what is produced in a country. Thus, commOdities
may be produced for the foreign markets even though there is no
ciemand for such goods in the home market. In other words, home
demand does not necessarily dictate the pattern of local pro—
duction. Furthermore, since countries which differ in terms
Of comparative advantage also differ in terns of their economic

Structures, trade should be most intense between countries

37

which have differing economic structures.

When analysing the factors which determine home pro—
duction, Linder reaches different conclusions. According to
Linder, entrepreneurs involved in the production of manu-
factured goods tend to respond first to the needs of the
society in which they live. Four main reasons are offered to
explain this behavior. First, when faced with uncertainty,
producers are more likely to respond to the profit opportunities
of which they have the greatest knowledge. These opportunities
inevitably appear in the home market. Second, local needs are
the most likely to stimulate technological change and innova-
tion, the result being that newly developed products are usually
unsuited to foreign needs and tastes. Third, the production
and marketing of manufactured goods, especially differentiated
commodities, generally requires development work that must be
carried out in close contact with the market. Crucial informa-
tion must be readily available to the entrepreneur during the
trial-and-error period that occurs during the early stages of
production. Finally, psychological as well as material factors
(e.g., administrative problems, high marketing costs, distance)
:further explain why an entrepreneur tends to learn about profit
capportunities in the domestic market before investigating
‘profit opportunities abroad.

In summary, it can be said that the production functions
0f manufactures demanded at home tend to be viewed as relatively

more advantageous by the entrepreneurs. Consequently, domestic

38

production tends to reflect domestic demand, and only in
special cases does it also reflect factors underlying orthodox
comparative advantage theories.

As the home market becomes more and more saturated,
leaving little room for further expansion, entrepreneurs
begin investigating the possibilities of exporting their
‘products to foreign markets. Linder sees international trade
as "nothing but an extension across national frontiers of a
country's own web of economic activity" (15, p.88). Production
being originally geared to the needs and tastes of the home
country, exported goods will have to satisfy similar needs and
tastes. Since the types and qualities of goods consumed
locally are in part a function of the stage of development of
a country, as expressed, for example, by its per capita income
(PCI) level, manufactured goods produced in one country are in
demand primarily in countries at a similar stage of development
(i.e., with a similar PCI level). Thus, according to the
Linder theory, export and import "baskets" for a given country
should be highly similar. Furthermore, the more similar the
economic structures and levels of development of two countries,
'the more similar one country's composition of exports will be
to the other country's composition of imports. Why does trade
take place between two countries when their production and
consumption patterns are highly similar? Linder answers this
question by hypothesizing that mutually beneficial trade stems

from the marginal satisfaction which differentiated products

39

bring to consumers in the two trading countries. Linder
States, in particular, that "the almost unlimited scope for
product differentiation - real or adVertised - could, in
combination with the seemingly unrestricted buyer idiosyn-
crasies, make possible flourishing trade in what is virtually
the same commodity" (15, p.102). As an illustration of this
argument, "buyer idiosyncrasy" could explain why Italy exports
Fiats to Germany and imports German VWs. I
Linder recognizes that while similarity in economic
structures constitutes a trade-creating force, actual trade
may not reach its full potential between two countries with
similar structures due to various trade-breading forces.
Distance between countries is the main trade—breaking force
advanced by Linder. Distance is considered a proxy for both
transport costs and limited trade horizons (i.e., entrepreneurs
are less likely to be aware of market opportunities in far—
away countries than in neighboring countries). Other trade-
breaking forces include cultural dissimilarity (e. g. , different
languages), political strains, various types of trade barriers.
Given the importance accorded by Linder to product
differentiation as a factor generating trade, the theory should
apply better to highly differentiated products (finished con-
finnner goods) than to standardized semimanufactures. Balassa
(16) provides support for the Linder theory but suggests that
its validity is restricted to finished manufactures. He pro-

Poses that, for small countries, orthodox trade theories

4o

(e.g., the H-0 theory) explain the export pattern of semi-
manufactures. The reason is that these products do not gene-
rally require the availability of a home market. He illus-
trates this point by stating that there are "numberous examp-
les, in countries as diverse as Belgium, Hong Kong, and Por-
tugal where domestic consumption plays only a supplementary
role as market outlet for standardized manufactures." (l6,p.203)
Like any other theory, the Linder similarity of pre-
ferences theory applies under a number of assumptions. The
most important assumption is that exports are produced by
local entrepreneurs rather than by foreign investors. If this
is not the case, a country may produce commodities which do
not reflect the home demand, and may export these commodities
to countries with dissimilar economic and industrial structures.
This outcome can be explained by the fact that foreign investors
are responding to demand conditions in theirigwn home country,
but are producing abroad in order to take advantage of some
relatively low priced factor of production (e.g., low wages).
Such production usually takes place only after development work
in the investors' home country has been completed. Very often,
production consists solely in plant assembly of components
produced in the investors' country. It may therefore be con-
cluded that the larger the number of foreign investors in a
country, the more its export pattern will tend to differ from
the one predicted by the Linder theory. In other words, the
existence of foreign investments tends to invalidate the Linder

theory in the same way as it invalidates the factor proportions

41

theory. In the former case, these investments make it possible
for a country to gear its production to satisfy foreign demand,
while in the latter case, they make it possible for a country
to produce and export commodities for which the country does
not enjoy a comparative advantage (e.g., capital-intensive
mining products, steel, chemicals). Linder reaches a similar
conclusion with respect to trade in raw materials, stating:

"If entrepreneurship could not move internationally, it is
quite possible that our proposition could be applied to trade

in primary products as well as manufactures" (15, p.93).

Previous tests of the Linder theory

Various trade models based on the Linder theory have
been tested in recent years.

The first test of the theory_was performed by Linder
himself. He attempted to show that the average propensity to
import (API) of one country from another is negatively correla-
ted with the absolute value of the difference in per capita
income between the two countries (IAPCID. Using a sample of
32 developing and developed countries, he plotted for each
country the (API) values against the (IAPCID values. To support
the theory, the graph should resemble an inverted V centered
around the per capita income level of the exporting country.
.Alimbugh the results of this test were not conclusive, they did
Suggest that the thesis isworthy of further consideration.

Hufbauer (10) attempted to test two different inter—

Pretations of the Linder theory. The first interpretation is

42

that similarity should exist between actual commodities im—
ported and exported by a given nation. The second interpreta-
tion is that similarity should exist between commodity charac-
teristics embodied in a nation's imports and exports.

To test the first interpretation of the theory,
Hufbauer used, as a measure of trade similarity, the following

index:

 

 

CosX.M. = n
l J
' 2 2
V 4;: Xin ' gmjn
where:
xin = exports of commodity n as a percentage
of total manufactured exports from
country i
mjn = imports of commodity n as a percentage

of total manufactured imports by
country j
When CoinMj equals one, there is complete identity between
exports and imports. On the other, a value of zero means
complete dissimilarity.
Values of CoinMj derived from the 1965 three-digit
‘trade statistics were then used in the following two regression

e quations:

CoinM.

3 C1 + alAj + blAi when Ang.

l
and

CoinMj 2 c2 + a2Aj + b2Ai when AJ.>,Ai

43

A. A. = per capita GDP of country i and
country j, respectively

The Linder theory implies that 01 should be smaller than 02,
that al should be positive, and a2 should be negative.1 The
converse results would, on the other hand, support the orthodox
comparative advantage theories.

Results from the above analysis supported neither
Linder nor the orthodox trade theories. Both a1 and a2 were
found to be positive, with approximately the same small value.
Furthermore, it was found that the richer the trading partners,
the more similar are imports and exports (i.e., the higher the
value of CoinMj). Hufbauer concluded: "Broadly speaking,
these findings represent nothing more than the diversification
of exports and imports which accompanies greater affluence.
Owing to concentration, especially export concentration, the
opposing trade vectors of two poor countries, say Hong Kong
and Portugal, will substantially differ, while thanks to diver-
sification the import-export vectors of two rich nations, for
example the United Kingdom and Sweden, will roughly coincide."
(10, p.201).

Hufbauer also tested the second interpretation of the
ILinder theory, i.e., similarity of characteristics embodied in

inmports and exports. This was done by ranking national imports

‘

lHufbauer included the variable A. in the two regression
éflluations because the value of the cosine should also be a func-
tlxon of the economic level of the exporting country.

44

and exports separately with respect to commodity characteristics.
A positive correlation between the two rank lists would support
the Linder theory, while a negative correlation would support
the orthodox theories. The Spearman correlations were all
found to be negative, but not significantly different from
zero at the 1% level. On the basis of the results of the above
two tests, Hufbauer concluded: "The comparatively weak showing
of Heckscher-Ohlin might be interpreted either as a modest
triumph for Linder or as the inevitable outcome of restrictive
tariffs and quotas. The major point, though, is that trade
gggg involve some exchange of characteristics?(10, p.207).
Fortune (17) tested the Linder theory for finished
manufactures in S.I.T.C. groups 7 and 8. He used the following

regression equation:

Mij/Yi = a + b-le / Nj - Yi / Nil + cDij
where:
Mij = imports of country i from country j
Yj’Yi = GNP of countries j and i, in dollars
Nj’Ni = population of countries j and i
Dij = distance between countries 1 and j, in miles

The regression equation was estimated for each of 25
exporting countries. The total number of observations for
each regression was equal to 50 importing countries.

Negative and significant values for the coefficients
‘b and c would support the Linder theory. However, the findings
Iprvided only limited support for the theory. The coefficient

(If the distance variable was negative in all cases, and was

45

significant at the 5% level in eleven of the 25 regressions.
The difference in per capita income variable was significant
at the 5% level in 9 cases, with the coefficient being
negative in 7 out of the 9 cases (It was negative, but non-
significant in 11 other cases.) Fortune concluded: "Never-
theless, the results do give some support to the Linder
hypothesis concerning similarities in income levels as a
prerequisite for trade in finished manufactures. However,
the low coefficients of determination for all the regressions
show that even in those cases where the per capita income
variable is a significant determinant of trade intensities,
it is hardly the only one." (17, p.317).

The validity of the Linder theory cannot be assessed
unless the theory is properly tested. The fact that the

above tests failed to support the theory or provided only
limited support, could be due to a too strict interpretation
of the theory, or to inadequate testing methodology.

Since it is probable that the theory applies best to
differentiated manufactures, testing of the theory should be
performed separately for standardized commodities and dif-
ferentiated commodities. Unfortunately, Linder tested his
theory for global exports, including both manufactured and
primary commodities, while Hufbauer and Fortune tested the
theory for global manufactures, including both standardized
and differentiated commodities.

Second, both Linder and Fortune failed to take into

46

consideration the size of the exporting country. Exports
from a large country A to a country C should be larger than
those of a small country B to the same country C, even though
per capita incomes in countries A and B are equal. Thus, im-
port intensity should be partly a function of the size of

the exporting country (e.g., GNP of the exporting country).

Third, income distribution should also affect a
country's average propensity to import. If two countries
have different income distributions, their average propensi-
ties to import from a third country may differ even though
their per capita incomes are equal. This is recognized by
Linder who Stated that median rather than average income
levels would better reflect the demand structure for imports
whenever "the distribution of income within the countries is
very uneven." (15, p.113). Thus, the fact that the above
tests did not support the theory could be explained by the
use of average rather than median PCIs.

Fourth, as suggested by Fortune, difference in per
capita income may not constitute the only major independent
variable. His test of Linder included a second variable,
distance. There may be other major trade-breaking and trade-
creating variables. For example, political relations between
two countries could either reinforce or impede trade while
cultural similarity should favor trade. The validity of the
Inain Linder explanatory variable ( IAPCID can only be
zissessed if it is included in a model which incorporates

I

all relevant variables.

47

To summarize, proper testing of the Linder theory re-
quires that the theory be tested separately for differentiated
and standardized commodities, that median rather than average
income levels be used, that the size of the exporting countries
be taken into consideration, and that all relevant independent
variables be included. We propose that the test be performed

for a Linder-based model of the following general form:

(API)ij f( IAPCIMlij, Yj, Dij, ij, ij, Tij)

where:

H

(API)ij average propensity to import of country i

from country j.

IAPCIMIij = absolute difference in median per capita
income between countries 1 and j.

Y. = GNP of exporting country j.

J
Dij = distance between countries i and j.
Pij = binary variable for political ties between
countries 1 and j.
Cij = binary variable for cultural similarity.
Tij = binary variable for trade agreement.

In chapter IV, a modified version of the above model
will be tested in order to determine whether the AAS export

pattern supports the Linder theory.

Linder theory and tariff preferences
From the perspective of the Linder theory, tariff pre-
.ferences should have a limited impact on exports of manufactured

(mammodities, especially differentiated manufactures, from the

4s

preferenCe-receiving country. Since, according to the
theory, potential comparative advantage does not determine
potential trade intensity between countries with dissimilar
economic structures, the removal of trade barriers should

not have a significant impact on actual trade intensities
between these countries. On the other hand, if trade
barriers exist between countries with similar economic
structures, they may constitute a major trade-breaking

force, and their removal should permit actual trade intensity
to coincide with potential trade intensity.

In the case of the EEC-AAS trade arrangement, the
theory predicts that the tariff preferences should have
virtually no impact on AAS exports of differentiated com-
modities to the EEC, and only a limited impact on exports
of standardized commodities. These two propositions will

be tested in chapter IV.

10.

ll.

12.

49

LIST OF REFERENCES

. J. Viner, The Customs Union Issue, New York: Carnegie

Endowment fbrflnternationaldPeace, 1950.

. J.E. Meade, The Theory of Customs Union, Amsterdam:

North HollanleublIShing COmpany, 1955.

. R.G. Lipsey, "The theory of customs unions: a General

Survey", The Economic Journal, September 1960.

. Harry G. Johnson, Economic Policies Toward Less

Developed Countries, Washington D.C.: The Brookings
Ihstitution,’l967.

. M.E. Kreinin, "Effects of the EEC on imports of

manufactures", The Economic Journal, September 1972.

EFTA, The Effects of BETA on the Economies of Member

States , Geneva:u.n.1969. .

. B. Balassa, "European Integration: Problems and Issues",

American Economic Review, Papers and Proceedings, May 1963.

. N.D. Aitken and R.S. Obutelewicz, "A cross-sectional

study of EEC trade with the association of African
countries", The Review of Economics and Statistics,
November 1976:

. C. Young, "Association with the EEC: Economic aspects of

the trade relationship", Journal of Common Market Studies,
December 1972.

G.C. Hufbauer, "The impact of National characteristics
and technology on the commodity composition of trade in

manufactured goods", in R. Vernon (Ed.), The Technolo
Factor in International Trade, New York: NBER, CqumBla
University Press, 1970.

H.B. Lary, Im ort of Manufactures from Less Develo ed
Countries, New York: NBER, Columbia UniverSlty Press,

1968.

P.B. Kenen, "Nature, Capital and Trade",
Journal of Political Economy, October 1965.

50

13. W.B. Brown, "Market segmentation and international
competitiveness: Trade theory and practice re—examined",
Nebraska Journal of Economics and Business, Summer 1972.

14. H.G. Grubel, "Intra-industry specialization and the
pattern of trade", Canadian Journal of Economics and
Political Science, August 1967.

15. S.B. Linder, An Essay in Trade and Transformation,
New York: John Wiley, 1961. ‘

16. B. Balassa, "Country size and trade patterns: Comment",
American Economic Review, March 1969.

17. J. Neil Fortune, "Some determinants of trade in
finished manufactures", Swedish Journal of Economics,
September 1971. -

CHAPTER III

EVALUATION OF THE AFRICAN ASSOCIATED STATES
RESPONSE TO TARIFF PREFERENCES GRANTED BY THE EEC

l. The evaluation framework .

The purpose of this chapter is to determine whether,
during the 1962-1969 period, the AAS responded in a positive
manner to tariff preferences granted its manufactured
exports by the EEC. Several types of analyses will be used
in order to evaluate the AAS response. Although no single
analysis yields fully conclusive answers regarding the AAS
response, the over-all information generated by the analyses
should provide an adequate basis for a reasonably conclusive
evaluation of the AAS response.

Section 2 of this chapter describes changes in the
pattern of total EEC imports and total AAS exports, and
applies the methodology developed in chapter II in order to
determine whether the AAS responded to the tariff preferences
by expanding its total exports to the EEC. Although this
study is concerned with manufactured AAS exports, the analy-
sis of over-all trade will be useful in qualifying findings
regarding trade in manufactures. Section 3 applies the same
methodology to global manufactured exports from the AAS.
lSection 4 analyzes the AAS export pattern for various groups
cxf manufactures in order to determine whether the growth of
iflle market share of exports to the EEC is higher or lower

ifllan that of exports to other groups of countries. Section 5

51

52

applies the methodology developed in chapter II to a number

of individual manufactured commodities at the 3-digit SITC ‘
level. Homogeneous data at the 5—digit level could not be
obtained for this analysis. .The data that could be obtained
for individual commodities at the 5-digit level are analyzed,
in section 6, using a modified version of the methodology
developed in chapter 11. Section 7 compiles a list of labour-
intensive, low-skill intensive commodities exported by the

AAS to countries other than EEC countries. Finally, section 8
pools the information generated in the previous sections in

order to evaluate the AAS response to EEC tariff preferences.

Period of analysis

Since the EEC-AAS trade agreement took place in 1958,
the period of analysis should cover a span from pre-agreement
through post-agreement years (e.g., 1950-1970 period). Un-
fortunately, homogeneous and comprehensive data for AAS ex-
ports are not available for years prior to 1962. This year
was therefore selected as the first year of the analysis
period. The length of the period of analysis was selected
on the basis of two criteria. First, it should be long
enough to allow for the impact of tariff preferences to
take place. Second, as far as possible, the analysis period
should not include years in which EEC has established trade
agreements with other countries. If these years were included,
the evaluation of the AAS response would become much more com-

1)lex because of lack of proper control countries.

53

A review of EEO trade agreements with countries
other than the AAS shows that, starting in 1969, the EEC
has granted tariff preferences to Tunisia and Morocco
(1 September 1969), and to the East African Community,
i.e., Kenya, Uganda, and Tanzania (Agreement concluded
on 24 September 1969, effective beginning 1971). Further-
more, on July 1, 1971, the EEC granted substantial tariff
preferences to the majority of LDCs by instituting its own
GSP. Under these circumstances, it was decided to adopt a
period of analysis covering the years 1962 to 1969. The
adopted period should be long enough (11 years from 1958)
to allow for the impact of EEO preferences to take place.
It could be argued that EEC tariffs imposed on manufactured
AAS exports were not fully removed until 1968, and thus,
the period should extend beyond 1969. However, as shown
in Table l, tariffs imposed in 1962 on EEC imports from
outside the AAS are double those imposed on imports from
the AAS. This tariff differential should constitute a sub-
stantial incentive for the AAS countries to expand their
exports to the EEC at a relatively higher rate than non-

beneficiary countries.

Estimation method for rowth rates of the value
and market sHare of exports
Growth rates of the value and market share of ex-

jports are used in the analyses presented in sections 2 to 6

017this chapter. The method used to estimate these growth

54

.iem .av puma .omm "maonmswm
.mwauacseeou on» no mmaua>auo¢ on» so phenom Hmuocoo awash .conmaEEoo 0mm «mouaom

 

ooa

ma

ma

mm

om

ow

OH

on

on

CA

ov

00

OH

on

on

0H

ow

ov

0H

05

on

OH

on

on

0H

ca

0H

0H

.m<<

ms» scum mason
ISA so venom
ISA ma noanz
mmaumu accumuxm
coesoo 0mm ecu
mo ommuzoouom

macauoaoou
m>auwasafizo

Ho>oa nmma
humusMH H no

moms meoHuoapou
HMQOH>H©GH

 

 

ow.b.a

 

no.5.a

 

mm.H.H

 

mo.H.H

 

mw.h.H

 

Nm.h.H

 

~o.a.a

Ho.H.H

 

ow.h.a

 

 

¢m.H.H

 

 

“undo away

Wand 03H EON“ mn«H0nmEH UQHDUOMHDCME GO ON“ ”SUV HO mCOHHUDUOH MNNHflalch WAQ‘E

 

55

rates is by fitting an exponential trendline to the time
series data for the 1962-1969 period, and calculating the
grthh rate on the bais of the estimated value of the slope.

Let us define:

yo, yt value of the dependent variable
(e.g., market share) at years c and
year t, respectively.
r = the average growth rate for the period.
Then:
y = y (l + 1")t
t 0
Taking the logarithm of the above expression gives:
Log yt = Log yo + t Log(l + r)
The above equation may be rewritten:
Log yt = a + b.t
Using the least-square method, we obtain an estimate

b for b, that is an estimate for Log(l + r). The estimated

average growth rate, r, is then equal to:

A

r = eb - 1

In applying the least-square method, the values of
t range from t = o for 1962 to t = 7 for 1969. The mean
for the period is obtained by calculating the expected value

of the dependent variable for t = 3.5.

2. Analysis of total AAS exports and EEC imports
, Although the purpose of this chapter is to evaluate
the AAS response to tariff preferences granted to their

Inanufactured exports, it is of interest to analyze the over-all

56

AAS response to tariff preferences. On the basis of this
analysis, it will be possible to determine whether the over-
all AAS response concurs with the AAS response in the case
of manufactured exports. The knowledge of whether the two
responses concur may help explain the role of tariff pre-
ferences with respect to export expansion from beneficiary
countries.

First, changes in the EEC pattern of imports will
be compared to changes in the AAS pattern of exports over
the period of analysis. Second, the methodology developed
in chapter II will be used in order to assess the over-all

AAS response to tariff preferences.

Changes in the pattern of EEO imports and AAS
exports over the period ofanaIysIs

Table 2 provides the growth of the share of EEC
imports from the world, the AAS and "other LDCs", as well
as the growth of the share of AAS exports to the world, the
EEC and the "Rest of the World" (ROW). First, it can be
seen that the growth rates of the shares of EEC imports
from the AAS and "other LDCs" are approximately equal (—3.04
versus -3.21), and both are negative. Thus, tariff pre-
ferences do not seem, on the basis of this limited analysis,
to have had an impact on AAS exports to the EEC. (This con-
clusion is based on the assumption that the growth rate of
the share of imports from "other LDCs" constitutes a proxy
of what the growth rate of the share of imports from the AAS

\would have been in the absence of preferences.

.A>uucsoo m<¢ gone now oco .noaumwumum opmuu mo mossHo> vHV
mwma new wood .omImmmH xoon use» I moum«00mm< I mcaumauoum ensue cmaouom
"paw AHH one H mossao>v onma .mowuaczaaou :omQOusm on» No ooauuo Houaumaumum "mdommsum
.mehme x009 use» I moutaUOmmd I moaumflunum opens cmamuOh .uoauacsseou cmomousm ecu mo scammo Hoowumauoum mmmmmmflMlmflfl Amy

 

 

I
Ame .m .H wesHo>v HhmH .mwuacsaiou cmomousm may no oofluwo Houaumauwum ”uaomuzum
.owma I mwma.xoon use» I mwuuaoonmd I mowumwumum mouse cmaouom

 

 

57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.moauacsssou suomowsm on» no ooawmo Hooaumauoum "muwmmsa 0mm Asa uoowsom
mm.mm em.v on.sm ov.mm o~.om oa.om mm.mn .Wv.mm v~.s~ mm.e~ m:
o-.oem mo.ma mmv.mam omm.oom www.mmm mom.vmm oom.ovm mmo.m~m Hsv.vam mmm.mmH > onoz
one no umom
N~.oo mm.m I on.~o oo.ao ve.mm om.mo se.sm Hm.om ma.~s No.~e m2
who.vcm mm.e saa.mwm omo.mom sem.omo vam.amm omo.omo www.mvm ave.~nm mmm.emv > can
oo.ooH I oo.ooH oo.ooH oo.ooa oo.ooa oo.ooa oo.ooH oo.ooa oo.ooa m:
qm~.voo.a ve.m omm.vmm.a coo.moM.H mov.mmo.H o~a.mmo.a omv.smm mmo.oem ~H~.sms asm.amo > oawoz
«mm
muummmm was
mv.o~ H~.m I av.ea oo.ma em.ma me.ma mm.ma ma.0m Ha.o~ mm.mm ms
mem.e~v.m em.c ooo.ama.ma ooa.mcm.aa ooe.amo.oa oom.mvv.oa ooa.mve.m oom.mmo.m oom.mma.m oov.emm.m > mood wonuo
om.H vo.m I ov.a om.H om.H mm.a Hm.a om.a ms.a mn.a m:
ooo.mam mm.m oov.ooo.a oom.0mm 00m.mmm oo.amm oom.ome ooH.oom oom.mmm oom.eao > was
oo.ooa I oo.ooa oo.ooa oo.ooa oo.ooa oo.ooa oo.ooa oo.ooa oo.ooa m:
ooo.vvn.om oa.oa oom.mem.ms ooo.mmm.am ooo.moo.mm ooa.emc.mm 00m.omm.me oom.omm.ve ooo.vae.ov ooo.mmc.mm > oawoz
. "Baum
mowmmaH can
any
as com: mum»
susowm mesa moma soma coma moms «mad mesa mesa m”
> and: omnum><
Aucmu Mom Gav mumcm umxumz n m:
loco.~ » sac msam> n >
Amwma I moody muuomxm mdd Hmuou 0cm muuomaw 0mm Hmuou m0 owmnm umxume paw 03Hm> O .
a Eros II N mama...

 

58

Second, an analysis of AAS exports, also provided in
Table 2, shows that the lack of over-all AAS response to
tariff preferences cannot be attributed to an incapacity of
AAS to expand exports so as to meet the EEC demand of im-
ports. The over-all growth of AAS exports to the world
(+9.74%) is approximately equal to that of EEC imports from
the world (+lO.lO%). Moreover, as shown in Table 2, the
growth rate of the market share of AAS exports to the EEC
decreased, on the average, by 2.29% while that of AAS ex-
ports to the rest of the world increased, on the average,
by 4.87% per annum. The lack of AAS response to tariff
preferences must therefore be attributed to factors other
than the incapacity of the AAS to expand exports to the EEC.

One such factor could be that the AAS is over-all
less price competitive than "other LDCs".. Thus, the compe-
titive edge provided by the tariff preferences could be
offset by a growth of AAS prices higher than that of "other
LDCs". In order to determine whether such factor did affect
the AAS export performance, we must find out whether the
growth rate of AAS exports to "other DMECs" (i.e., Developed
Market Economy Countries other than the EEC Countries) is
also lower than that of "other LDCs". This will now be done

by applying the methodology developed in chapter II.

AAS response to tariff preferences
for over-all exports

It may be recalled, from chapter 11, that in order

to isolate the impact of tariff preferences, one must control

59

m manmfi ca manomxo m<¢ you uwsw mu meow

“In canoe .msma .nseozo ”xwoa 3oz

 

"muwomxo was Am.

 

 

 

smbma I acqumauoum ucmBmon>oa can dough HmcoHuocuuucH mo xoonccum .Q<Buza "muuamfim,monq AH¢ Adv "condom
mam.~mp.m~ mmm.amm.aa one.vva.mm soa.em~ Ada.mcm ku.m-.a ooo.omo.¢m ooo.o-.~H ooo.oem.om mmma
mmv.me>.dm vem.s~m.m www.cem.am mam.os~ ome.mom Hea.mso.a ooo.o~o.- ooo.oms.oa ooo.ome.~m moma
vmm.moe.mH mmv.meo.m ss¢.m¢m.m~ ooo.m~m sam.omo mmm.oam ooo.omm.mH ooo.ose.m ooo.ocv.m~ bead
«mm.omv.ma mmm.mv>.m omo.mma.e~ www.mmm «Hm.amm omm.oam ooo.omo.ma ooo.0me.m ooo.omo.m~ coma
aem.amo.sa omm.mm~.m amm.mm~.mm m~o.mma omm.m~o mvm.vmm ooo.0m~.ea ooo.omm.m ooo.oaa.m~ mood
sme.oem.oa mmm.vve.n -m.omo.e~ mmm.mma mvo.meo mea.mmm ooo.oem.ma ooo.omm.m ooo.omm.em coma
mhm.aaa.ma mm~.emm.m Hmm.mso.- mum.maa ave.~em moo.amm coo.0m~.ma ooo.ovm.e ooo.oew.- mood
omm.mmm.ma mam.~mm.m ~vo.mea.o~ osv.mm www.5mv wmm.omm ooo.omm.ma ooo.oms.m ooo.osn.om «cad
“Mmmwm one woman “Mmmum can moans mmmmwm one woman

 

 

 

 

 

 

 

 

sou =mooq wosuog «0 wuwodxm

 

"on mad mo muuomxu

 

no» swung Has: no uuwomxm

 

 

“coo.a » say

Ammma I Nmmav :mumzo Hmnuo: van Dam cu mwuomxm sauna Honuos wan mdd Hmuoa II.m mqmtfi

 

 

l‘IIIIIl‘I

 

 

 

60

for changes in both supply and demand conditions in the
preference-giving and preference-receiving countries over ,
the period of analysis. To do so, one must use two groups
of "control" countries: "other LDCs" needed to control for
changes in demand conditions in the EEC, and "other DMECs"
needed to control for changes in supply conditions in the
AAS. The growth rates of AAS and "other LDCs" exports to
the EEC and "other DMECs" are then compared in order to
determine whether the AAS responded to EEC tariff preferences.
It may be concluded that the AAS response was positive when-

ever (AG) is greater than zero, with:

_ (AG) = (G11 " G21) " (G12 ' G22)
where:
G11, G12 2 Growth rates of AAS exports to
respectively the EEC and "other DMECs"
G21, G22 = Growth rates of "other LDCs" exports

to respectively the EEC and "otherDMECs".

Table 3 provides the value of exports from the AAS and

"other LDCs" to the EEC and "other DMECs"l over the 1962-1969
period. I

The following growth rates and mean values were
estimated:

‘

l"Other DMECs" include the following countries:
11.8., Canada, EFTA countries, Finland, Ireland, Greece,
ESpain, Turkey, Yugoslavia, Japan, Australia, New Zealand
61nd South Africa.

61

Growth rate Mean value

 

 

0%) (1n.$1.000)
AAS exports to EEC : G11 2 7.24 619,187
AAS exports to "Other DMECs": G = 15.12 162,903

12

"Other LDCs" exports to EEC : G 7.96 7,792,036

21

"Other LDCs" exports to
"Other DMECs" . : G = 7.65 16,686,465

22

The above growth rates yield a value for (11G) of
-8.19%. Thus, it may be concluded that, in the case of over-
all exports, the AAS did not respond to EEC tariff preferences.
Indeed, the above findings indicate that, for some unknown
reasons, the AAS performed better in the markets of "Other
DMECs" than in those of the EEC countries.

At this level of commodity aggregation test findings
should be considered with caution. Since there may be many
commodities which are exported by "Other LDCs" but not by
the AAS - and vice-versa -, the impact of tariff preferences
on commodities exported by both groups of countries may be
"hidden" by changes in the export of the former commodities.
In order to reduce this problem,the evaluation methodology

will be applied, in section 5 and 6, to disaggregated commo-

dities at, respectively, the 3-digit and 5-digit SITC levels.

3. Evaluation of AAS response for over-all
manufacturedexports

The methodology used for global exports is applied,
jLn this section, to total manufactured exports, excluding

E>etroleum products, unworked non-ferrous metals, and a number

62

of manufactured commodities in SITC groups 6 and 7
(e.g., 711 and 735 when exported by LDCs).l

In this analysis, the "control" group "Other DMECs"
is limited to 13 majOr countries: the 8 EFTA countries,
USA, Canada, Australia, New Zealand, and Japan. Due to
lack of readily available data, it was not possible to
include 7 additional countries defined by the U.N. as DMECs
(i.e., Finland, Iceland, Greece, Turkey, Spain, Yugoslavia
and South Africa).

Table 4 provides the value of exports from AAS and
"Other LDCs" to EEC and the above group of "Other DMECs"
over the 1962-1969 period. The following growth rates and
mean values were estimated:

Growth rate Mean value
(%) (in $1,500)

AAS exports to EEC G11 2 13.04 61,540
AAS exports to "Other DMECs": G12 2 19.19 16,964
"Other LDCs" exports to EEC : G21 = 9.04 933,773
"Other LDCs" exports to

"Other DMECs" : G22 = 15.56 2,310,553

Given the above growth rates, we obtain a value for
( AC) of +.37%. This value is positive but very close to

zero. Thus, it may be concluded that in the case of total

*

lThe manufactured exports analyzed in this section
(rorrespond to those included in UNCTAD's "Total A" (i.e.,
rnanufactured goods included in SITC groups 0 to 9, excluding
3531.02, 332, 341.2, 351, 513.65, 667, 681, 682.1, 683.1,
6585.1, 686.1, 687.1, 689, 711, 735, and 961.0).

63

.oump o: .mu50Iucaum newsmaoo .a¢soz= un>ocoo

 

.monsuomwscmaIaaom can mousuuumscma mo mummmmm .Q¢BUZD “mousom

 

 

oom.mmc.e coo.~oc.a co~.om~.c mec.~m cmm.maa mmm.cea mcc.oam.v ccc.m~m.~ mam.c~v.c mcma
mmm.ecc.m emc.mma.a eH~.mvo.m cao.c~ mcm.~oa Hom.m~H Hem.oam.m mme.oc~.a cHo.HmH.m mcmH
cmc.mca.m cmo.moo.H ~mc.cmd.v cmm.m~ moH.Hc mmm.voa mmc.mma.m mma.cmo.a Hmc.Hm~.v mcma
mmo.¢vm.~ mom.cua.a mcm.~mm.m «mm.om mv~.mc mso.cm Ham.emm.~ mmm.mma.a vcc.ccm.m ccmH
cme.mae.~ mmc.mmm ace.mcm.m cmm.m~ oma.mm mam.mc cmc.meq.~ mmc.aao.a mmm.mmc.m mcma
acm.oma.m oao.mam Hmm.mco.m mam.cH moa.mm cao.qs vec.mca.m mHH.omm mcm.nma.m ecma
mo~.mmc.a mmc.~mm cmo.~ec.~ mvo.HH ~mc.mc emm.cc Hm~.omc.a Hem.mcm www.mme.m mcma
vac.mmc.a mmm.mmm mmm.mmv.m Hmc.m Hcc.mv www.cm cm~.mcc.a omc.avm ccc.m0m.~ mcma
mwmmum can momzo ma “Mmmum one means ma “Mmmwm one woman ca

 

 

 

 

 

 

 

 

sou =mooq wosuo: «0 uuwomxu

 

“on mad mo manomxm

 

"on gmooq Mac: co cowomxu

 

 

Aooo.H a adv

Ammmd I Nomav

..MUQ.H 80390: UCM m:
ecu Bonn smumzo nonuos can own ha mwuduomuscmalaaom can nmuzuoduscna mo.muuomEH II.v.mAm49

 

64

manufactures, the AAS did not respond to EEC tariff pre-
ferences.

This finding should be considered with caution for
two reasons. First, as in the case of total exports, the
test is applied to highly aggregated export data. Second,
the 221k of AAS manufactured exports may consist of commo—
dities for which the EEC common tariff is low, and thus
does not favor a positive AAS response to preferences. The
results of the analysis does not exclude the possibility of
a positive AAS reSponse limited to a few commodities for

which EEC tariffs are relatively high.

4. AAS exports of various groups of manufactures
to the EEC and other groups of countries

This section will provide an over—all view of the
pattern of exports from the AAS with respect to various
groups of manufactured commodities. It constitutes only a
partial evaluation of the AAS response to tariff preferences
since "Other LDCs" exports to EEC and to "Other DMECs" are
not taken into consideration. Due to lack of data, it was
not possible to apply the methodology developed in chapter II
to groups of manufactured exports. Nevertheless, findings
from this section will be helpful in qualifying those ob-
tained in other sections of this chapter. The analysis in
this section is designed to determine: (i) whether, for
various groups of manufactured commodities, there was a

shift in the geographical pattern of AAS exports in favor

65

of the EEC over the 1962-1969 period, and (11) whether the
size of this shift was a function of the type of commodity
being exported (i.e., finished commodities versus semi-

manufactures).

The data

Coverage of manufactured exports. The definition
of "manufactures" used in this analysis is somewhat less
restrictive in coverage than that usually adopted by
international organizations and trade economists. Speci-
fically, it covers a number of products from groups 0 and 2
of the CST Commodity Classification1 which are usually
classified as agricultural materials: for example, in
Group 0, frozen or chilled meat and simply prepared (dried
or salted) fish; in Group 2 wood sawn length wise and rail-
. way sleepers. Such goods are included in this study as
"manufactures" because their production requires a certain
level of processing, albeit a fairly uns0phisticated one.
The technological level of the AAS is so low that a pro-
cessing level which may be considered as non-significant
in developed countries should be considered as significant
in the AAS. Consequently, simply processed raw materials

are classified as manufactures.

‘1

lCST stands for "Statistical and Tariff Classification
For International Trade". This is the classification used by
the Statistical Office of the European Communities with res-
pect to AAS export statistics.

66

Exclusion of re-exports. Comprehensive data on
exports from the AAS during the 1962-1966 period were
published for the first time by the Statistical Office
of the European Communities in 1968. Export data for the
years 1967 to 1969 were published two years later in 1970.
Exports are coded according to the Statistical and Tariff
Classification (CST) by groups (1 digit classification),
sub—groups (3 digit classification) and by individual
commodities (5 digit classification). The published
tables provide AAS exports to the world, the EEC and in-
dividual countries. These EEC publications constitute the
only source of comprehensive foreign trade statistics for
all the years included in the period of analysis.

It was noted in the forewords of these publica-
tions that in order to improve the reliability of the
compiled foreign trade statistics various adjustments of
the original data had been carried out. These adjustments
did not, however, include the exclusion of re-exports. A
cursory look at the export statistics indicates that a
number of exported commodities could not have been pro-
duced by the AAS countries. The obvious examples are air-
plane engines and machine tools. Such items either consti-
tute re-exports or are being shipped abroad for repairs.
Export items of the above type may be easily identified
and excluded from the analysis. However, there also exist

questionable export items which cannot be rejected off—hand

67

without checking whether they are actually produced in the
exporting country. Examples of such items include bicycles,
cars, various chemicals, and even cigarettes.

In order to identify and exclude from the analysis
various types of re-exports, it was necessary to compile
for each country within the AAS a list of manufactured
commodities produced in the country. It was then assumed
that if a commodity is actually produced in the country,
it constitutes a potential export commodity. If it was
subsequently found that this commodity was exported, the
export was assumed to be genuine.

Industrial statistics for each AAS country were
obtained from a 1971 publication1 which provides complete
and detailed information on the manufacturing industries
of each country, including the name, location and size of
firms, the starting year and volume of production, and in -
some cases, the destination of the production of the firms
(e.g., local market, foreign market in Africa, EEC). Manu—
factured exports which could not be identified from the
above list were excluded on the assumption that they con-
stituted re-exports. In cases where the list shows that
the production of a given manufacture started after 1962,
exports between 1962 and the starting year of production

were also rejected as re-exports.

—

lEdiafrie-Service (1)

68

Table A-1 of the Appendix provides a complete list
of true exports from each country within the AAS,-including
the year in which exports of each commodity began.

Grouping of manufactured commodities. The aggrega—
tion of manufactured commodities into various groups pre-
sents a double interest. First, the separate analysis of
individual groups will indicate which type of exports to
the EEC the AAS succeeded in expanding or failed to expand.
Second, since tariff preferences are generally of greater
interest for finished goods than for simply processed raw
materials or intermediate goods,1 it is of interest to
determine whether AAS exports of finished goods to the EEC
grew at a higher rate than AAS exports of intermediate
goods. Tentative conclusions may then be drawn from such
cross-sectional analyses as to whether-tariff preferences
have been successfully used by the AAS to expand their
exports of finished goods to the EEC.

In this section, manufactured exports are aggregated
into 11 groups: total exports, simply processed raw materials

and intermediate goods, finished goods, and each of the CST

 

1Since the impact of tariff preferences on export ex-
pansion is a function of the level of tariff duties, and
SIINBG these duties tend to be higher as the value added by
nmnnlfacturing increases, the impact of tariff preferences
should.be larger for finished goods than for raw materials
cxr intermediate goods. Moreover, this impact should be re-‘
:hmforced by the fact that the difference between the effective
talfixff and the nominal tariff is much larger for finiShed
nunnlfactures than for semi-manufactures.

69

groups 0 to 8, excluding<3roup 3.1, CST groups 0, 5 and 6
include both intermediate and finished goods. CST groups
1, 7 and 8 include finished goods only, while CST groups
2 and 4 include intermediate goods only. The classifica-
tion of individual commodities into finished or inter—
mediate goods is provided in Table A-1 of the Appendix.
Table 5 provides the over-all share of each of the two
types of goods within each group.

In some cases, a commodity could be classified
either as an intermediate good or as a finished good,
depending on whether it is to be further processed in
industrial establishments, or is to be retailed as a con-
sumer good. For example, if palm oil is exported already
refined and bottled for home consumption it would be classi—
fied as a finished good. On the other hand, if it needs
further processing before it could be retailed, it should
be classified as an intermediate good. Since the 5-digit
CST classification is not always sufficiently disaggregated
to permit an unambiguous distinction between finished and
intermediate goods, a certain degree of arbitrariness could

not be avoided when goods were being classified.

Grouping of importing countries. Countries im-
porting from the AAS are grouped according to specific

geographical and economic characteristics. The AAS export

 

lSince exports of commodities in CST group 3 did
not take place until the end of the period, they were not
included in the analysis. '

70

TABLE 5.--0ver-all share of intermediate goods and of
finished goods in individual groups of

manufactured AAS exports

 

 

 

 

 

(Per Cent)
Comgggity . Share of ' Share of
Group intermediate goods finished goods
0 53.57 46.43
1 0.00 100.00
2 100.00 0.00
4 100.00 0,00
5 57.06 42.94
6 75.60 24.40
7 0.00 100.00
8 0.00 100.00

 

Source: Same as Table 2.

 

71

performance to various groups of countries may then be
compared, and some tentative conclusions reached regarding
the relationships between the pattern of AAS exports to
groups of countries and the characteristics of the8e count-
ries.

Countries are aggregated into 5 groups: Africa,
"Other LDCs", EEC, "Other DMECs" and Socialist countries
of Europe. The group "Africa" consists of all countries
in the African continent, including the AAS. The charac-
teristics of this group are (i) geographical proximity to
the AAS, and (ii) a development level generally similar to

that of the AAS.1

The group "Other LDCs" includes all
developing countries in the Middle East, Latin America and
Asia. The characteristics of this group are: (1) large
geographical distance from the AAS, and (11) with some
exceptions in the Middle East and Latin America, 8 develop-
ment level similar to that of the AAS. The EEC constitutes
a special group among the developed countries because of

its special trade relationship with the AAS. The "Other
DMECs" group includes all developed market economy countries
with the exception of the EEC. The characteristics of this

group are: (i) large geographical distance from the AAS, and

(ii) a high level of development relative to that of the AAS.

 

1An additional characteristic is that some of the AAS
countries are grouped within two customs unions (see section 2
of chapter IV), a factor which may favor trade among these
countries. Thus, if tariff preferences were to be effective
in promoting trade, the growth rates of AAS exports to the EEC
and Africa should be higher than those of exports to other
groups of countries.

72

These two characteristics also apply to the EEC. The fifth
group includes all soCialist countries ("SOC. countries") in
Europe. The characteristics of this group are similar to
those of "Other DMECs", but its trading policies differ

significantly from those of "Other DMECs".l

Evaluation of AAS export performance
during the 1962—1969 period

This evaluation is based on estimates of growth
rates summarized in Table 6. These estimates were cal-
culated from export statistics (values and market shares
for the 1962-1969 period) provided in Tables A-2a and A-2b

of the Appendix.

Total manufactured exports. As shown in Table 6,
total AAS exports to the world grew at the relatively high
rate of 8.72% per annum. Close to 99% of total exports
were distributed among the EEC (71.43%). Africa (14.52%)
and "Other DMECs" (12.72%). The growth rate of exports to
Africa (16.45%) is higher than that of exports to the EEC
(6.10%) and than that of exports to "Other DMECs" (14.90%).

The growth rate of the share of exports to Africa
(7.11%) is higher than that of exports to the EEC (-2.4l%)
and than that of exports to "Other DMECs" (5.68%). It may

therefore be concluded that total manufactured exports grew

 

1Although the countries included in each group are
not fully homogeneous with respect to the characteristics
listed in each case, it was decided to limit the analysis
to these five groups.

73

TABLE 6.--AAS exports of individual groups of manufactured
commodities - Growth rate of value and market

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

share
V = Value of exports
MS = Market share of exports
l = Mean value (in $ 1,000)
2 = Growth rate of V (in Per cent)
3 = Mean value of MS (in Per cent)
4 = Growth rate of MS (in Per cent)
Destination
Commod i ty Social .
Group . ”Other "Other countr.
World Africa LDCs" EEC DMECs" of
Europe
Total V 1 380,141 54,586 3,127 268,470 47,810 1,843
Manu- V 2 8.72 16.45 14.14 6.10 14.90 23.95
factured MS 3 100.00 14.52 .83 71.43 12.72 .49
exports MS 4 - 7.11 5.07 - 2.41 5.68 13.44
Semi- V 1 289,675 15,558 1,656 228,904 37,751 1,790
manu- V 2 7.31 13.15 14.62 5.61 16.31 24.16
factured MS 3 100.00 5.45 .58 80.13 13.22 .63
goods MS 4 - 5.43 6.63 - 1.58 8.39 15.90
Finished v 1 89,677 38,606 1,173 39,441 9,611 24
goods V 2 13.12 17.55 22.89 9.06 10.27 52.36
MS 3 100.00 43.45 1.32 44.39 10.82 .02
MS 4 - 3.92 8.75 - 3.59 -2.52 98.63
CST V 1 134,064 24,464 1,297 86,187 20,333 351
Group 0 V 2 12.27 9.70 6.99 12.67 11.30 77.21
MS 3 100.00 18.45 .98 64.98 15.33 .26
MS 4 - -2.29 - 4.79 .35 - .87 58.21
CST V 1 2,292 1,871 4 365 0 0
Group 1 V 2 14.02 17.52 80.39 - 2.05 .00 .00
MS 3 100.00 82.90 .93 16.17 .00 .00
MS 4 - 3.07 84.51 -14.10 .00 .00
CST V 1 121,679 5,918 492 96,539 15,318 668
Group 2 V 2 4.42 6.43 29.17 2.55 21.49 - 3.40
MS 3 100.00 4.98 .41 81.17 12.88 .56
MS 4 - 1.93 24.30 - 1.79 16.35 - 7.19

 

 

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 6.-- (Continued)
Destination
Commodity Social.
Group . "Other "Other counI.
World Africa LDCs" EEC DMECs" of

Europe
CST V 1 58,820 2,336 23 52,647 1,040 10
Group 4 V 2 3.56 7.90 -11.97 2.01 80.90 .00
MS 3 100.00 4.17 .04 93.92 1.86 .02
MS 4 - 4.21 -16.00 - 1.49 73.22 .00
CST V 1 5,493 1,467 33 2,301 1,371 0
Group 5 I V 2 17.50 49.88 7.68 8.27 8.75 .00
MS 3 100.00 28.36 .64 44.49 26.51 .00
MS 4 - 27.22 - 8.33 - 7.86 -7.44 .00
CST V 1 44,244 9,893 748 25,539 6,440 17
Group 6 V 2 8.09 28.84 32.25 2.27 2.00 17.04
MS 3 100.00 23.20 1.75 59.90 15.10 .05
MS 4 - 19.19 22.30 - 3.39 -3.64 6.05
CST V 1 .3,002 1,709 14 1,052 77 7
Group 7 V 2 16.76 20.02 - 4.46 9.29 59.90 31.74
MS 3 100.00 59.78 .49 36.80 2.69 .24
MS 4 - 2.79 -18.15 —6.40 37.10 37.48
csr ' v 1 3,270 2,953 13 234 22 0
Group 8 V 2 32.33 34.28 21.05 14.29 39.45 .00
MS 3 100.00 91.65 .40 7.26 .69 .00
MS 4 - 1.47 1.39 -13.62 5.11 .00

 

Source: Growth rates estimated on the basis of export
statistics provided in Tables A-Za and A-Zb of
the Appendix.

 

75

‘at a relatively high rate, and that this export expansion

favored Africa and "Other DMECS" rather than the EEC.

Exports of semi-manufactured commodities. Over-all
exports of intermediate commodities grew at a relatively
high rate (7.31% per annum). Close to 99% of total exports
were distributed among Africa (5.45%), the EEC (80.13%)
and "Other DMECs" (13.22%). The growth of exports to the
latter group of countries (16.31%) is higher than that of
exports to Africa (13.15%) and than that of exports to the
EEC (5.61%). The growth rate of the share of exports to the
EEC (-1.58%) is much lower than that of exports to Africa
(5.43%) and than that of exports to "Other DMECs" (8.39%).
Therefore, the conclusions regarding total manufactured ex-

ports also apply to semi-manufactured exports.

Exports of finished commodities. Over—all exports
of finished commodities grew at a much higher rate (13.12%)
that exports of intermediate commodities. Thus, it would
seem that the AAS has applied a greater effort and shown
a greater capability in expanding exports of finished goods
than in expanding exports of intermediate goods. Close to
99% of total exports of finished goods were distributed
among Africa (43.45%). the EEC (44.39%) and "Other DMECs"
(10.82%). While Africa ranks third (in terms of the share
of exports) in the case of semi—manufactured goods, it ranks
second in the case of finished goods. Furthermore, the growth
rate of exports of finished goods to Africa (17.55%) is much
higher than that of exports to the EEC (9.06%) and than that

76

of exports to "Other DMECs" (10.27%). The growth rate of the
share of exports to Africa (3.92%) is also much higher than
that of exports to the EEC (-3.59%) and than that of exports
to "Other DMECS" (4.52%).1 It may also be noted that the
growth rate of the share of finished exports to the EEC
(-3.59%) is lower than that of semi-manufactured exports to
the EEC (-l.58%). Since tariff duties for finished exports
are usually higher than those for semi-manufactured exports,
it would seem that the granting of tariff preferences to the
AAS has failed to have the expected impact (i.e., relatively
higher expansion of exports of finished commodities to the

EEC than that of semi-manufactured commodities).

Exports of Group 0 commodities (Processed agricultural
materials). The growth rate of Group 0 exports (12.27%) is
higher than that of over-all exports (8.72%). The growth rate
of exports to the EEC (12.67%) is higher than that of exports
to Africa (9.70%) and than that of exports to "Other DMECs"
(11.30%). Africa, the EEC and "Other DMECs" account for
close to 99% of total exports in this group. The growth rate
of the share of exports to Africa (—2.9%) is lower than that
of exports to the EEC (.35%) and than that of exports to "Other
DMECs" (-.87%). Exports in Group 0 are approximately divided

between semi—manufactured goods and finished goods (see Table

 

1The fact that preferential trade agreements exist
between some AAS countries and some African countries should
not provide a full explanation for this high trade intensity
of intra-African trade.

77

5). Since tariff duties are relatively high (20% to 80%)
for the finished exports,l there may have been an attempt
by the AAS to respond positively to tariff preferences.
However, given the low growth rate of the share of exports

to the EEC (.35%), no firm conclusion can be reached.

Exports of Group 1 commodities (Manufactured tobacco
and beverages). Exports to the world grew at a fairly high
rate (14.02%), with over 99% of total exports going to Africa
(82.90%) and to the EEC (16.17%). The growth rate of exports
to Africa (17.52%) is very much higher than that of exports
to the EEC (-2.05%). Decreasing exports to the EEC account
for the marked decrease of the EEC share of exports (-l4.10%),
as against the share of Africa (3.07%). These are to some
extent surprising results since tariff duties on manufactured
tobaccos and beverages are usually very high. For spirits,
the main Group 1 export to the EEC, a tariff duty of 107% is
imposed on imports from "Other LDCs" (see Table 8). One ex-
planation — assuming that the AAS is as competitive on the
EEC market as other groups of countries - is that the AAS is
unable to produce beverages or tobacco products of the type

and quality which cater to the tastes of EEC consumers.

Export of Group 2 commodities (Processed raw
gmaterials). Over—all exports of Group 2 commodities grew at

aa fairly low rate (4.42%), with over 99% of these exports

k

1Tariff duties imposed by the EEC on imports from
"0ther LDCs" are presented in section 6, Table 8.

78

going to Africa (4.98%), the EEC (81.17%) and "Other DMECS"
(12.88%). The growth rate of exports to the EEC (2.55%) is
lower than that of exports to Africa (6.43%) and than that
of exports to "Other DMECs" (21.49%). Consequently, the
share of exports to the EEC decreased by 1.79%, while the
share of exports to Africa increased by 1.93%, and that to
"Other DMECs" by 16.35%. Given the generally low level of
tariffs imposed on Group 2 commodities (2% to 7%),1 tariff
preferences would not be expected to have a significant im-

pact on AAS exports of these commodities.

Exports of Group 4 commodities (Oils and fats).
Group 4 exports are of particular interest to the AAS since
these commodities are natural resource intensive, and the
AAS is a main producer of the raw materials used in their
production. Tariff duties on these commodities are moderate,
with a tariff of 7% applying to the main Group 4 commodities
exported by the AAS (see Table 8). Thus, preferences could
be expected to have a moderate impact on the expansion of
AAS exports to the EEC.

The over-all growth rate of Group 4 exports is
fairly low (3.56%), with over 99% of total exports going to
the EEC (93.92%), Africa (4.17%) and "Other DMECs" (1.86%).
The growth rate of exports to "Other DMECs" (80.90%) is
much higher than that of exports to Africa (7.90%) and than

 

lTariff duties for Group 2 commodities were obtained
from the Statistical Office of the European Communities (2).

79

that of exports to the EEC (2.01%). Consequently, the share
of exports to the EEC declined (-l.49%) while the shares of
exports to Africa and to "Other DMECs" increased by 4.21%
and 73.22%, respectively. These results suggest that tariff
preferences, albeit moderate, did not provide an incentive
for the AAS to expand substantially its Group 4 exports to
the EEC.

Export of Group 5 commodities (Chemicals). A larger
number of chemical products are exported to Africa than to
the EEC. The main chemicals exported to the EEC are glycerin
(CST 512-26), essential oils and resinoids (CST 551-10),
superphosphates (CST 561-29) and starches (599-51). For
starches, the tariff duty is 28.1%. For the other commodi-
ties duties are low to moderate (0.3% — 6%).

Over—all exports of Group 5 commodities grew at a
fairly high rate (17.50%), with over 99% of total exports
going to Africa (28.36%), the EEC (44.49%), and "Other DMECs"
(26.51%). The growth rate of exports to Africa (49.88%) is
much higher than that to the EEC (8.27%) and than that to
"Other DMECs" (8.75%). Consequently, the share of exports
to Africa increased substantially (27.22%) while the shares
of exports to the EEC and to "Other DMECs" decreased by
7.86% and 7.44%, respectively. It would therefore seem
that tariff preferences, albeit moderate to low, failed

to expand AAS exports to the EEC.

80

Exports of Group 6 commodities (Mainly processed raw
materials). The main Group 6 exports to the EEC include
various leather and wood products, products made of vegetable
fibers, and unwrought aluminium and aluminium alloys. As
shown in Table 8, tariff duties on these commodities are
moderate to high, ranging from 8% to 21%.

Over-all Group 6 exports grew at a relatively high
rate (8.09%), with over 98% of total exports going to Africa
(23.20%), the EEC (59.90%), and "Other DMECs" (15.10%). The
growth rate of exports to Africa (28.84%) is very much higher
than that of exports to the EEC (2.27%) and than that of ex-
posts to "Other DMECs" (2.00%). Consequently, the share of
exports to Africa increased substantially (19.19%) while the
shares of exports to the EEC and to "Other DMECs" decreased,
respectively, by 3.39% and 3.64%. It would therefore seem
that tariff preferences, although moderate to high, failed

to induce the AAS to expand substantially Group 6 exports
to the EEC.

Exports of Group 7 commodities (Machinery and trans-
pprt equipment). It is suspected that some of Group 7 ex-
ports to the EEC constitute re-exports. This suspicion is
basead on the unevenness of exports statistics for individual
commodities over the 1962-1969 period.' However, since the
AAS produces all Group 7 commodities shown in Table A—1 of

the Appendix, it was not possible to exclude certain commo-

dities as re-exports.

81

Over-all Group 7 exports grew at a fairly high rate
(16.76%), with over 99% of total exports going to Africa
(59.78%), the EEC (36.80%) and "Other DMECs" (2.69%).

The growth rate of exports to "Other DMECs" (59.90%)
is higher than that of exports to Africa (20.02%) and than
that of exports to the EEC (9.29%). Consequently, the share
of exports to "Other DMECs" and to Africa increased, respec-
tively, by 37.10% and 2.79%, while the share of exports to
the EEC decreased by 6.40%. Since tariff duties on Group 7
commodities are usually high (13% to 20% for imports origina-
ting from "Other LDCs"),1 it would seem that tariff preferen-
ces failed to induce the AAS to expand substantially its ex—

ports to the EEC.

Exports of Group 8 commodities (Mostly non-durable con-
sumer goods). This group of exports is of particular interest
to the AAS since it includes a large number of labor-intensive
commodities for which the AAS should enjoy a comparative advan-
tage in production and trade, and since tariff duties imposed
on these commodities are usually high (on the average, l5-25%).

Over-all Group 8 exports grew at a very high rate
(f52.33%). Close to 99% of total exports were divided be-

tWeenAfrica (91.65%) and the EEC (7.26%). The growth rate
Of‘ exports to Africa (34.28%) is much more higher than that
Of exports to the EEC (14.29%). Thus, while the share of

GXPDOIts to Africa grew by 1.47%, that of exports to the EEC

dec2reased by 13.62%. It therefore seems that Group 8 exports

__~

_ lTariff statistics were obtained from the Statistical
folmze of the European Communities (2).

82

represent a clear case where tariff preferences failed to
induce the AAS to expand substantially its exports to the
EEC.

Summary of findings on the AAS export performance

The preceeding analysis of AAS exports shows that,
with exception of Group 4, the growth of exports of manu-

factures was moderate to high. The over-all export perfor-
mance of the AAS can therefore be considered as more than
satisfactory. Its export performance did not, however,
particularly favor the EEC. With the exception of Group 0,
where the growth of the share of exports to the EEC was
marginally higher than that of exports to "Other DMECs"
(.35% versus -.87%), the EEC ranked last, in terms of the
growth rate of the market share of exports, after.Africa
and "Other DMECs".l This result prevails not only when EEC
tariff duties are low (e.g., Group 2 and Group 5 commodities),
but also when duties are moderate to high (e.g., Group 1, 4,
6, 7 and 8 commodities).

On the basis of the analysis conducted in this
Section, it may be concluded that the high expansion of AAS
eacports during the 1962-1969 period cannot be attributed to
arl expansion of exports to the EEC, and that the findings in
thnis section are at odds with those which would be expected

on the basis of tariff preferences. It must be noted, however,

¥

lSince socialist countries and "Other LDCs" accounted
tOg'ether for generally less than 1% of total exports, they need
not be considered when evaluating AAS export performance.

83

that the above analysis does not permit us to conclude, un-
ambiguously, that the AAS did not respond to EEC tariff pre-
ferences. In order to reach firm conclusions about the AAS
response, one must also take into consideration "Other LDCs"
exports to the EEC and to "Other DMECs". Furthermore, the
analysis should be conducted for individual commodities.
This type of analysis will be carried out in the following

two sections.

5. Evaluation of AAS response with respect totindividual
commodities at the 3-digit level of the SITC classification

 

In this section, the evaluation methodology developed
in chapter II is applied to 13 major manufactured commodities
at the 3—digit level of the SITC classification. Lack of
readily available trade data prevented application of the
methodology to other major commodities classified as manu-
factures in the previous section.1 A few additional commodi-
ties were excluded from the analysis because the value of ex-
ports to "Other DMECs" was equal or close to zero for the
whole period of analysis, and the evaluation methodology
could not be reliably applied. These commodities are, how-

ever, analysed (at a higher disaggregation level) in section 6.

L

lImport statistics used in this section were obtained
.from unpublished UNCTAD statistics on manufactured imports by
iflle EEC and 13 other DMECs. The UNCTAD definition of manu-
Ifiictures is more restrictive than the one adopted in the pre-
‘Vious section. Thus, a number of major AAS exports, such as
frozen meat, salted or dried fish, milled rice, raw sugar,
gill-cake, and various vegetable oils, could not be included
III the present analysis.

84

Commodities analyzed in this section include the
following main SITC sub-groups: O53 (fruit, preserved and
fruit preparations), O55 (vegetables, preserved or prepared),
072-3 (coc021 butter and paste), 243 (wood, shaped or simply
worked), 431 (animal and vegetable oils and fats, processed,
and waxes of animal or vegetable origin), 512 (organic-
chemicals), 551 (essential oils, perfumes and flavor materialsL
599 (chemicals, n.e.s), 611 (leather), 631 (veneers, plywood
boards), 632 (wood manufactures, n.e.s.), 657 (floor coverings),
and 684 (aluminium).

The two groups of "control" countries used in this
section are the same as those used in section 3 (i.e., "Other
LDCs" and "Other DMECs", the latter group including 13 major
countries).

Table 7 provides the growth rates of EEG and "Other
DMECs" imports from the AAS and "Other LDCs". These growth
rates were estimated on the basis of import statistics pro-
vided in Table A-3 of the Appendix. Table 7 also provides
estimates of the parameter (24G) which indicates whether the
AAS responded to EEC tariff preferences. As shown in the
above table, (.40) is positive for only 4 commodities (SITC
sub—groups 431, 551, 631, and 632) out of 13.

In order to determine whether the AAS response is

Positively correlated with the height of the EEC common tariff}

¥

1Tariffs provided in Table 7 are the average tariffs
fOI‘the commodities included in each sub-group.

ANv.moHuwcoEEoo commousm on» NO mofimmo Hmoaumaumum on» Eoum mowumaumum wwwume
xaosoaaa one No mus oases ca oooa>owm
mowumfiumum down» no mammn ecu co poumHSUHmo moumu nuzoum pom mooam> com: "mousom

85

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

om.m mo.v I mm.nm nmw.m vm.v I «ma ma.nh mnm.m om. vHN.mH vwm
om.m~ Hm.>a I wo.m mmm.~w «5.0 av ha.ma mmv.hm mo.H use hmo
ow.ma mm.m + mm.om ommsva mm.>~ h om.h Hom.m oo.NN .mh Nmm
om.HH mm.ma + mo.HN mhvshm ww.v mamsv om.wN omms¢ vv.m~ mamsv Hmm
om.m om.m I mawm ome.mm mm.oaI h mm.m~ mmm.~m v0. mwm Ham
om.m vm.m I mo.~ mmwsmm mH.H Nv no.m I bamso vw.h vav mam
om.H mv.v + vv.~ hamshm wo.oa mmmsm mo. hmmsha on.~a vmmsa Hmm
ow.mm mv.mm I wh.HN mmhsmm mm.Hm m mo.ma mesha hm.h mm Nam
om.o mm.om + cm. I mmwsma vv.hHI ma hm.a ommsv vm.m mmm Hmv
om.H mn.mH I vo.m mmm.HNH w¢.mm awhsv mN.m mmoshm mm.ma vmo.h mum
oo.mm wN.mm I oo.oa momtam mm.mv mphsa mw.mm whosm mv.hm mom.m m.Nho
om.ma mm.ma I Nw.va Ham.mm NN.wH NN m¢.m mamsmv vw. mvmsa mmo
oo.mm mo.NHHI mm.HH «mmsvm Nh.mma NH mm.m Hamsov mm.ma HHH.m mmo
Amy Avv Amy Amy AHV

muse avvuxmvv mo >2 mo >2 mo >2 mo >2
WWW—AHMW AA NV I A.__.. a : MUCH H$£HO= mg : mom—A ngUO: Mdd @mVOU
cam on. "Eouw muuomaa momzo ma "Eouw muuomEH 0mm UBHm
usmu mom Ga "mono“ £u3ouo H mm

ooo.a » ca "modam> coo: u >2

mead I mmma poaumm on» How woven susoum can mosam> some
I smooq porno: one m<< sown mquQEa soomzo assoc: was 0mm II.n mamas

86

( AG) was regressed on the tariff level - which measures the
margin of preferences. The following estimated regression

equation was obtained:

( Ag) : 15,50 - 2.29T
(tz2,562, significant at p.<-O5)
where:
T = tariff level.

The above cross—section findings indicate that the
over-all AAS response to EEC tariff preferences is at best
weak: the AAS reSponded positively for only four commodities
out of 13, and the estimated regression equation indicates

that the AAS response is negatively correlated with the tariff

 

level.

It may be noted that three out of the four SITC sub-
groups for which the AAS responded positively include manu-
factured commodities traditionally exported by the AAS to the
EEC. The three sub-groups are 431 (animal and vegetable oils
and fats), 551 (essential oils, perfumes), and 631 (veneers,
plywood boards). Thus, the AAS response seems to have been
limited to an expansion of its traditional exports to the
EEC. Furthermore, as shown in Table 7, the EEC tariffs for
the four commodities are very low (0.9% for sub-group 431,
1.5% for sub-group 551), or moderate (11.9% for sub-group
631, 13.8% for sub—group 632).

87

6. Evaluation of AAS response with respect to individual

commodities at the 5-digit level of the CST classification

As stated earlier, the evaluation of the AAS response

to tariff preferences should be undertaken preferably with
respect to highly disaggregated commodities (e.g., commodities
at the 5-digit or 7-digit level). This section presents an
evaluation of the AAS response for 29 commodities at the
5—digit level of the CST classification. Due to lack of
data, the evaluation methodology developed in chapter 11
could not be applied to these commodities. Consequently, it
was necessary to develop a modified evaluation methodology

applicable to the available data.

The commodities

The 29 commodities to be evaluated are shown in
Table 8. Fifteen of the commodities belong to CST Group 0,
and the remainder to CST Groups 1, 4, 5 and 6. Together,
these commodities account for close to 80% of the total value
of manufactured exports from the AAS.1 The only major ex—
ports not included in Table 8 are those for which the common
IEEC external tariff on imports from "Other LDCs" is equal or

Close to zero (e.g., CST 013.30, CST 081.30, CST 243.31,

CST 431.42).

¥

1These commodities were selected on the basis of a
tabulation of total AAS exports for each commodity exported
by the AAS. The computer print-outs may be made available
on request.

88

TABLE 8.--Major 5-digit level commodities exported by
the AAS to the EEC

 

 

 

 

 

Common EEC
tariff im-
CST Descri tion posed on
Code p imports from
"Other LDCs"
1964
011-10 Meat of bovine animals, fresh, 20.0
chilled or frozen
013-80 Other prepared or preserved meat or 26.0
' meat offals
031-10 Fish, fresh, chilled or frozen 10.7
031-20 Fish, dried, salted or in brine: 15.0
smoked fish
031-30 Crustaceans and molluscs, fresh, 16.4
frozen, salted or dried
032-01 Fish, prepared or preserved, n.e.s. 24.6
042-20 Rice, semi-milled or wholly milled 15.5
053-50 Fruit juices, vegetable juices, 21.2
unfermented
055-45 Tapioca 25.6
061-10 Sugars, beet and cane, raw, solid 80.0
061-50 Molasses, whether or not decolorized 57.2
072-31 Cocoa paste, whether or not defatted 25.4
072-32 Cocoa butter (fat or oil) 20.0
081-20 Bran, sharps and other similar 16.0
residues
112.40 Spirits 107.0
(421.40 Groundnut oil 7.0
422-20 Palm oil 7.0
422-40 Palm kernel oil 7.0
‘122-90 Fixed vegetable oils, n.e.s. 7.0
L“

 

 

89

TABLE 8.--(Continued)

 

 

 

 

 

Common EEC
tariff im-
CST Descri tion posed on
Code p imports from
"Other LDCs"
1964
'561-29 Superphosphates, other phosphatic 6.0
fertilizers‘
599-51 Starches, inulin 28.1
611-40 Leather of other bovine cattle, and 9.5
equine leather
631-10 Wood sawn lengthwise, sliced or 8.1
peeled, 5 mm or less
631-21 Plywood or blockboard 14.9
632-89 Other articles of wood, n.e.s. 13.2
655-61 Twine, cordage, ropes and cables 13.0
656-10 Sacks and bags, of textile materials, 21.1
for packing
657-80 Plaiting materials bound together 10.2
684-10 Aluminium and aluminium alloys, 9.0
unwrought

 

Source: Statistical Office of the European Communities (2).

 

9O

Tariff duties for the 29 commodities range from 6.0%
(CST 561.29) to 107.0% (CST 112.40). For the majority of
the commodities, the tariff range is 10% — 30%. Thus,.
tariff preferences could be expected to have a moderate to

high impact on AAS exports to the EEC.

Evaluation methodology

The evaluation of the AAS response to tariff pre-
ferences for 29 individual commodities will include two,
major parts: first, a description of the findings derived
from an examination of EEC import data and AAS export data,
and second, a presentation of the results of two regression

analyses.

Regression analysis on EEC imports. The first re-
gression analysis pertains to the relationship between the
growth of EEC imports from the AAS and the price competi—
tiveness afforded the AAS by tariff preferences. This
analysis makes use of the "control" group "Other LDCs" as a
proxy for AAS export performance in the absence of tariff
preferences. It isolates the impact of tariff preferences
from the impact of differential changes in the C.I.F. price
competitiveness of the AAS and "Other LDCs" by using the
multiple regression technique. This approach was adepted
because lack of data prevented the use of the group "Other

DMECs" in order to control for differential changes in price

91

competitiveness.

Under the usual assumptions of product homogeneity
and perfect competition, it may be stated that, for a given
commodity, the difference between the growth of EEO imports
from the AAS and that of imports from "Other LDCs", (Gl - G2),
is a function of EEC tariff preferences and the difference in

2

price competitiveness between the AAS and "Other LDCs". The

above statement may be expressed as follows:
(G1 - G2) : f(Tv(GPl - GP2))

where:

T : tariff preferences (i.e., level
of duties imposed on imports from
non-beneficiary countries)

G 2 Growth rate of the C.I.F. price
of EEO imports from the AAS
G 2 Growth rate of the C.I.F. price
of EEC imports from "Other LDCs".
(GP - GP ) measures the difference in pasig price compe-

l 2
tiveness (not including tariff preferences) between the AAS

 

1The evaluation methodology deve10ped in chapter II
is preferable because it controls for additional factors not
taken into account by the methodology used in this section
(e.g., factors such as bad weather which constrain the suppl
of agricultural products used as inputs to the food industry .

2Since changes in EEC supply and demand conditions
apply equally to the AAS and "Other LDCs", they need not be
taken into consideration when estimating (G - G ). Similarly,
changes in supply and demand conditions in J'Otheg‘ DMECs" need
not be taken into consideration.

92

and "Other LDCs". Given the schedule of EEC cummulative
tariff reductions in favor of the AAS (see Table l), the
EEC common external tariff imposed on imports from "Other
LDCs" and (GP1 — GPZ), one may estimate the difference in

total price competitiveness between the AAS and "Other LDCs"

when tariff preferences are taken into consideration,

(GP1 - GP2)D'
The above relationship may then be rewritten as
follows:
(G -G>=f[[(G -G ) —(G —G )MG ~01]
l 2 P1 P2 D P1 P2 P1 IE

The first term between brackets in the right-hand side of the
above relationship is the additional price competitiveness
afforded the AAS by EEC tariff preferences, while the second
term between brackets is the difference in.b§sig price com-
petitiveness.

Let us define:

G -G - G -G =X
[( P1 92%) ((131 1323 1
G —G :X

P1 P2 2

(G1 - G2) = Y

In order to determine whether Y (difference in growth
of EEO imports from the AAS and "Other LDCs") is a function

of basic price competitiveness (X2) as well as additional

 

price competitiveness due to tariffs preferences (X1), the
following multiple linear regression equation may be

estimated:

Y1 : 30 + leil + B2Xi2 + 8i

93

where Y, X1 and X2 are defined as above, and 1 refers to the
ith commodity included in the sample. In the present ana-
lysis, the value of i ranges from 1 to 29. If the regression
coefficient 31 is found negative and significant, it would
indicate that the AAS exports had increased in reSponse to
the EEC tariff preferences.l

Let us now show how X1 may be estimated on the basis
of the growth rate of the C.I.F. price of EEC imports from
the AAS, the EEC tariff imposed on non-beneficiary countries,
and the schedule of EEC tariff reductions on EEC imports from
the AAS.

The reduction of tariff duties imposed on imports by
the EEC from the AAS reached 100% of the duty rate on
1 August 1968. This is also the same date at which the in—
ternal tariff reduction of the EEC reached the 100% target
agreed on. Table 1 provides the schedule of individual re—
ductions, the cummulative reductions and the percentages of
duty rates which applied to manufactured imports from the
AAS'during the 1959-1968 period.

Table 1 shows that the average cummulative reduction
in 1962 - the base year for this study - is 45%. The cummu-
lative reduction for the year 1969 is 100%. Therefore, the
Percentage of the duty imposed on imports from the AAS de-

creased from 55% in 1962 to 0% in 1969.

E

1It would be logical, in this case, for 3% also to
be :negative and significant.

94

It will now be shown how, on the basis of the above

data, to estimate the growth rate of the price of imports

from the AAS when tariff preferences are taken into considera-

tion.

Let us

0’ 7

0’ 7

define:

C.I.F. prices of imports from the AAS at
year 0 (1962), and year 7 (1969).

the EEC common external tariff on imports
from "Other LDCs". T is the average ex-
ternal tariff for the 1962—1969 period.
fraction of the EEC common external tariff
which is imposed on imports from the AAS at
years 0 and 7 (i.e., .55 and 0).

growth rate of the price of imports from
the AAS when tariff preferences are not
taken into consideration (i.e., growth rate
of the C.I.F. import price).

growth rate of the price of imports from
the AAS when tariff preferences are taken

into consideration.

The price at year 7, including tariff duties, is

giVen by the following relationship:

P7(l + d7.T) : PO(1 + dO.T)(1 + r;)7

Taking the logarithm of the above relationship gives:

L98 P7 + Log(1 + d7.T) 2 Log PO + Log(1 + dO.T) + 7L0g(1 + r;)

95

Rewriting gives:

LogP - LogP Log(1 + d7.T) - Log(1 + dO.T)
+

0
7 7

The first term of the left-hand part of the above expression

7

 

: Log(1 + I?) ..(1)

is equal to Log(1 + r1) since we have:

P7

Log P7 : Log PO + 7Log(l + r1)

Po” + r1)7

and therefore: Log P _ Log P

7 0

Log(1 + r1) :

 

7
Equation (1) may then be rewritten:

Log(1 + dZ.T) - Log(1 + dO.T)

 

 

Log(1-t r1) + = Log(1 + r;)
7
Since (1.7 = O and d0 = .55, then:
Log(1 + r1) - Log(1 ; '55T) = Log(1 + r;)
An estimate of r; may be obtained once rl and T are

known.

When using an average tariff1 for the whole period
of analysis (i.e., T constant over the whole period), the
growth rate of the C.I.F. price of imports from "Other LDCs",
r2, is the same as the growth rate of import prices when

*

lTariff duties did not vary much over the period of
analysis. Thus, the use of an average tariff in place of
lIldividual tariffs for each year should not affect the
f1ndings of the analysis.

96

. . . . 1
tariffs are taken into cons1deratlon, ré.

Given r1, r; and r2, we can obtain Xl from the
following relationship:

._ _ _ _ —— , _ _ :3 I _
X1 ‘ [(GP GP )D (GP GP )1 ” r1 r2 r1 + r2 r1 r1
1 2 l 2
If T was not assumed to be constant over the period
of analysis, the functional relationship for X1 would be

much more complicated, and would include r2 as well as rl

and T.

Regression analysis on AAS exports. A second re-
gression analysis pertains to the relationship between the
growth of AAS exportsto the EEC and the level of EEO tariffs
imposed on non-beneficiary countries. Since tariff pre-
ferences should make the AAS relatively more competitive on
the EEC market than on "Other DMECs" markets, the difference
between the growth rate of AAS exports to the EEC and that
of AAS exports to "Other DMECs", (G3 - G4),shou1d be a
function of the level of the EEC tariff imposed on non-
beneficiary countries. In addition, (G3 - G4) should be a

function of the change in supply and demand conditions in

1This may be shown as follows:
_ . 7
P7(l + T) — FO(1 + T)(l + r2)

 

Log P7 + Log(1 + T): Log PO + Log(1 + T) + 7Log(l + r2)
Log P - Log P ,
7 O = Log(1 + r2) = Log(1 + r2)
7
Thus r2 = r2

97

the EEC and "Other DMECs" over the period of analysis.1 The
above considerations lead to the definition of the following
functional relationship:

(G3 " G4) = f[T'( 68D)EEC’( 68D)"Other DMECs"]
where:

p3
II

EEC tariff imposed on imports from non—
beneficiary countries
( CSD)EEC,( 68D)"Other DMECs" 2 change in demand and
supply conditions in, respectively, the
EEC and "Other DMECs".
If it is assumed that ( 68D)EEC and ( 68D)"Other DMECs" are
equal in terms of their impact on AAS exports to each group
of countries,2 (G3 - G4) becomes a function of T only. Thus,
in order to determine whether tariff preferences had an
impact on AAS exports to the EEC, the following regression
equation may be estimated:
((43—04) = a+BTi+ 8i
A significant and positive 3 will indicate that tariff pre-

ferences had a positive impact on AAS exports to the EEC.

 

1Changes in supply conditions in the AAS need not be
taken into consideration since they apply equally to the
EEC and "Other DMECs".

2The growth rates of EEC imports from LDCs of groups
of manufactured commodities which include the 29 commodities
analysed in this section are generally close to those of im-
ports by the 21 major DMECs (including the EEC countries)
over the 1962—1969 period (see UNCTAD (4), Tables 4 and 8).
Thus, the above assumption may not be unrealistic.

98

Findings for individual commodities

The findings for individual commodities are derived
from a descriptive analysis of EEC imports (from the AAS
and "Other LDCs")and of AAS exports (to the EEC and the
rest of the world).

The analysis of EEC imports is based on data in
Tables 9 and 10. Table 9 provides the growth rate of the
value of EEC imports from the AAS and "Other LDCs" (column 2),
and the growth rate of the market share of imports from the
above two groups of countries (column 4). It also provides
the growth rate of the C.I.F. price of imports from the AAS
and "Other LDCs" (column 6), and the growth rate of the price
of imports including the tariff duty (column 7). The growth
of value, market share, and prices of imports from two other
groups of countries — the EEC, "Other DMECs + socialist
countries" - are also included for information purposes.

Table 10 summarizes the findings from Table 9.
Column (1) provides the difference between the growth rate
of the value of imports from the AAS and that of imports
from "Other LDCs", (Gl - G2). In 18 cases out of 29 (62% of
all cases), the difference in growth rates is negative.

Columns (2) and (3) provide the difference between
the growth rate of the C.I.F. price of imports from the AAS

and that of imports from "Other LDCs", (GP - GP ),and the
2

1
difference in the growth of import prices including tariff

Preferences, (GP - GP )D. In 17 out of 29 cases (59%), the
1 2

‘l|‘{

 

 

99

TABLE 9.-- EEC imports of 29 major commodities from the

AAS,

(1962 - 1969)

"Other LDCs", EEC, and "Other DMECs + soc. countries"

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1 = AAS
2 = "Other LDCs"
3 = EEC
4 = "Other DMECs + soc. countries"
Import value Market share C.I.F. price Growth
rate
CST °f
code Mean Growth Mean Growth Mean Growth (CIF
($ rate rate rate price+
1.000) (%) (%) (%) ($/t) (%) duty)
(1) (2) (3) (4) (5) (6) (7)
011.10 1 1,426 2.39 .38 -12.16 928 .07 - 1.41
2 90,694 7.40 24.40 - 7.94 589 5.41 5.41
3 150,475 22.01 40.49 3.81 983 6.77 5.19
4 129,072 16.82 34.73 .72 856 2.04 2.04
013.80 1 1,881 9.83 3.29 - 8.74 902 .12 - 1.77
2 5,886 13.26 10.30 - 5.87 597 3.39 3.39
3 25,743 26.12 45.06 4.81 1,127 5.53 3.53
4 23,625 15.74 41.35 - 3.82 597 .13 .13
031.10 1 1,408 - 1.14 .86 -10.23 442 .53 - .29
2 6,918 17.60 4.21 6.74 502 5.12 5.12
3 49,130 15.98 29.91 5.26 445 5.72 4.86
4 106,787 7.20 65.02 - 2.71 363 6.48 6.48
031.20 1 37 -18.88 .07 -20.32 1,762 29.11 27.66
2 705 10.53 1.37 8.00 496 10.53 10.53
3 14,771 3.94 28.81 1.68 379 7.73 6.53
4 35,764 1.67 69.75 - .78 528 - .33 - .33
031.30 1 1,368 53.81 3.37 35.54 1,736 5.03 3.74
2 5,037 7.42 12.42 - 5.36 1,195 8.68 8.68
3 12,598 9.60 31.06 - 3.44 171 8.54 6.10
4 21,556 15.50 53.15 1.76 557 6.65 6.65
032.01 1 6,605 4.93 .7.51 2.77 930 - 4.16 - 5.88
2 18,158 - 6.21 20.66 - 8.15 726 1.56 1.56
3 4,686 18.48 5.33 21.18 534 7.28 5.35
4 58,449 3.28 66.50 .99 749 5.99 5.99

 

 

100

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 9.-- (Continued)

Import value Market share C.I.F. price Growth

rate

CST Of

code Mean Growth Mean Growth Mean Growth (CIF
($ rate rate rate price+

1.000) (%) (%) (%) ($/t) (%) duty)

(1) (2) (3) (4) (5) (6) (7)

042.20 1 3,817 - 8.30 16.27 -10.80 278 - .30 — 1.49
2 10,150 -15.07 43.25 -17.37 120 3.02 3.02

3 5,340 36.98 22.76 33.09 181 7.28 6.00

4 4,159 39.27 17.72 35.38 171 8.56 8.56

053.50 1 1,732 2.52 3.34 -10.02 255 - 7.12 - 8.59
2 13,927 16.81 26.83 2.52 236 3.04 3.04

3 18,511 20.61 35.67 5.86 210 7.86 6.15

4 17,728 7.29 34.16 - 5.83 300 - 1.85 - 1.85

055.45 1 1,184 - 1.87 79.09 - 3.99 194 2.33 .42
2 183 - .66 12.22 - 2.82 138 - 1.27 - 1.27

3 120 38.82 8.02 35.82 293 -lO.48 -12.15

4 10 78.12 .67 118.85 233 -20.32 -20.32

061.10 1 3,374 -10.37 4.01 - 3.39 117 -13.37 -17.77
2 73,963 - 6.89 81.25 .38 164 - .64 - .64
3 3,039 1.38 3.61 9.25 106 1.23 - 3.91
4 3,816 -17.60 4.53 -11.16 76 -14.81 -14.81

061.50 1 384 32.42 1.99 30.28 30 - 1.22 - 4.65
2 10,035 .39 51.91 - 1.21 30 - 2.13 - 2.13

3 3,326 - 2.06 17.20 - 3.62 41 - 2.69 - 6.07

4 5,588 3.49 28.90 1.84 32 — 2.94 - 2.94

072.31 1 861 109.39 55.84 28.25 319 19.00 16.80
2 85 - 2.70 5.51 -40.38 173 8.92 8.92

3 544 39.20 35.28 -14.75 313 20.69 18.46

4 52 37.52 3.37 -15.53 205 2.19 2.19

072.32 1 7,690 28.85 24.14 1.00 1,264 7.17 5.58
2 6,683 37.59 20.98 7.85 1,168 6.90 6.90

3 15,509 24.01 48.69 - 2.79 1,260 6.15 4.58

4 1,972 12.16 6.19 -12.06 1,192 6.90 6.90

081.20 1 805 26.75 1.60 17.21 54 .62 - .59
2 32,975 10.44 65.46 2.21 56 .45 .45

3 7,710 14.60 15.31 6.07 64 1.93 .71

4 8,881 - 6.29 17.63 -13.27 66 - 2.28 - 2.28

 

 

101

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 9.-- (Continued)
Import value Market share C.I.F. price ngzzh
CST Mean Growth Gro th G th (CIF
code Mean w Mean row .
($ rate rate rate price+
1.000) (%) (%) (%) ($/t) (%) duty)
(1) (2) (3) (4) (5) (6) (7)
112.40 1 361 2.24 .50 - 7.61 476 1.20 - 5.23
2 12,938 2.53 17.94 - 7.33 523 .48 .48
3 21,187 12.70 29.37 1.86 1,275 .46 - 5.97
4 37,641 12.47 52.19 1.65 1,035 - 3.05 - 3.05
421.40 1 30,315 15.64 60.08 3.88 351 - 5.07 - 5.58
2 12,422 4.19 24.61 - 6.41 279 .44 .44
3 4,224 24.11 8.37 11.48 327 - 1.08 - 1.61
4 3,500 15.52 6.94 3.80 290 - .22 - .22
422.20 1 4,930 - 3.71 8.69 - 3.56 226 - 3.77 - 4.29
2 48,734 3.98 85.85 4.15 215 - 3.53 - 3.53
3 2,767 2.96 4.88 3.13 262 - 1.45 - 1.98
4 327 -58.42 .58 -55.59 164 -12.52 -12.52'
422.40 1 1,067 66.41 12.24 46.06 287 1.43 .88
2 4,934 6.96 56.63 - 6.15 277 3.42 3.42
3 2,549 13.32 29.25 - .63 281 3.82 3.26
4 164 -12.76 1.88 -22.90 328 .45 .45
422.90 1 3,657 -41.81 24.57 -35.52 338 - 4.98 - 5.49
2 4,141 - 3.68 27.82 6.75 384 -11.20 -11.20
3 4,359 26.57 29.28 40.27 310 - .74 - 1.27
4 2,729 14.25 18.33 26.62 341 - .56 - .56
561.29 1 632 19.82 3.23 15.35 24 2.03 1.56
2 2,365 - 5.43 12.10 - 8.95 54 1.23 1.23
3 12,929 1.51 66.14 - 2.27 40 3.66 3.18
4 3,622 15.82 18.53 11.50 45 1.29 1.29
599.51 1 366 - 9.50 3.09 -15.27 123 - .15 - 2.18
2 317 - 8.08 2.67 -14.23 110 6.83 6.83
3 9,841 9.65 83.01 2,34 126 - .52 - 2.54
4 1,331 - 3.56 11.23 -10.00 110 3.33 3.53
611.40 1 345 - 2.90 .76 -14.33 1,102 7.71 6.92
2 4,222 43.81 9.32 26.81 926 - 5.67 - 5.67
3 33,188' 12.93 73.23 - .42 3,124 3.42 2.66
4 7,564 .81 16.69 -11.14 2,137 .93 .93

 

 

102

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 9.-- (Continued)
Import value Market share C.I.F. price Growth
rate
CST °f
Mean Growth Growth Growth (CIF
code Mean Mean .
($ rate rate rate price+
1.000) (%) (%) (%) ($/t) (%) duty)
(1) (2) (3) (4) (5) (6) (7)
631.10 1 2,519 39.00 5.30 24.93 220 .72 .10
2 2,179 49.75 4.59 34.64 593 16.18 16.18
3 28,724 7.93 60.48 - 2.94 1,089 1.17 .55
4 14,073 5.92 29.63 - 4.88 480 - 3.74 - 3.74
631.21 1 3,198 13.32 8.89 - 8.66 341 - 1.90 - 2.99
2 1,512 15.13 4.20 - 7.20 261 1.71 1.71
3 15,589 29.46 43.32 4.35 336 - .12 - 1.23
4 15,687 22.29 43.59 - 1.43 226 2.34 2.34
632.89 1 66 9.10 .58 - 3.34 122 - 4.32 - 5.28
2 176 -14.10 1.55 -23.67 540 - 9.96 - 9.96
3 6,614 15.49 58.32 2.59 422 - 1.30 - 2.29
4 4,485 9.72 39.55 - 2.53 397 - .25 - .25
655.61 1 72 -21.08 .55 -22.18 389 - 6.55 - 7.47
2 368 9.45 2.81 3.77 340 - 8.69 - 8.69
3 10,080 4.58 76.86 - .90 416 - 4.91 - 5.84
4 2,594 '10.49 19.78 4.70 700 - 1.60 - 1.60
656-10 1 451 9.18 1.54 3.75 130 - .68 - 2.23
2 11,619 4.31 39.77 - .84 337 .84 .84
3 10,821 3.83 37.03 - 1.30 398 2.00 .41
4 6,330 9.65 21.66 4.23 307 .36 .36
657.80 1 439 - .99 11.34 - 9.15 2,090 - .30 - 1.07
2 450 28.79 11.62 18.17 1,275 11.51 11.51
3 553 8.01 14.29 - .92 147 2.53 1.73
4 2,429 7.67 62.75 - 1.21 275 4.35 4.35
684.10 1 19,346 .74 8.29 -14.33 479 2.45 1.75
2 3,988 92.25 1.71 63.32 485 3.06 3.06
3 71,391 21.72 30.58 3.51 511 2.09 1.39
4 138,718 16.04 59.42 - 1.32 502 1.20 1.20

 

Source: Growth rates estimated on the_basis of statistics

provided in Tables A-4a and A-4b of the Appendix

 

103

TABLE 10.-- Summary table of the evaluation of AAS response
to tariff preferences for 29 major commodities

 

 

 

 

 

 

 

 

CST Code (oi-CZ) (Giff-513) (Gpl— sz)D (3)2:(2) (G3’G4’
(1) (2) (3) (4) (5)
011.10 - 5.01 - 5.34 - 6.82 - 1.48 + 64.21
013.80 - 3.43 - 3.27 - 5.16 - 1.89 - 21.59
031.10 - 18.74 - 4.59 - 5.41 - .82 - 29.65
031.20 - 29.41 + 18.58 + 17.13 - 1.45 — 48.99
031.30 + 46.39 - 3.65 - 4.94 - 1.29 - 61.45
-032.01 + 11.14 - 5.72 - 7.44 - 1.72 - 20.71
042.20 + 6.77 - 3.32 - 4.51 - 1.19 + 5.39
053.50 - 14.29 - 10.16 - 11.63 - 1.47 - 4.72
055.45 - 1.21 + 3.60 + 1.69 - 1.91 + 22.98
061.10 - 3.48 - 12.73 - 17.13 - 4.40 — 69.15
061.50 + 32.03 + .91 - 2.52 - 3.43 — 96.83
072.31 +112.09 + 10.08 + 7.88 - 2.20 + 94.37
072.32 - 8.64 + .27 - 1.32 - 1.59 — 10.87
081.20 + 16.31 + .17 - 1.04 - 1.21 + 21.44
112.40 - .29 + .72 - 5.71 - 6.43 -128.61
421.40 + 11.45 - 5.51 - 6.02 - .51 + 17.69
422.20 - 7.69 - .24 - .76 — .52 + 4.84
422.40 + 59.45 - 1.99 - 2.54 - .55 -191.16
422.90 - 38.13 + 6.22 + 5.71 — .51 - 22.77
561.29 + 25.25 + .80 + .33 - .47 + 59.20

 

 

104

 

 

TABLE 10.-- (Continued)
(CY-(=2 > (le—Z (G — (3 (352-72) (GB’G4’
CST Code 1 2 PI P2 pl 92):)
(l) (2) (3) (4) (5)
599.51 - 1.42 - 6.98 - 9.01 - 2.03 - 20.64
611.40 - 46.71 + 13.38 + 12.59 - .79 - 97.80
631.10 - 10.75 - 15.46 - 16.08 - .62 - 1.66
631.21 - 1.81 - 3.61 - 4.70 - 1.09 + 17.45
632.89 + 23.20 + 5.64 + 4.68 - .96 + 14.53
655.61 - 30.53 + 2.14 + 1.22 - .92 - 28.67
656.10 + 4.87 - 1.52 - 3.07 - 1.55 - 17.01
657.80 - 29.78 - 11.81 - 12.58 - .77 + .08
684.10 - 91.51 - .61 - 1.31 - .70 - 48.46

 

 

 

 

 

 

 

 

105

AAS is more price competitive than "Other LDCs" when tariff
preferen es are not taken into consideration (pgsig com-
petitiveness). It becomes more competitive than "Other LDCs"
in 21 cases (72%) when tariff preferences are taken into
consideration (£3331 competitiveness). The data suggest
that the AAS performance is to a large extent independent
of its relative competitiveness vis—a-vis "Other LDCs". In
13 of the 18 cases (72%) where (G1 — G2) is negative, the
total competitiveness of the AAS is superior to that of
"Other LDCs". A similar percentage (73%) is obtained in the
11 cases where (G1 - G2) is positive.

The above analysis of EEC imports from the AAS and

"Other LDCs" is summarized in the following table:

 

 

 

 

(Gl - G2) (GPl — GP2)<O (GPl - GP2)D<O
Sign Cases
_ 18 11 cases 13 cases
cases (61%) (72%)
+ 11 6 cases 8 cases
cases (54%) (73%)
29 17 cases 21 cases
TOTAL cases (59%) (72%)

 

 

 

 

 

 

The examination of EEO imports tends to indicate that the
AAS did not respond to the EEC tariff preferences: in the
majority of cases, the growth rate of imports from "Other
LDCs" is higher than.that of imports from the AAS, and AAS
export performance does not seem to be a function of its
relative price competitiveness vis—a-vis "Other LDCs".

An analysis of AAS exports yields similar conclusions.

106

Table 11 provides the rate of growth of the value of AAS ex-
ports to the EEC and the rest of the world (ROW). The group
"ROW" includes DMECs other than the EEC countries, socialist
countries, and non-African LDCs.’ Since the fraction of ex-
ports to the last two sub-groups is very small (less than
1% - see section 4), the growth rate of exports to "ROW"
should be almost equal to that of exports to "Other DMECs".l
Column 5 in Table 10 indicates the difference be-
tween the growth rate of AAS exports to the EEC and that of
AAS exports to "ROW", (G3 — G4). In 18 out of 29 cases,
(G3 - G4) is negative, implying that the market share of
AAS exports to the EEC has declined over the period of
analysis in favor of "ROW". As indicated in Table 11, only
in 8 cases was the growth of the share of AAS exports to the
EEC positive, and only in 5 cases was this growth rate
higher than the growth rate of the share of exports to "ROW"
and "Africa". It can also be seen in Table 11 that exports
to "Africa" grew at a higher rate than exports to "ROW" and
"EEC" in 11 cases out of 29 and that, in 16 cases, the
average share of exports to "Africa" ranks first or second
as compared to the shares of exports to the EEC or "ROW"

(see column 3).

 

1Given the fact that the growth rate for "ROW" should
closely approximate that for "Other DMECs", it was not deemed
necessary to undertake the substantial additional data col-
lection which would have been needed to obtain "Other DMECs"
rather than "ROW" statistics.

TABLE 11.-- Growth
commodities to the

107

BBC, ROW,

rate of AAS exports of 29 major
and Africa (1962 - 1969)

 

Export value

j

Market share

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.38

 

CST Code Desti- Mean Growth Mean Growth
nation rate rate
($ 1,000) (%) (%) (%)
(1) (2) (3) (4)
011.10 EEC 963 25.19 27.82 12.83
ROW 192 39.02 5.54 44.60
AFR. 2,307 29.27 66.64 16.50
013.80 EEC 1,782 10.91 76.12 1.98
ROW 304 32.50 12.99 17.08
AFR. 255 10.25 10.89 2.56
031.10 EEC 353 13.67 51.16 3.96
ROW 129 43.32 18.70 11.15
AFR. 208 14.06 30.14 2.61
031.20 EEC 11 18.91 .27 16.09
ROW 30 30.08 .75 32.23
AFR. 3,964 3.49 98.98 .25
031.30 EEC 751 37.97 76.40 4.80
ROW 100 99.42 10.17 26.48
AFR. 132 23.22 13.43 14.44
032.01 EEC 6,468 4.36 99.37 .04
ROW 14 25.07 .22 18.62
APR. 27 7.38 .41 11.30
042.20 EEC 3,666 4.51 63.32 10.61
ROW 129 9.90 2.22 14.61
AFR. 1,995 24.52 34.46 16.56
053.50 EEC 1,329 5.85 93.39 .60
ROW 12 10.57 .85 6.22
APR. 82 21.13 5.76 13.74
7055.45 EEC 1,028 1.05 98.66 .47
ROW 10 24.03 .96 22.73
APR. 4 21.68 20.62

 

 

108

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 11.-- (Continued)

Export value Market share
Desti- Growth Growth

CST Code nation Mean rate Mean rate

($ 1.000) (%) (%) (%)

(1) (2) (3) (4)
061.10 EEC 2,443 - 20.22 73.41 - 17.79
ROW 683 48.93 20.52 192.71
AFR. 202 46.84 6.07 41.36
061.50 EEC 241 5.53 97.18 - 13.75
ROW 0 102.36 -2.82 513.09
AFR. 7 .00 .00 .00
072.31 EEC 619 110.05 66.50 22.22
ROW 298 15.68 31.50 - 32.69
AFR. 18 2.61 2.00 - 55.60
072.32 EEC 7,159 31.97 73.00 - 1.97
ROW 2,648 42.84 27.00 6.11
AFR. 0 .00 .00 .00
081.20 EEC 649 15.59 46.19 10.42
ROW 669 - 5.85 47.62 - 10.07
AFR. 87 .13 6.19 - 4.36
112.40 EEC 304 4.93 83.52 - 6.97
ROW 1 133.54 .27 412.19
AFR. 59 62.70 16.21 41.04
421.40 EEC 47,083 1.03 96.77 - .10
ROW 72 - 16.66 .15 - 14.07
AFR. 1,501 22.11 3.08 20.70
422.20 EEC 2,582 - 2.81 93.69. - 1.15
ROW 11 - 7.65 .40 - 33.61
AFR. 103 8.62 5.91 10.49
422.40 EEC 1,110 51.38 84.22 - 11.55
ROW 175 242.54 13.28 282.84
AFR. 33 10.41 2.50 - 35.75
422.90 EEC 577 - 1.90 71.23 - .34
ROW 138 20.87 17.04 22.81
AFR. 95 - 13.64 11.73 - 12.26

 

 

 

 

 

 

 

 

5' I l I
Ilr.‘ll|ll' l-.“l".(llllll\.lu|

.4 ‘II~'|.,[|I|r

109

TABLE 11.-- (Continued)

 

Export value

Market share

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Desti- Growth Growth
CST Code nation Mean rate Mean rate
($ 1,000) (%) (%) (%)
(l) (2) (3) (4)
561.29 EEC 377 14.90 91.06 4.47
ROW 10 - 44.30 2.42 67.28
AFR. 27 17.96 6.52 7.25
599.51 EEC 254 - 15.04 75.15 5.74
ROW 67 5.60 19.82 17.15
AFR. 17 7.52 5-03 19.26
611.40 EEC 290 4.53 84.80 9.00
ROW 20 102.33 5.84 76.16
AFR. 32 31.76 9.36 14.70
631.10 EEC 1,557 24.10 33.30 .81
ROW 3,070 25.76 65.67 .79
AFR. 48 9.45 1.03 12.72
631.21 EEC 2,503 15.28 35.04 10.91
ROW 3,559 - 2.17 49.83 5.88
AFR. 1,081 - .31 15.13 4.10
632.89 EEC 15 19.91 41.67 4.71
ROW 7 5.38 19.44 12.52
AFR. 14 17.82 38.89 2.88
655.61 EEC 49 - 30.36 25.93 30.50
ROW 4 - 1.69 2.11 3.73
APR. 136 11.22 71.96 10.99
656.10 EEC 241 - 18.66 44.63 21.22
ROW 83 - 1.65 15.37 4.74
APR. 216 45.36 40.00 40.79
657.80 EEC 434 .76 78.06 .42
ROW 47 .68 8.45 .50
AFR. 75 2.14 13.49 .94

 

 

110

 

 

 

 

 

 

 

 

 

TABLE 11.-- (Continued)
Export value Market share
Desti- Growth Growth
CST Code nation Mean rate Mean rate
($ 1,000) (%) (%) (%)
(l) (2) (3) (4)
684.10 EEC 14,787 - 6.03 96.80 .28
ROW _472 42.43 3.09 72.75
AFR. 17 - 13.55 .11 - 14.04

 

Source 3

provided in Table A-4c of the Appendix.

Growth rates estimated on the basis of data

 

111

In summary: the descriptive analyses of EEO imports
and AAS exports tend to indicate that, for the 29 selected
commodities, the AAS failed to respond significantly to the

EEC tariff preferences.

Findings from regression analyses

In the first regression analysis, based on the
values of Yi’ Xi1 and X12 in Table 10, the following
estimates of the regression coefficients were obtained:

Y = 6.44971 - 4.55611X1 + 0.23616X2

(t = -.84566)(t = +.24995)

Multiple R = 0.16832
Both regression coefficients are non-significant. Thus,

neither basic price competitiveness nor additional com-

 

petitiveness afforded by EEC tariff preferences explain
the AAS export performance vis-a-vis the EEC over the

1962-1969 period. It may therefore be concluded that,
with respect to the 29 major manufactured commodities

selected for this analysis, there is no evidence of a
positive AAS response to tariff preferences.

The second regression analysis provides further
support for the above conclusion. Using the values for
T from Table 8 and the values of (G3 — G4) from Table 10,
the following estimates of the regression coefficients
were obtained:

(G3 - G4) = .427 — .953T

(t = -2.0575. p<.05)

112

Thus, the difference between the growth rate of AAS exports
to the EEC and that of AAS exports to "ROW" is, contrary to
expectations, negatively correlated with T. Since the
above regression equation does not control for differential
changes in demand and supply conditions in the EEC and in
"ROW", the result of this analysis should be considered as
an additional, yet not fully conclusive, indication that
EEC tariff preferences did not have a positive impact on

AAS exports.

7. Manufactured commodities exported to groups
of countries other than the EEC

The commodities described in the previous section
represent the main AAS manufactured exports to the EEC.
Other manufactured commodities were exported to the EEC in
very low amounts, with frequent interuptions occuring
during the 1962-1969 period. In other words, these exports
are to some extent incidental, with the EEC playing a minor
role as an importer. The AAS does, on the other hand, ex-
port a large number of manufactured commodities to various
groups of countries other than the EEC, especially Africa.
Although the over-all value of these exports is lower than
that of the exports described in section 6, these exports
cannot be considered as incidental since they took place
each year during the 1962-1969 period, and have tended to
grow at a fairly high rate.

Table 12 provides a non-exhaustive list of manu-

1iictured commodities exported mostly to Africa, and the EEC

113

Africa (1962 - 1969)

TABLE 12.--Manufactured commodities mostly exported to

 

 

 

EEC tariff
CST duty imposed
Description on imports
Code
from
”Other LDCs"
011-60 Edible offals of the animals, 17.9
chilled or frozen
022-10 Milk and cream, preserved, concentra- 23.8
ted, sweetened
046-01 Flour of wheat or of meslin 30.0
061-20 Refined sugar, of beet or cane 80.0
061-90 Sugar syrups, artificial honey, 22.2
caramel
062-01 Sugar confectionery, not containing 27.3
cocoa
111-01 Soda waters 7.1
111-02 Lemonades 16.7
122-30 Beer 30.0
122-20 Cigarettes 180.2
276-30 Common salt 68.9
513-11 Oxygen 7.7
533-32 Varnishes and lacquers 15.4
541-63 Vaccines and serums 11.9
541-70 Medicaments 12.7
553-00 Perfumery, cosmetics and toilet 15.4
preparations
599-20 Disinfectants, insecticides, 12.8
fungicides '
631-83 HOOpwood, piles, pickets and stacks 8.1
of wood
651-42 Cotton yarn, for retail 16.4

 

 

 

 

114

TABLE 12.--(Continued)

 

 

 

. \EEC tariff
duty imposed
CST . .
Description on imports

Code

from

"Other LDCs”

652-13 Woven cotton fabrics, non-finished 15.9
652-29 Woven cotton fabrics, finished 15.9
656-61 Wool blankets 19.0
656-91 Bed linen, table linen, curtains 22.0
674-92 Sheets or plates of iron or steel 5.9
677-11. Iron or steel wire 10.0
694-11 Nails, tacks, staples 12.5
733-11 Cycles 17.0
821-09 Furnitures 16.9
831-00 Travel goods 15.3
841-11 Outer garments, men's and boys' 18.5
841-12 Outer garments, women's and girls' 19.2
841-13 Shirts, men's and boys' 20.0
851-01 Footwear, of rubber 20.0
851-02 Footwear, of leather 19.7
899-32 Matches 14.0

 

 

 

Source: Statistical Office of the European Communities (2),
and source provided in Table 2 for AAS exports.

 

 

 

 

11".4 1.!ll'lllll‘l I

115

tariff duties applying to these commodities. This list in-
cludes 6 commodities in Group 0 (tariff duties ranging from 18%
to 80%), 4 in Group 1 (tariff duties ranging from 7% to 180%),

l in Group 2 (tariff duty equal to 68.9%), 6 in Group 5 (tariff
duties ranging from 7% to 15%), 9 in Group 6 (tariff duties
ranging from 8% to 22%), 1 in Group 7 (tariff duty equal to 17%)
and 8 in Group 8 (tariff duties ranging from 14% to 20%). Thus,
EEC tariff duties on these commodities tend to be moderate to
high, and the AAS could have taken good advantage of tariff
preferences in order to expand their exports to the EEC.

Manufactures in Groups 1, 7 and 8 constitute good
examples of commodities which are exported primarily toward
Africa. As shown in Table 6, Africa accounts for 83%, 60%
and 91% of Group 1, 7 and 8 exports, respectively. High
shares of exports to Africa are also found for commodities
in Groups 0, 5 and 6. The commodities in these six groups
have in common a certain number of characteristics, summarized
below:

(1) Most of them are classified by Lary (3, p.186-210)
as labor-intensive commodities, the only exceptions being re-
fined sugar (CST 061—20), cigarettes (CST 122—20), oxygen
(CST 513—11) and iron or steel wire (CST 677-11). Thus, it
appears that the AAS did not make use of its potential com-
parative advantage by producing and exporting these commodi-
ties to the EEC.

(ii) They are often new commodities, first produced in

the most developed AAS countries during the late fifties or

.5{‘||‘lllllll
‘ llllll'l'i"

116

early sixties. The main exporters are Senegal, Cameroon,
Ivory Coast and Malagasy Republic.

(iii) The value added in the manufacturing of these
commodities is, in general, higher than that of traditional
commodities. In other words, these new commodities are less
natural-resource intensive than are traditional commodities.

(iv) For exports of these new commodities, product
differentiation may be as important, or even more important
than price differentiation. It is possible that the quality
of AAS exports is such that they could not penetrate and ex-
pand in the EEC market, while in African countries, demand
for these exports is particularly high. The quality
variable may be particularly important for goods in CST
Groups 0, 1, 5, 6, 7 and 8. Examples include sugar confec-
tionery (CST O62-Ol), beer (CST 112-3o), cigarettes (CST 122—20),
perfumery and cosmetics (CST 553—00), varnishes and lacquers
(CST 533—32), cotton fabrics (CST 652-13 and 652-29), bed and
table linen (CST 656-91), cycles (CST 733-11), furnitures
(CST 821-09), clothing (CST 841—00) and shoes (CST 851-00).

(v) AAS exports of new commodities grew, in general,
at a higher rate than exports of traditional commodities. The
growth rates of exports in Groups 8, 5, 7 and l, which include
large numbers of new commodities, are 32.33%, 17.50%, 16.76%
and 14.02%, respectively, while the growth rates of Groups 2
and 4, which include mostly traditional exports, are 4.42%

and 3.55%, respectively.

117

A number of conclusions may be derived from the in-
formation presented in this section. First, the AAS did not
take advantage of tariff preferences in order to expand

exports to the EEC of a large number of highly promising
commodities characterised by moderate to very high tariff
duties. Second, the export potential of new commodities seems
to be much greater than that of traditional commodities, as
shown through a comparison of their growth rates. Third, the
quality variable may have constrained expansion of new ex—
ports towards the EEC. This could explain the fact that the
majority of these commodities are exported to Africa rather

than to the EEC.

8. AAS response to EEC tariff preferences: Summary of
findings and concluding remarks

 

It was attempted in this Chapter to determine whether
the AAS responded to EEC tariff preferences by expanding its
exports to the EEC at a relatively higher rate than it would
have done in the absence of preferences. Given the complexity
of the problem, and the lack of several types of data, a
number of evaluation methodologies were developed and applied
to AAS manufactured exports at various levels of commodity
aggregation. Although each analysis has certain weaknesses,
the findings from the various analyses are mutually supportive,
and it is possible to reach a fairly conclusive assessment of
the AAS response.

It was shown in section 2 that, with respect to total

exports, the AAS did not respond to EEC tariff preferences.

118

Indeed, for some unknown reasons, the AAS performed better
in the markets of "Other DMECs" than in those of the EEC
countries. The analysis in section 3 indicated a lack of
response for over-all manufactured exports, although the
findings in this case are less pronounced than in the case
of total exports.

In section 4, AAS exports of various groups of manu—
factures were analyzed, and their growth rates to various
groups of countries were compared. It was shown that three
groups of countries, the EEC, "Other DMECs", and Africa
accounted for close to 99% of total AAS exports. It was
also found that the growth rate of the share of AAS exports
to the EEC was marginally higher (.35%) than that of exports
to the other two groups of countries gply in the case of
Group 0 commodities. For finished manufactures, the growth
rate of exports to Africa was higher than that of exports
to the EEC and "Other DMECs". For semi-manufactures, the
growth rate of exports to "Other DMECs" was the highest.
Thus, the analysis of the various groups of manufactured ex-
ports tends to indicate that the AAS failed to respond signi-
ficantly to EEC preferences.

Section 5 presented analyses of AAS exports of 13
manufactured commodities at the 3-digit level SITC classifi—
cation. When the evaluation methodology developed in
chapterll'was applied to these commodities, it was found that
the AAS failed to respond for 9 out of the 13 commodities.

Furthermore, it was found that the difference between the

119

growth rate of EEC imports from the AAS and that of imports
from "Other LDCS" was negatively correlated with the level
of EEC tariff imposed on imports from non-beneficiary
countries.

-In.section 6, the regression analyses applied to
EEC imports and AAS exports of 29 major commodities at the
5-digit level of the CST classification further indicated
that tariff preferences did not play their expected role.
First, it was found that the additional price competitiveness
afforded the AAS by EEC tariff preferences was not a determi-
nant of the difference between the growth of EEC imports from
the AAS and that of imports from "Other LDCs". Second, it
was shown that the difference between the growth rate of AAS

exports to the EEC and that of exports to "ROW" was negatively

 

correlated with the level of EEO tariffs imposed on imports
.from non-beneficiary countries.

Finally, it was shown in section 7 that a large
number of manufactured commodities in CST Groups 0, l, 5, 6,
7 and 8 were exported almost exclusively to African countries.
Most of these commodities are relatively labour-intensive,
and are subjected to relatively high tariff duties by the EEC
(15% to 30% on average). An expansion of AAS exports of
these commodities to the EEC could be expected for two reasons.
First, the AAS has a comparative advantage in the production
and export of labor-intenSive commodities. ,Second, the re-

latively high EEC tariffs provide the AAS with a substantial

preferential margin on "Other LDCs" in the EEC market. Yet,

120

despite these two potential advantages, the AAS failed to
reSpond to EEC preferences in the case of the above commodi-
ties.

All the above findings tend to yield the same con-
clusion: the AAS failed to respond, in any significant way,
to the EEC tariff preferences. This chapter does not provide
an explanation for this lack of response. However, on the
basis of available data, three potential explanations may be
rejected. First, the lack of AAS response cannot be attribu-
ted to a low level of EEC tariffs imposed on imports from
non-beneficiary countries. As shown in Table 8, the EEC
tariff range for the 29 major commodities exported by the
AAS is 10% — 30%. Second, it cannot be argued that the growth
rate of the price of EEO imports from the AAS was higher than
that of imports from "Other LDCs", thus offsetting the price
Competitiveness afforded the AAS by tariff preferences. As
shown in section 6, the granting of tariff preferences made
the AAS more price competititve than "Other LDCs" for 20 out
of the 29 analyzed commodities (72%). Moreover, it was shown
in the same section that the AAS reSponse was not a function
of either pasig competitiveness or additional competitiveness
afforded the AAS by tariff preferences.

Finally, it could be argued that the lack of AAS res-
Ponse was due to an incapacity to expand exports so as to meet
the growth of EEC demand. .Although the absence of production

Statistics for the AAS and "Other LDCs" prevents verification

121

of this argument, available AAS export data tend to raise
doubts as to its validity. As shown in Table 10, AAS ex-
ports to "ROW" grew at a higher rate than exports to the
EEC in 18 out of 29 cases. Thus, the lack of AAS response
is associated, in part, with a shift of AAS exports from
the EEC to "ROW" over the period of analysis. This finding
tends to indicate that the lack of AAS response cannOt be
attributed to an incapacity to expand exports to meet the
growth of EEC demand.

The above three reasons are the first that come to
mind when attempting to explain the lack of AAS response to
EEC tariff preferences. Since available data do not support
these explanations, one must look for less obvious explana-
tions for this lack of AAS response. Several may be proposed.
First, the manufactured commodities exported by the AAS may
face a low EEC demand because they are not of the type or
quality sought by EEC consumers and producers. In other
‘words, price differentiation may be less important than pro-
duct differentiation in explaining the lack of AAS response
to tariff preferences. This explanation would apply, in
jparticular, to finished manufactures described in section 7.

Another explanation is that the AAS may have neither
'the technological level nor the technical and managerial
skills needed to produce and successfully market manufactured
products in the EEC. The small number of foreign investors

engaged in manufacturing in AAS countries would not compensate

122

for this lack of technical and managerial skills. This ex-
planation would apply both to finished and to semi-finished
manufactures. -

Finally, the Linder similarity of preferences theory,
exposed in chapter 11, could offer an explanation of the lack
of AAS response. A test of a trade model based on this

theory will be undertaken in the following chapter.

123

LIST OF REFERENCES

. Ediafric—Service, "L'industrie Africaine en 1970",
Bulletin de l'Afrique Noire (Numéro Spécial),
Paris: EdiafricéService, 1971 (2 separate volumes

covering each of the 14 African countries included
in this study).

. Statistical Office of the European Communities,
Tariff Statistics-Import—l964—Tab.4-5,
Brussels: Statistical Office of the European
Communities, no date.

. H. Lary, Import of manufactures from less developed
countries,7New York: NBER, Celumbia UnIVersity Press,
1968.

. UNCTAD, Trade in manufactures of developing countries -
1970 Review, New York : United Nations, 1971.

CHAPTER IV

DETERMINANTS OF THE GEOGRAPHICAL INTENSITY 0F AAS EXPORTS

It was shown in chapter III that the AAS response to
EEC tariff preferences must be considered, at best, as weak.
This weak response applies to all types of manufactured ex-
ports, and is especially pronounced for export of finished
commodities, such as labor-intensive commodities in CST
Group 8.

These findings are at odds with those implied by
comparative advantage theories. The AAS, like any other
group of countries, should enjoy a comparative advantage in
the production and export of a number of manufactured commo-
dities (e.g., labor-intensive commodities, low-skill inten-
sive commodities, etc.). Thus, the AAS should respond
positively to tariff preferences by increasing its share of
EEC imports for at least those goods for which it enjoys a
comparative advantage, however this may be defined. Factors
which could conceivably offset the potential impact of
tariff preferences - i.e., low AAS competitiveness on the
EEC market despite tariff preferences (and, more importantly,
despite comparative advantage in production), and/or an
incapacity to expand exports to the EEC - were shown, in
chapter 111, to be largely absent. Thus, the weak AAS res-
ponse to tariff preferences must be explained by other

factors.

124

125

In this chapter, a trade model will be described and
tested, and it will be shown that the geographical intensity
of AAS exports is a function of a number of variables other
than those proposed by the H-0 theory or recent neo-technology
theories. This model is based on the Linder similarity of
preferences theory described in chapter II. One main impli-
cation of this trade theory is that the AAS exports to the
EEC should not be significantly affected by EEC tariff pre-
ferences. Thus, if under testing, the model is shown to be
valid, the weak AAS response to tariff preferences will be,
at least in part, explained.

The first section of this chapter will describe the
model. Section 2 will describe the data used in testing the
model. Section 3 will present and analyze test results for
the AAS as a whole. Section 4 will present a Similar analysis,
but in summary form, for individual countries within the AAS.
Finally, section 5 will provide comments and concluding re—
.marks on the performance of the model in explaining the geo-

ggraphical intensity of AAS exports.

1. The model

According to Linder, trade intensity between two
countries, as measured by the average propensity to import
(API), is primarily a function of the absolute difference in
:per capita income ( IAPCII) between the two countries, with
Other variables playing a lesser role. Thus, a formal

functional statement of the basic Linder model may be expressed

126

as follows:

(API)ij f(' IAPCII ij)

where:

i exporting country

j = importing country
Given the particular characteristics of the countries within
the AAS, the addition of a number of variables to the basic
Linder model should enhance significantly the explanatory
power of the model. These variables are:

(i) Distance (D) between the exporting and

importing countries, in miles.
(ii) Language (L) of the importing and exporting

country. L is a binary variable with a '
value of 1 if the official languages of the
2 countries are the same, and a value of 0
if they are different.

(iii) Special trade agreements (P) between the
exporting and importing countries. P is
a binary variable with a value of 1 if a
special trade agreement exists between the
2 countries, and a value of 0 if no such
agreement exists.

The variable D is used primarily as a proxy for
transport costs. Transport costs constitute a trade-breaking
force, and should therefore be negatively correlated with API.
It would have been preferable to use actual transport costs

instead of distance. Unfortunately, such data are not easily

127

obtained since shipping companies do not usually divulge
their Shipping rates. Furthermore, data on inland transport
costs in Africa, Latin America and Asia do not exist in a
readily available form.

The variable D can also be considered as a composite
proxy variable for both transport costs and the trade horizon
of AAS exporters. The term "trade horizon" refers to the
level of awareness by AAS exporters of market opportunities
outside the home country. It is explicitly assumed by Linder
that as distance to a potential market increases, awareness
of export possibilities, and therefore the level of actual
exports to a market, decreases.

Variable L may be considered as a proxy for special
political and/or cultural relationships which exist between
the importing and the exporting country. This variable is
included in the model because trade flows are greatly in-
fluenced by the existence of such relationships. A common
language facilitates negociations between importers and ex-
porters. Close political links favor the establishment of
Treasures to encourage trade expansion between countries. In
'the context of this study, it is probable that the long-
eestablished political ties between France and individual
czountries whithin the AAS have favored trade between these
countries. Of particular importance are the existing bilateral

Inonetary arrangements between France and its former colonies,

‘

1For details on these arrangements, see International
MOnetary Fund ( 1) .

128

arrangements which introduce a bias in favor of an expansion
of trade between the former and the latter countries. A
Similar bias probably exists in favor of trade expansion
among countries within the AAS since they use a common
official language (French), belong to a common political
entity (the French Community), and, during the 1962-1969
period, adopted the same monetary unit (the CFA Franc).
Finally, the use of the same language may have also favored
trade between the AAS and North African countries, as well
as Belgium and Switzerland.

In the Linder perspective, special trade agreements
may accentuate the trade intensity between countries with
similar soci-economic structures and levels of development,
but should not have a significant impact on the trade in—
tensity between countries which are highly dissimilar in the
above two respects. The variable P is included in the model
in order to find out whether in fact it constitutes a signi-
ficant explanatory variable of trade intensity. If it does
not, this would tend to confirm the implication of the Linder
theory that special trade agreements play a minor role in
expanding trade between highly dissimilar countries.

It would have been preferable to use actual tariff
rates rather than a binary variable. Unfortunately, tariff
Schedules for a large number of importing countries are not
readily available. Furthermore, since the model is tested

for groups of commodities, the estimation of an average tariff

129

for each group of commodities and for each country would

have constituted a formidable task, well outside the scope

of this study. It may be further noted that it is only
intended to find out whether the variable P is significantly
and positively correlated with the dependent variable (API),
rather than to obtain an accurate estimate of the correlation
coefficient in order to predict the impact of trade preferen-
ces. For this purpose, the use of a binary variable should
be adequate.

IAPCII, D, L, and P constitute the independent
variables of the model. Two other variables could have been
included in the model if data had been available. These are
"Income distribution" (1), and "Border" (B).

The variable I was already discussed in chapter 11 in
relation to the Linder theory. It would have been of interest
to include this variable since, according to Linder, median
rather than average income better reflects the demand structure
for imports. Unfortunately, data on income distribution
(e.g., Ginni coefficients) were not readily available for the
Inajority of the countries importing from the AAS. Thus, it
\Nas not possible to include this variable in the model.

The variable "Border" is of great interest and should,
Lif included in the model, increase significantly its explana-
'tory'power. The importance of this variable rests on a parti-
cnilar characteristic of geographical borders in Africa. These

bOrders very often divide homogeneous tribal groups. Thus,

130

consumer goods on both sides of the borders are often identi-
cal in terms of type and quality. This identity of consump-
tion patterns could, and usually does, give rise to an
important exchange of simply manufactured goods, of the
cottage industry type, between populations on both Sides of
the border. It is important to distinguish the variable
"Distance" from the variable "Border". The former is a proxy
for shipping costs while the latter would be used as a proxy
for an identity of consumption patterns of populations on
both sides of a border.

It would be relatively easy to include "Border" as a
binary variable in the model. There is, however, a good
reason for not doing so. It is well known, and even documented,
that border trade in Africa is very largely an illicit type of
trade which is not included in export and import statistics.
Since trade intensity - as reflected by the API — does not
take into consideration actual trade across borders, it was
decided to exclude the variable B from the model.

Given the independent variables IAPCII, D, L, and P,
and the dependent variable (API), it is proposed to test the
following model of the determinants of the geographical
intensity of AAS exports:

fP..
e lj

_ a b c dL..
(API)ij — e ( IAPCIIij) (Dij) e lj

Tflle ordinary least square method is used to estimate the
HKJdel parameters on the basis of the following regression

equation:

 

 

 

 

I it."
[If Ill,

131

Log (API)ij = a + b-Log( thCIlij) + C-LogDi. + dL.. + fP.

J 1:1 13'
where:
i : importing country, i = 1,2, ...... 140
j = exporting AAS country, j = 1,2, ...... l4
(API)ij = average propensity to import

of country i from country j
IAPCIIij = absolute value of the difference
in per capita income between

country i and country j, in 3

Dij = Distance, in miles, between
country i and country j
Lij = binary variable (values of O or 1).

If the official language of country i
is the same as that of country j,

Lij = 1. If it is not the same, Lij : O.
Pij = binary variable (values of O or 1). If

a special trading agreement exists between

country i and j, Pij = 1. If it does not,

P.. = 0.
13

a, b, c, d, and f are the regression coefficients.
In the above relationship, API is obtained from the

fC>Lllowing relationship:
: Exports from j to i (in $1,000) .
GNP of i (in million 8)

104

 

(API)ij

Export values are used insteadof import values for
thee estimation of API because import values may be assessed

in different ways by different countries. Since API values

132

are usually very small, these values are multiplied by 104
in order to avoid a large number of cumbersome zeros. Thus,
API is given in terms of dollars per 10 million dollars of
GNP.

A logarithmic form is used for the regression equa-
tion because, theoretically, API should decrease exponentially
as D and/or IAPCII increase.

In the above regression equation, Log (API) is ex-
pressed as a function of Log(thCID rather than of Log (.APCI).
The absolute value of the difference in PCI's is used in order
to express the argument in the Linder theory that everything
else being equal, the API'S of two countries from a third
country will be equal if their PCI's are either higher or
lower than that of the third country by the same amount.

The sign of the regression coefficients b and c is
expected to be negative,while that of coefficient d is ex-
pected to be positive. The sign of the regression coefficient
f would be positive if special trade agreements have a signi-
ficant impact on the geographic intensity of AAS trade. If,
(N1 the other hand, such agreements have no impact on trade
instensity, the sign of f may be either positive or negative,
bu't statistically non-significant.

The estimation of the model parameters in section 3
iS :not undertaken to predict future values of API's, but to
Verflify empirically the validity of the model. Thus, the

goail is to obtain unbiased estimates of the parameters by

133

including all theoretically specified variables for which
data are available in the regression equation rather than
to obtain a regression equation with the least residual
variance by excluding those variables which decrease the
value of R2. The prediction of future values of API is

outside the scope of this study.

2. The data

Grouping of exports

The Linder theory should apply best to manufactured
consumer goods, and should have little relevance for raw
materials. Its relevance for intermediate manufactures
would depend on the nature of the goods. In general, the
closer a traded commodity is to a consumer good, the better
the Linder theory would apply.

In order to verify the above pr0positions, AAS manu-
factured exports are divided into intermediate goods and con-
sumer goods. The regression equation is estimated for each
of the following 13 groups of exports:

- Total intermediate goods.

- Total consumer goods.

- Intermediate goods in each of the CST Groups

0, 2, 4, 5 and 6.
- Consumer goods in each of the CST Groups
0, l, 5, 6, 7 and 8.
Groups 2 and 4 contain only intermediate goods, Groups

1, 7 and 8 contain only consumer goods, and Groups 0, 5 and 6

134

contain both. The classification of commodities into inter-
mediate and consumer goods is provided in Table A-1 of the
Appendix. Export statistics,1 at the 5-digit level of the
CST Classification, were used in order to estimate the
value of each group of exports from each AAS country.

API's are estimated on the basis of the average

value of exports for the 1962-1969 period, i.e.:

(API) [1 2 ]
. . = - X
13 8 w t iI'
where:
xt = value of exports for year t.

The average value of exports is used instead of
yearly values for two reasons. First, the goal of the ana-
lysis is to verify empirically the Linder theory, and not to
predict the future geographic intensity of AAS exports.
Second, given this goal, it is important to use an average
value in order to minimize export fluctuations of an inci-
dental nature which may distort the findings of the test.

If, for a given group of exports, the value of ex-
ports to a country is equal to zero, the country is not in-
cluded in the test. Obviously, given the economic size of
the AAS countries, it cannot be expected that they will ex-

port all commodities to all countries of the world. For most

 

lExport statistics were obtained from the Statistical
Office of the European Communities (2). These statistics are
available in the form of computer print-outs. Although they
have not been included as an appendix, they can be made
available on request.

135

groups of commodities, the number of countries which do not
import from the AAS is in fact larger than the number of
countries which do. The inclusion of a large number of

zero APIs in the analyses would "hide" the relationship
which may exist between the dependent variable, API, and the
explanatory variables. By excluding zero APIS, the analysis
aims at identifying the factors which determine the existing
distribution of given amounts of exports among the trading
partners of the AAS.

GNP and PCI statistics for each importing and ex-
porting country are provided in Table A-5 of the Appendix.
The values of GNP and PCI used in the analysis are those of
1967. It would have been preferable to use average values
for the 1962-1969 period. Unfortunately, GNP and PCI data
were not available for all years and all countries. The 1967
values should, however, closely approximate the average
values for the period under study.

Distances between importing and exporting countries
were obtained from a publication of the U.S. Navy (3) for
coastal countries, and from rail and road maps for land-
locked countries. In each case, the shortest distance was
selected. Whenever a rail line exists between a port and
the capital of a land-locked country, inland distance was
measured on the basis of the rail line. Distances are re-
ported in Table A-6 of the Appendix.

The value of the variable P (special trade agreements)

136

is equal to l for each pair of countries composed of one AAS
country and one EEC country. Variable P is also equal to l
for various pairs of AAS countries. Twelve out of 14 AAS
countries are grouped within two customs unions: (1) Union
Douaniere des Etats de l'Afrique de l'Ouest, and (ii) Union
Douaniere et Economique de l'Afrique Centrale. The first
customs union includes Mauretania, Mali, Senegal, Upper Volta,
Niger, Ivory Coast, and Dahomey. The second customs union
includes Gabon, Central African Republic, Congo-Brazzaville,
Tchad and Cameroon. A number of countries have left these
unions since 1969. In general, tariff preferences offered
to members of these customs unions are similar to those

granted by the EEC to the AAS.
3. Test of the model for the AAS as a whole

Test findings

The purpose of this section is to test the model
desoribed in section 1 for the AAS as a whole. Thus, j -
the exporting country - covers a range of l-l4, while i -
the importing country — covers a range of 1-140 (140 being
the total number of countries importing AAS manufactures,
including the 14 AAS members).

Table 13 provides the results of the test for the 13

1

export groups described in the previous section. For ease

of presentation, intermediate goods in Groups 0, 5 and 6

 

1The ordinary least square method is used to estimate
the regression equation.

137

TABLE 13.--Resu1ts of the test of the LINDER model for
the AAS as a whole

 

 

 

 

 

 

 

 

 

 

 

 

 

Regression Coefficients Over-all Degr.
CST Regression of
Group Desig- Beta Free-
nation value Weight t p R2 p dom
Consumer Const. 13.956
Goods ' IAPCII - .473 - .216 -5.285 .0005
D -1.161 - .352 -8.879 .0005 .4311 .0005. 499
L 2.374 .313 8.447 .0005
P - .234 - .025 - .686 .5000
Inter- Const. 7.208
mediate IAPCII - .100 - .063 -1.250 .2090
Goods D - .420 - .185 -3.690 .0005 .2019 .0005 493
L 1.413 .270 5.948 .0005
P .654 .104 2.366 .0170
Inter- Const. 5.029
mediate IAPCII - .208 - .135 -1.165 .2460
Goods in D - .096 - .043 - .390 .6970 .0647 .0350 154
Group 0 L .716 .125 1.337 .1830
(O-I) P - .722 - .123 -1.360 .1760
Consumer Const. 11.023
Goods in IAPCII - .444 - .219 -3.969 .0005
Group 0 D - .767 - .271 -5.109 .0005 .3425 .0005 322
(0-C) L 2.279 .327 6.604 .0005
P - .693 - .086 -1.685 .0890
Consumer Const. 8.409
Goods in IAPCII - .667 - .292 -2.401 .0190
Group 1 D - .740 - .259 -2.215 .0300 .4852 .0005 68
L 3.934 .436 4.689 .0005
P .104 .013 .137 .8920
Inter- Const. 6.402
mediate IAPCII - .195 - .125 -2.108 .0340
Goods in D - .356 - .160 -2.735 .0060 .1784 .0005 377
Group 2 L 1.045 .199 3.695 .0005
(2-I) P .773 .132 2.538 .0110

 

 

TABLE 13.--(Continued)

138

 

 

 

 

 

 

 

 

 

 

 

 

 

Regression Coefficients Over-all Degr.
CST Regression of
Group Desig- Beta Free-
nation Value Weight t p R2 p dom
Inter- Const. 6.308
mediate IAPCII - .094 - .056 - .561 .5750
Goods in D - .565 - .259 -2.876 .0050 .3244 .0005 156
Group 4 L 2.466 .422 5.543 .0005
(4—I) P - .299 - .050 - .668 .5050
Inter- Const. 5.873
mediate LAPCII - .382 - .239 -1.288 .2040
Goods in D - .248 - .121 - .628 .5330 .2536 .0090 45
Group 5 L ,1.556 .305 1.920 .0610
(5-I) P - .596 - .117 - .751 .4570
Consumer Const. 8.407
Goods in IAPCII - .627 - .292 -3.335 .0010
Group 5 D - .626 - .235 -2.831 .0050 .4243 .0005 134
(5-C) L 2.789 .371 5.134 .0005 '
P - .151 - .020 - .271 .7870
Inter- Const. 6.173
mediate IAPCII - .473 - .254 -3.117 .0020
Goods in D - .133 - .054 - .687 .4930 .2001 .0005 216
Group 6 L 1.529 .255 3.680 .0005
(6-1) P - .092 - .014 - .199 .8430
Consumer Const. 11.796
Goods in IAPCII - .529 - .258 -5.021 .0005
Group 6 D -1.041 - .363 -7.297 .0005 .5093 .0005 332
(6-C) L 2.066 .290 6.782 .0005
P .205 .026 .615 .5470
Consumer Const. 12.705
Goods in IAPCII - .453 - .235 -4.228 .0005
Group 7 D -1.232 - .457 -8.565 .0005 .5490 .0005 201
L 2.097 .320 6.105 .0005
P - .423 - .059 -1.105 .2700

 

 

TABLE 13.--(Continued)

139

 

 

 

 

Regression Coefficients Over-all Degr.
CST Regression of
Group Desig- Beta Free-
nation Value Weight t P R2 p dom
Consumer Const. 9.967
Goods in. LAPCII - .602 - .280 -4.443 .0005
Group 8 D - .854 - .298 -4.897 .0005 .4972 .0005 221
L 2.679 .357 6.820 .0005
P - .422 - .053 -1.009 .3140

 

 

 

 

 

 

 

 

 

 

 

140

amadmngnated as (O-I), (5—1) and (6-I), while consumer
gmfls.h1these same groups are designated as (O-C), (5-C)

ami(6-C). {Test results are summarized below.

The regression coefficient for the difference in

percmpita income is, as postulated by Linder, negative for

allgmpups of exports. Difference in per capita income is

a shyflficant variable at the .005 level1 for over-all con-

sumer goods and for five of the six consumer goods groups

(O-C, 5-C, 6-C, 7 and 8). It is significant at the .03

level for Group 1. With respect to intermediate goods, the

difference in per capita income is a significant variable

at the .005 level for one group (6-I), and at the .03 level

for a second group (2). Thus, IAPCII is significant at the

.03 level for all groups of consumer goods, and 2 out of 6

groups of intermediate goods.

The regression coefficient for the distance variable
is, as postulated by Linder, negative for all groups of ex-
,ports. The distance variable is significant at the .005

level for over-all consumer goods and for five consumer goods

groups (O-C, 5—C, 6-C, 7 and 8), and at the .03 level for

(}rou;>ll. This variable is significant at the .005 level for
(averuérll intermediate goods and for Group 4, and at the .03

level for Group2. Thus, the distance variable is significant

eat ‘thee .03 level in all cases relating to consumer goods, and

 

1In this section, two levels of significance are used
‘to Cliscxribe the results: .005 level and .03 level.

 

 

lIIj'llll'll‘lIl‘ll

141

in 3 out of 6 cases relating to intermediate goods.

The regression coefficient for the variable language
is, as implied by the Linder theory, positive for all groups
of exports. The language variable is significant at the
.005 level in all cases relating to consumer goods. It is
significant at the .005 level for over-all intermediate
goods, and for intermediate goods Groups 2, 4 and 6. Thus,

the language variable is significant at the .005 level in
all cases relating to consumer goods, and 4 out of 6 cases
relating to intermediate goods.

The regression coefficient for the variable "Special

trade agreement", P, is both positive and significant at the
.03 level for over-all intermediate goods and for Group 2.
In the remaining 12 cases, P is non-significant (with a
positive sign in 2 cases and a negative sign in 9 cases);
The value of R2 is relatively high for over-all
consumer goods (.4311), and for most consumer goods groups,
ranging from a value of .3425 for Group 0-0 to a value of
.5490 (for Group 7. The values of R2 for most groups of
intermediate exports are much lower than those for consumer
goods (.4311 versus .2019 for over-all exports, .3425 versus
.0647 for exports in Group 0, .4243 versus .2536 for exports
in Group 5, and .5093 versus .2001 for exports in Group 6).
7he only moderately high R2 for intermediate exports is that
‘elating to Group 4 (.3244).

Given the above findings, it may be concluded that

142

the three Linder type variables ( IAPCII , D, and L) provide

a significant "explanation" of the geographical intensity

of AAS exports for all groups of consumer goods. Further-

more, as postulated, the Linder model does not apply as well

for intermediate goods as for consumer goods. R2 values for

consumer goods are approximately the double of those for

intermediate goods. In the particular case of Group 0, the

R value for consumer goods is approximately five times the
R2 value for intermediate goods within this group.

An additional finding of the test is that the inde-
pendent variable P is non-significant for all groups of

consumer goods, and for 4 out of 6 groups of intermediate

exports. This finding should be considered with some caution,
P being a binary variable rather than the actual tariff level.

However, it is interesting to note that P is not a significant
determinant of the geographical intensity of AAS export of

intermediate goods in Groups 0, 5, and 6 despite the high

duties which are usually imposed on a large number of commo-
tariffs do not seem to

This

dities within these groups. Thus,

have consituted an important trade-breaking force.

finding is implicitly postulated by the Linder theory.

Examples of API values predicted by the
regression emlation for the AAS as a whole

The regression equation can be used to predict the

verage trade intensity between the AAS as a whole and any

 

143

importing country. The predicted estimate is obtained by

first calculating the average IAPCII and average distance

(D) between the importing country and the AAS as a whole,

and then calculating the average API on the basis of the

average values of IAPCII and D, and the values of P and L

which apply. The regression equation may, for example, be
used to estimate the trade intensity between the AAS and

France on the basis of an average IAPCII of 2,338 dollars,

an average distance of 3,529 miles, and a value of l for both

the variables P and L.
API values were estimated for the following three

hypothetical cases:

Case I : IAPCII = 8100 ; D = 500 miles; L = O; P = 0
Case 11 : IAPCII = 82,000; D = 3,500 miles; L = O; P = O
- l; P = 1

3,500 miles; L —

Case III . IAPCII : $2,000; D :
Case I may be considered as being a typical case of

trade between the AAS and a non-French speaking African

country. Case 11 may be considered as being a typical case
of trade between the AAS and a non-French speaking European

country outside the EEC. Case III may be considered as being

a typical case of trade between the AAS and a French speaking

:ountry (France or Belgium) within the EEC.

y

lThe estimated regression equation cannot, however,
? used to estimate the average trade intensity between the
S and individual AAS countries which belong to a customs
1011 since the value of the variable P is not, in this case,

_ique.

144

Table 14 shows that API values for consumer goods

are higher for Case I than for Cases II and III by a very

large margin. API values are approximately 5 to 40 times
higher for Case III than for Case 11, the difference being
The difference

attributed primarily to the variable language.

in APIs between Case I and Case II is so large that predicted

total exports in Case I generally exceeds predicted total

exports in Case II. To illustrate this point, let us assume
that the importing country in Case I has a GNP of 1,500

million dollars - a typical GNP value for an English speaking

African country. Let us also use, for our example, API
values for consumer goods in Group 0 (i.e., 67.5 and 4.0).

What would the GNP of a country in Case II have to be in

order for its imports from the AAS to equal those of the

English speaking African country? This GNP may be obtained

from the following equation:
our = %(1,500) =

Among non-EEC countries within Europe, only the U.K. had,

25,312 million dollars
in

1967, a GN'P higher than 25,312 million dollars, the GNPs

for other non-EEC countries ranging from 546 million dollars
Thus, AAS

for Iceland to 24,143 million dollars for Sweden.
exports of consumer goods in Group 0 are predicted to be

higher for a typical English-speaking African country than
for non-EEC countries in Europe, with the exception of the

Similar calculations would show that AAS exports of

V.K.
onsumer goods to the above typical African country are

145

TABLE l4.--Examp1es of API values predicted by the regression
equation for the AAS as a whole

 

 

 

 

 

 

 

 

API
CST Group
Case I Case II Case III

Consumer Goods in:
Group 0 67.5 4.0 19.6
Group 1 2.1 .1 3.8
Group 5' 5.1 .2 3.2
Group 6 27.1 .5 4.7
Group 7 19.3 .5 2.4
Group 8 _ 6.6 .2 2.0

Intermediate Goods in:

Group 0 32.1 14.4 14.3
Group 2 26.9 7.5 46.2
Group 4 10.6 2.7 17.7
Group 5 13.1 2.6 6.7
Group 6 23.8 4.4 18.7

 

 

146

generally predicted to be lower than AAS exports to EEC
countries.
APIS for intermediate exports are also higher for

Case I than for Case 11 and Case III. The difference is
not, however, as large as the difference for consumer goods.
It may be shown that, in general, predicted AAS exports of
intermediate goods to European countries are higher than
predicted exports to a typical African country outside the

AAS .

4. Test of the Linder model for individual AAS countries
The proposed model of the geographical intenSity of
AAS exports is tested, in this section, for each individual
AAS country and each of the 13 groups of commodities des—
cribed in Section 2 of this chapter. The estimated regression
equations permit more accurate predictions of the geographical
intensity of exports from an individual AAS country than that
permitted by the over-all regression equation estimated in
the previous section. The number of data points for indivi-
dual AAS countries is not, however, always large enough for
testing purposes. Furthermore, for a number of commodity
groups, non-zero APIS relate primarily or exclusively to AAS
exports to African countries. This latter circumstance tends
:o distort the test findings for the following reason. When
on—zero APIS relate primarily to AAS exports to African
Duntries, the IAPCII values associated with these API values

rl’ld to be Clustered within a narrow range. Under these

147

(flicumstances, the distance and language variables may over—
shadow the importance of the variable thCII as a determinant
of trade intensity (i.e., API values may become mainly a
function of D and L).

The variable IAPCII postulated by Linder may not be
tested properly unless the range of values associated with
this variable is fairly wide. In order to avoid the problems
raised above, the Linder model was tested only in cases where
the number of data points was large enough to insure a wide
range of LAPCII values. By observation of the data, it
could be seen that 20 constitutes an adequate number of data
points. Consequently, the test results are provided only in

cases where the number of degrees of freedom is equal to or
larger than 15.
Table 15 provides a list of the commodity groups

selected for testing, and a list of the AAS countries asso-

ciated with each commodity group.

Test findings

Detailed test results for each AAS country are pro-
‘vided in Table A-7 of the Appendix. These results are
ennmmarized in Table 16. For each commodity group, the table
Ixrovides the number of regression equations which are tested,
tlue number of regression coefficients which are non-signifi-
ceuit (defined here as p2,10), as well as the number of
coexfficients which are significant at levels in the following

reunges: .05 to .10, .01 to .05, and .01 or less. The table

148

TABLE 15.--Commodity groups and AAS countries selected
for testing

 

 

 

 

Selected Number of
commodity Selected countries selected
groups countries
Intermediate
Goods:
Total Togo, Cameroon, Central African 13
Republic, Congo, Chad, Dahomey,
Gabon, Ivory Coast, Madagascar,
Mali, Niger, Senegal, Upper Volta
In Group 0 Senegal, Cameroon, Ivory Coast 3
In Group 2 Cameroon, Ivory Coast, Senegal, 12
Madagascar, Upper Volta, Central
African Republic, Congo, Chad,
Dahomey, Gabon, Mali, Niger
In Group 4 Ivory Coast, Senegal
In Group 6 Cameroon, Ivory Coast, Senegal,
Gabon
Consumer
Goods:
Total Togo, Cameroon, Senegal, Ivory .12
Coast, Madagascar, Congo, Chad,
Dahomey, Gabon, Mali, Mauretania,
Upper Volta
In Group 0 Ivory Coast, Senegal, Madagascar, 4
Mauretania
In Group 5 Ivory Coast, Senegal 2

In Group 6

In Group 7

In Group 8

 

Senegal, Ivory Coast, Cameroon,
Madagascar, Togo, Congo, Upper Volta

Ivory Coast, Senegal, Cameroon,
Madagascar

Ivory Coast, Senegal, Cameroon,
Madagascar

 

 

 

.xHoaommfl ecu mo nId mance "mousom

 

 

 

 

149

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

me O N OH OO 5N O m N am mN O O HO ON HH m ON O a O me NO Hosea azamo
ON N O H Om m N N N aN OH O N N ON a H H O m m AH mm .o .mzoo Hayes
OH O N NH OH ON O H O AN m m N ON N O O NN O N O ON em .0 .ezH H4909
O O O O O O O O O O N O H H N H H O H H O N O O moose
O O O O O O H O O N O O O O O O O O H H H H O O moose
A O O O a O H O H m m N O N O H O O H N H m N O moose
N O O O N O O O O N N o O O H H O O H O O H N m asowo
N H O O N N O H O m m H O O N N O O O O H N O O asowo
OH H O H HH H O H H OH O m H O O N O H N H N a NH Hobos
umpoom Headmcoo
O O O O m H H H O N H H H H O O O O O H H N a O moose
N O O O N O H O O H H O O H H H O O O O O N N O cacao
H O O a H HH H O O HH O N H O O N H O O O N OH NH N cacao
H O O N H N H O O N O O O m H O O N O H O N N O moose
s O N O O OH N O O HH N O O OH H H N O O O H NH NH Hobos
“mooom .EumucH
mO. H. O. H. mO. H. O. H. . O. H.
a. 0» on HA EA xvao 0» cu HAS on Oh HAS. 0» on dASMOu on HA
HO.mO. O.nO. O.mO. HO.wO. HO.mO. oson
HanoHuHs Hm Has HHH Hoe oumou mosouo
. . m 3 _Hoa<_ mo soHooEEoo
HHMIHm>o m moocmuommnm ommsmcmq oocmumao Honesz
muHSmou codmmmum. Adv COGMOHMHcmHm "mucoaoammooo GOammoumom
Ion HHMIHC>O .

 

 

 

 

incesszc

Hmoauucsoo med Hospe>flocav HooOE “wooed on» «0 anon on» no muasmmu mo humEEsm II.oH mamma

 

150

almnprovides the number of R2 values which are in the
ramms <.4 and>t4, as well as the number of regression
emunuons with over-all significance levels in the same
ramym as those specified for the regression coefficients.

Results from Table 16 are summarized below.

The total number of regression equations is equal

to 34 for intermediate goods and to 33 for consumer goods.

Due to an insufficient number of observations, intermediate

goods in Group 5 as well as consumer goods in Group 1 are

not included in Table 16.

The regression coefficient for the variable IAPCII
is significant at the .10 level or less in.6 out of 34 re-
gressions performed on intermediate goods (i.e., 17% of the

cases), and in 16 out of 33 regressions performed on con-

sumer goods (i.e., approximately 50% of the cases). Six of

the coefficients associated with these latter goods are sig-

nificant at the .01 level. Thus, as already shown in the

previous section, the variable LAPCII is a more powerful

determinant of trade intensity for consumer goods than for

iirhermediate goods. Results for individual AAS countries

sire, however, less conclusive than those for the AAS as a

IMhole. This is not unexpected since the sample size for

iiuiividual AAS countries is much smaller than that for the

AAS as a whole.
The regression coefficient for the variable distance

(D) .is significant at the .10 level or less in 11 out of 34

151

regressions performed on intermediate goods (i.e., 33% of
the cases), and in 32 out of 33 regressions performed on
consumer goods (i.e., 97% of the cases). 24 of the coeffi-
cients associated with these latter goods are significant at
the .01 level while only 3 coefficients associated with
intermediate goods are significant at this level. Thus,
test results for individual AAS countries confirm those of
the AAS as a whole, the variable distance constituting a
more powerful determinant of trade intensity for consumer
goods than for intermediate goods.

The regression coefficient for the variable language
(L) is significant at the .10 level or less in 10 out of 34
regressions performed on intermediate goods (i.e., 29% of
the cases), and in 26 out of 33 regressions performed on
consumer goods (i.e., 79% of the cases). 18 of the coeffi-
cients associated with these latter goods are significant at
the .01 level while only 5 coefficients associated with
intermediate goods are significant at this level. These

findings confirm those of the AAS as a whole.
The regression coefficient for the variable P (Special

‘trade agreement) is significant at the .10 level or less in

'7 out of 34 regressions performed on intermediate goods

(i.e., 20% of the cases), and in 6 regressions out of 33

pxarformed on consumer goods (i.e., 21% of the cases). Two

(If the coefficients associated with these latter goods are

ssignificant at the .01 level while 6 coefficients associated

152

‘with intermediate goods are significant at this level. Thus,
as shown in the previous section, special trade agreements
do not seem to constitute a significant determinant of the
geographical intensity of AAS exports. Results for inter-
mediate goods are slightly better than those for consumer
goods. Differences in the findings are, however, of minor
importance, and do not suggest any particular trend.

The value of R2 is, in general, higher when the
Linder model is tested for individual AAS countries than
when it is tested for the AAS as a whole. As shown in
Table A—7 of the Appendix, R2 values of .7 and .8 are fairly
common, while R2 values for the AAS as a whole do not exceed
.5. It can be seen from Table 16 that R2 values for inter-
mediate goods are higher than .4 in 10 regressions out of
34 (i.e., 29% of the cases), while those for consumer goods
are higher than .4 for 30 regressions out of 33 (i.e., 91%
of the cases). Thus, as shown in the previous section, the
tested regression equation explains a higher percentage of
the total variation of the dependent variable, API, when
applied to consumer goods than when applied to intermediate
goods.

The number of regression equations having an over-all
significance level of .10 or less is equal to 21 (out of 34)
for intermediate goods, and to 32 (out of 33) for consumer
goods. 30 of these 32 regressions are significant at the .01

level, while 15 regressions are significant at this level in

Ill/Ill III-l ..|

 

153

the case of intermediate goods.
In summary: the findings reported in this section

with respect to individual AAS countries confirm and re-
inforce those pertaining to the AAS as a whole.

5. Concluding remarks on the test of the Linder model

Results of the test, both for the AAS as a whole

and for individual AAS countries, are in concordance with

the Linder theory predictions regarding the geographical

The regression model

intensity of manufactured exports.
For these goods,

applies best to exports of consumer goods.
the regression coefficients are generally significant for
all three Linder type variables: absolute difference in per

capita income, distance, and language. The model is less

successful in predicting trade intensity for intermediate

For these goods, the variables language and distance

goods.
are generally significant, but the main Linder variable,

IAPCII , is often non-significant.
The test results also indicate that the variable P

(Special trade agreement) does not constitute a significant
determinant of the geographical intensity of AAS exports.
These results, which are based on a cross-sectional analysis

of AAS exports, are consistent with the results of the time-
series analysis undertaken in the previous chapter. This

finding concurs with the implication of the Linder theory
that special trade agreements will, at best, reinforce the

trade intensity between countries of similar economic levels,

 

 

 

 

154

and should only marginally affect the trade intensity between
countries of dissimilar economic levels.

It must be emphasized that the test results apply
strictly to the AAS and cannot be generalized to any other
group of developing countries. A number of the AAS countries
are classified by the United Nations among the least developed
of the develOping countries. Obviously, this characteristic
of the AAS countries may have had an impact on the test re-
sults. In other words, if the test were to be performed on
a group of LDCS at a higher level of development than that
of the AAS (e.g., Latin American countries), different re-
sults might be found.

It is, however, suspected that the Linder—based model
would yield positive results if tested for other countries
than the AAS. One form of preliminary evidence to support
this position can be derived from an analysis of world ex-
ports and imports of manufactures, on an inter-regional and
Lntra-regional basis. A partial analysis of such trade is
‘undertaken below for illustrative purposes.

Let us first redefine trade intensity. In the pre-
‘Vious sections of this chapter, trade intensity was defined
as erverage propensity to import, taking into consideration
tine size of the GNP of the importing country. This definition
(xf trade intensity is valid as long as the analysis is limited
to 21 single exporting country, or to a group of exporting

cmnurtries (such as the AAS) which have quite similar GNPS.

155

When the analysis is to include exporters which differ

greatly in size, the GNP of the exporting country must also

be taken into consideration. Obviously, the exports of a

very large country A (e.g., the U.S.A.) to a given country C
(e.g., France) will be much larger than those of a small
country B (e.g., Finland). When comparing exports from A

to C to exports from B to C, the size of the exporting country
should be taken into consideration by dividing API by the GNP
of the exporting country. The new measure of trade intensity,

(WAPI), may be defined as follows:

WAPI .. = exports of j to i
( )lJ (GNP)j-(GNP)i

where:
i = importing country
j = exporting country
(WAPI) = the trade intensity between i and j

weighted by the GNPs of countries i and j

Estimates of trade intensities between various re-
gions of the world are provided in Table 17 for manufactured
goods in SITC Groups 6 and 8 (excluding SITC 67 and 68), for
‘the year 1969. The WAPI values in the Table are multiplied
'by 1015 in order to avoid a cumbersome number of zeros.

First, it may be seen from a vertical comparison of
WAPI values that the highest values are, in all cases, those
assunciated with intra—regional trade (e.g., 50.56 for intra-
.African.trade). Thus, geographical trade intensities for

Lunports are higher for intra-regional trade than for

156

.w>OQM mfl OOHDOm 086m

.man

"conmu sumo Mom moaumwuoum mzw

 

.mcoaumz copes: "gnaw zoz .msma pacemadmsm

I mofiumflumum ucoEmoHo>mQ was moons HmcoHuocuwuaH mo xoonvcmm .mcoHumz confine

tha

smGOHUMZ UwflflfiD uxhow 3mz

.moaumfiumum ucmEmOHm>ma was moons HmcofiumanwunH mo xoonvcmm .mnoaumz cmufico

"ECHM mOHumflfimum 30H“ wvmua "00H50w

 

 

mw.mm vm.ov nm.n~ mh.~ HH.N mm.m mHm< ummmunusom
~m.va mm.mma oo.mm oa.H mm. vm.HH mamfl :uoumoz
vs. ~o.m~ wm.om up. ov.m mw.o mofiumd
oo.o oo.o vo. mm.m mo. mm. moauoam sauna
vv.¢ om.aa mm.m mv.a hv.> on.a mmfluucsou umflHmaoom
om.h vs.qq om.¢ vm.m m~.~ mo.em oaounm :uoummz
mfimd umom wand moauw4 moauma¢ mmauucsoo omousm
Inusom :uoummz :wumq umaaoaoom cumummz

 

 

 

 

 

 

 

 

maloav x Ammsam> Hm<sv

mama .Amm + new

. m + m masons oaHm

moauamcmucfi momma Hmconouumuuca can HoseammuuuwucHnl.nH mqmda

 

157

inter-regional trade. Second, it can also be seen that
geographical trade intensities for exports are, in most
cases, higher for intra-regional trade than for inter-
regional trade (horizontal comparison of WAPI values). The
only exceptions are those for exports from Western Europe,
Socialist countries and South-East Asia to Western Asia.
Obviously, the above analysis is too limited in
scope to permit firm conclusions regarding the general
validity of the Linder theory. Nevertheless, the fact that
intra-regional trade (among countries having fairly similar
socio-economic structures) is much more intense than inter-
regional trade (involving countries that differ widely in
socio-economic structures) is generally in conformity with

the Linder theory.

 

 

 

Jullll!

I‘ll, ‘1

l.

2.

158

LIST OF REFERENCES

International Monetary Fund, African Department
Study Group, Financial Arran ements of Countries

usin the CFA Franc, IMF Sta 'Papers,.volume_YVI,
No. g, Wasfiington D.C.: IMF, 1969.

Statistical Office of the European Commnities,
Foreign Trade-Associates - Yearbook 1959-1966,

(14 velumes, one for each AAS country), and
Foreign Trade - Associates - Yearbook 1967-1969,

volumes‘l and 2{EBrussels: Statistical Office of
the European Communities, 1968—1969 and 1970.

U.S. Department of the Navy, Oceanographic Office,
Distance between Ports, H.O. Pub. No. 151,
waShington D.C.: U.S. Government Printing Office, 1965.

CHAPTER V

DETERMINANTS OF THE AAS EXPORT PATTERN
AND EVALUATION OF TRADE STRATEGIES
FOR THE AFRICAN ASSOCIATED STATES

The evaluation of AAS export performance undertaken
in chapter III provided no evidence of an expansion of AAS
exports in response to the tariff preferences granted by EEC.
The lack of AAS response could not be attributed to low EEC
tariff duties or to an over—all lack of price competitiveness
on the part of the AAS. Moreover, it was shown in chapter IV
that trade intensity between the AAS and other LDCs - especial-
ly African countries - is much higher than that between the
AAS and deve10ped countries. The geographical intensity of
AAS exports was found to be a function of three Linder type
variables: LAPCII, D, and L. The variable "Special trade
agreement", P, did not seem to constitute an explanatory
Variable for trade intensity. Thus, the findings from the
cross-sectional analysis confirm those of the time series
analysis in the particular case of variable P.

This chapter has two main purposes. First, in the
following two sections, several possible explanations for
the lack of AAS response and the findings from the test of
the Linder-based model will be presented. Second, two main
trade strategies for the African Associated States will be
examined in the light of the findings from chapters III

and IV.

159

 

 

III! I will"
0'].

 

 

160

1. Possible explanations for the lack of
AAS response to EEO tariffpreferences

In the absence of relevant data, one can only hypo-
thesize about the factors that explain the lack of AAS res-
ponse. The explantions provided in this section are
associated with certain major characteristics of AAS countries.

With respect to finished manufactures, the lack of
AAS response could be explained by the small size of the AAS

l The Bela Balassa hypothesis (presented in

countries.
chapter II) is that small countries have a comparative
advantage for semi-manufactures and a disadvantage for finished
manufactures. In the case of the AAS, a comparative disad-
vantage in the production and export of finished manufactures
could partly offset the potential impact of tariff preferences,
and thus contribute to the lack of response for such manufac-
tures.

The Balassa argument about the size of countries is
valid as long as a country attracts little foreign investment.
Foreign investors bring in not only financial assets but also
technical skills, advanced technologies, and an extensive
marketing network. Thus, small countries, such as Hong Kong,

could become successful exporters of finished commodities.

Foreign investment may be undertaken for four reasons. First,

 

1The population of the AAS countries ranges from
approximately 500,000 people in Gabon to approximately
6 million people in Madagascar, with an average of 2—3
million (1967 data).

161

investors may wish to minimize transport costs by estab—
lishing themselves close to sources of raw materials. The
distance factor also applies whenever a foreign investor
exports to many countries within the same world region. In
this case, the investor will choose the country within the
region associated with minimum transport costs. A second
reason for investing in a foreign country is to take advantge
of the cheapness of some factor of production (e.g., low
wages) or special fiscal benefit. A third reason is to cir—
cumvent existing trade barriers (e.g., tariffs, quotas),
while being protected by these barriers against outside
competitors. Finally, by investing in a preference-receiving
country, it is possible to export the output to the preference-
granting country and thus avoid the tariffs that would have
been imposed if the goods were produced elsewhere.

The above reasons are not, however, sufficient to
induce foreign investment in a given country. If the main
reason for investment is to take advantage of low wages, the
foreign investor will also need an adequate infrastructure
and a reliable, trained labor force. Otherwise, benefits
derived by low wages could be offset by the costs of im-
proving the infrastructure and training the local labor
force. In the case of large projects (e.g., large mining
projects), these additional costs may be economically
justified. For small to medium size projects - which

encompass many of the consumer goods industries - foreign

162

investors will not generally be willing to finance such
costs.

. If the reason for investment is to decrease shipping
costs to a given region, the regional market must be large
and the transport facilities within the region must be ade-
quate. If these circumstances are not present, foreign in-
vestment may not take place.

Finally, if the reason fir foreign investment in a
given country is to circumvent existing tariff barriers, it
is necessary that the size of the country - in terms of
population - be large and/or the purchasing power of the
people high. If these conditions are not present, dis-
economies of scale or unused production capacity will greatly
diminish investment profitability.

Most of the AAS countries lack the characteristics
needed to attract foreign investment. The majOr exceptions
pertain to the production of raw materials (e.g., Gabon,
Congo), or agricultural materials (e.g., Ivory Coast). The
small size of the AAS countries, the inadequacy of transport
facilities within and between these countries, the inadequate
infrastructure (in terms of service industry, power and
telecommunication networks, etc.) and the low skill level of
the labor force are probably responsible for the low number
of foreign investments in the manufacturing sector. Lack of
foreign investment may constrain expansion of this sector,

and therefore contribute to the lack of AAS response to tariff

163

preferences.

Lacking foreign investment, a country must rely on
indigenous resources. Technical, managerial, and marketing
skills are essential for a successful trade strategy based
on the expansion of manufactured exports to developed
countries. The AAS countries may lack the managerial and
technical skills needed to produce manufactured goods of
the quality standard required by industrialized countries.
They may also lack the skills needed for the development of
adequate marketing channels in these countries. Without
these skills, it is not possible to produce the "research
intensive" commodities — i.e., commodities requiring a high
imput Of research and development as well as highly skilled
labor and management - which are highly demanded in indus—
trialized countries. Thus, the lack of AAS response could
be explained by lack of required skills to produce and market
those manufactures with the greatest demand in the EEC.

With respect to finished goods, it has been hypo-
thesized that product differentiation is more important than
jprice differentiation as a determinant of trade (see chapter'II)
This hypothesis implies that tariff preferences - which lead
to price differentiation - would have a minimal impact on the
export of finished manufactures from the AAS. Thus, the low
expansion of.AAS manufactured exports to the EEC could be
explained either by a lack of specialization in production,

or by’a specialization yielding manufactured goods which do

164

not cater to the tastes of EEO consumers.

Finally, the large diStances between EEC countries
and AAS countries could also explain the weak AAS response.
The test of the Linder-based model of chapter IV provides
empirical backing for the above explanation. Distance, and
therefore transport costs, may constitute a particularly
severe trade barrier for trade between land-locked AAS

countries and EEC countries.

2. Possible explanations for the high trade intensity
within the AAS and between AAS countries and non-AAS
African countries

The Linder-based model tested in the previous
chapter predicts that trade intensity within the AAS and
between the AAS countries and non-AAS African countries
shOuld be relatively high since differences in per capita
income between these countries are generally low, distances
are short to moderate, and, inthe case of trade among AAS
countries, a common language exists. Examination of API
values confirms the predictions of the model.. API values
associated with AAS exports to Africa are generally much
higher than.API values associated with AAS exports to
countries outside Africa.

Several factors hypothesized by Linder could explain
the relatively high intra—African trade intensity. These
factors, already presented in chapter II, may be summarized
as follows: similarity of the consumption patterns of the

trading partners, similarity between imports and exports,

165

entrepreneurial risk—minimizing behavior.

Second, with the exception of land—locked countries,
transport costs between African ports are relatively low
and could favor trade between these countries.

Third, with a very few exceptions, political re-
lations among AAS countries, and between these countries
and non-AAS African countries are good. Such relations have
favored the establishment of a large number of bilateral or
regional trade agreements. Although these agreements have
not generally led to the creation of customs unions or free
trade-areas, they have probably encouraged trade expansion
between African countries. Regional organisations, such as
the Organisation of African Unity, the Economic Commission
For Africa, and the African Development Bank may have also
contributed to the development of trade relations among
African countries.

Fourth, the high trade intensity between the AAS and
.African countries may be the result of explicit or implicit
regional planning. In some cases, the size of African
countries may not be large enough to justify investments in
large industrial projects. In other cases, these countries
may not have the financial means to undertake such projects
even if they were economically feasible. In order to circum-
vent the size and/or financial constraints, African countries
have agreed, in a number of instancea.to plan jointly large

industrial projects. These agreements include provisions

166

about the location of the project, the way the output is to
be divided among the participating countries, and the way
the project is to be financed. They constitute explicit
attempts toward intra—regional planning of industrial.
activities, and, as such, would tend to favor an expansion
of intra-regional trade in a number of manufactured commodi-
ties.

There are other cases where countries undertake
industrial investments without prior consultation with
neighboring countries. However, once these investments
have been implemented, they are implicitly taken into
account by neighboring countries when formulating their own
investment plans. In order to avoid uneconomical duplication
of existing industrial projects, a country may decide to im-
port manufactured goods from countries which are already pro-
ducing them. This type of situation constitutes implicit
intra-regional planning which would favor trade expansion

among neighboring countries.

3. Trade strategies for the AAS

In the previous three chapters, an attempt was made
to explain AAS trade patterns from the perspective of trade
theories. Findings from these chapters may help to clarify
the determinants of geographical trade intensity, but are most
probably of limited interest to policy makers in the AAS
countries. Policy makers would probably be primarily interes-

ted in obtaining answers to the following four questions:

167

(i) What are the export strategies which could be adopted?
(ii) How should each strategy be implemented (i.e., what

are the policy measures needed to implement a strategy)?
(iii) What are the chances for each strategy to succeed? and
(iv) What would the impact of each strategy be with respect
ito economic growth, balance of payments, employment and in-
come distribution?

' While the above questions are legitimate and highly
relevant, it is outside the scope of this study to provide
complete and accurate answers. This section will, however,
make use of limited empirical evidence, conventional economic
analyses and various assumptions about economic relationships,
in order to offer a preliminary evaluation of two contrasting
trade strategies.

Two distinct trade strategies could be adopted by the
AAS. The first is a strategy geared to increasing manu-
factured exports to industrialized countries and, in particular,
to the EEC countries where the AAS enjoys special tariff pre-
ferences. This strategy is backed by some implications of
comparative advantage theories. The second strategy is to
emphasize intra-regional trade in the way implied by the
Linder theory. The purpose of this section is to examine the.
two strategies without making any specific recommendation as
to which should be adopted. Moreover, it should be borne in
mind that the optimum approach may be one which mixes, in

varying proportions, the above two strategies.

‘.

168

Evaluation of an export strategy based on an

expansion of manufactured exports to

industrialized countries

The AAS could attempt to increase manufactured ex-
ports to industrialized countries by expanding the production
of manufactured goods for which they enjoy one type or
another of comparative advantage. Such goods would include,
depending on the comparative advantage theory being considered,
labor-intensive commodities, resource-intensive commodities,
low—skill intensive commodities, standardized commodities,
etc. In order to promote this strategy, the AAS would need
to implement policy measures of the type described below.

Some of these measures have already been partially implemented
by a few AAS countries (e.g., Ivory Coast).

The AAS could expand investment in export industries
by allocating less funds to infrastructural and social in-
vestment projects which are not essential to the export
strategy. Scarce foreign exchange could be used in priority
for the import of equipment, technical services, and material
inputs required by the export industries. Investment in
training programmes could be carried out in order to create
skilled labor needed for the production of manufactured
commodities of the quality required for markets in indus-
trialized countries. Finally, the AAS countries could improve
marketing channels in industrialized countries by develoPing
an efficient network of trade information centers, and by

organizing national trade fairs in order to promote the sale

169

of their manufactured goods.

In the case of consumer goods produced for export,
the AAS countries could specialize in a restricted number
of commodities (e.g., tropical food products, particular
types of textiles and clothing items, etc.). They could
thus use their limited means - financial, technical - in
order to exploit to the fullest extent possible the compara-
tive advantage which they may enjoy in the production of
certain commodities.

If the AAS countries were unable to implement the
above measures, they would need to attract foreign investors
through the granting of various fiscal incentives, guarantees
against business risks (e.g., expropriation), and, most im-
portantly, through the maintenance of a stable political
regime.

Chances for such a strategy to succeed in expanding
manufactured exports to industrialized countries do not'
appear to be very good for the following reasons. The
development of an indigeneous export industry faces serious
obstacles. In order to expand manufactured exports of the
quality required by industrialized countries, the AAS
countries would need to increase the capital-intensity of the
manufacturing sector, absorb new technologies, and create the
managerial and technical skills for designing, producing, and
“marketing products demanded in the markets of industrialized

countries. The deve10pment of the technological and managerial

170

basis needed to produce and market such products cannot be
created overnight. Indeed, a main characteristic of under-
development is the existence of various constraints which
impede the creation of the needed managerial and technical
basis. To promote manufactured exports of the above type,

a country needs to have already reached a relatively advanced
state of deve10pment which does not yet exist in most of the
AAS countries.

It is also doubtful that foreign companies will
make substantial investments in the AAS countries in the
foreseeable future. In the case of finished manufactures,
the main advantage in establishing manufacturing plants in
the AAS countries is the relatively low level of wages.

This advantage is, however, offset by several disadvantages:
inadequate infrastructure, lack of trained labor, and high
shipping costs to Europe. Foreign investors are likely to
prefer investment in North African countries which have a
well developed infrastructure, trained labor force, and are
much more closer to European markets. It can, in fact, be
shown that foreign investments in Tunisia and Morocco have
grown recently at a much higher rate than foreign investments
in AAS countries. Thus far, investments in the AAS have
been concentrated in the processing of raw and agricultural
materials. It is therefore doubtful that the AAS could
expand manufactured exports to industrialized countries by

relying on foreign investors.

171

Few developing countries have succeeded in sub—
stantially expanding manufactured exports to industrialized
countries. Table 18 provides a list of major LDC exporters
of manufactures to the main developed countries for the years
1962 and 1972. As shown in the table, 19 countries account
for over 80% of total LDC manufactured exports to developed
countries. Furthermore, 14 of these LDCs are listed in
both years, meaning that only 5 new countries were able,
between 1962 and 1972, to expand their exports sufficiently
to be included in the list. Similar findings characterize
exports of major groups of manufactures. Table 19 shows
that a small number of countries (3 to 13) account for a
major proportion of exports of individual groups of commodi-
ties (53.85% for drink and tobacco products to 90.27% for
clothing). The only AAS countries included in this table are
Ivory Coast (food products and wood products) and Cameroon
(worked non-ferrous metals).

The above data suggest that successful examples such
as Bong-Kong or South Korea may not be easily duplicated.
GiVen the limited capacity of industrialized countries to
absorb labor-intensive, low-skill intensive commodities pro-
duced by LDCs, only a small number of countries will be able
to succeed in substantially expanding their exports of manu-
factures to developed nations. The AAS countries are unlikely,
in the foreseeable future, to be included among the success-

ful exporters.

172

 

 

 

 

 

 

 

os.oH oo.~ s.mom smsH mm.a m.~oH osausomue
os.o oo.~ H.BQM omonoe + nonsense mo.~ a.oHH coma
om.o~ as.“ m.omn manage aosom ~H.~ m.~HH ssossom
om.s os.~ m.a~q mason a~.~ ~.H~H anon
om.o oo.m o.mse modasuse .suoz eo.m o.HmH omonoe + oooasass
o~.m oa.m m.oae oaaso mm.m m.ama oosxoz
om.o~ ov.m H.m~m ouomoasam as.m o.oo~ ms>oamomsw
oa.m oe.m H.omm oaosoaoz oa.m o.ao~ snowman:
oe.m os.m o.msm masses eo.v H.mH~ oases
om-o om.v ”.mso ososH Hm.m m.aa~ mauomaa
oo.mm oq.v «.mme Hanoum mm.m m.aom masses
om.¢ o~.m m.mam oaosnoso> om.o s.amm mango
oa.oH om.m n.eoa . ooaxoz om.o «.mom oaosH
o~.ma oa.m ~.moo.a oa>oamooss .ma.» H.vae msox meow
os.mo oo.s H.moa.a sosox mo .aom as.m s.ooe noaasuse .sooz
oq.aH oo.ma H.mme.~ macs meow o~.oa e.amm oaosnoso>
~smanwoa~ Houou A.s as suouassou so Houou A.e av suouasaou no
ommouosa «o a auucsoo msauuomxm no a .auussoo mcauuomxm
ommucmouom

omnum>m ascend

 

 

meH

 

Neaa

 

Nsaa oso «was .nomzo am on mousuoomssoe mo msouuomxo some: uu.ma mamas

 

173

 

.mhma .mcodumz wmuaco uxnow 3oz .3ma>mu mnma I mmauouauuma
cam mwauucsou mCaQOHm>ma mo mmusuommscmz ca momma .adeoza "moumom

 

 

 

 

 

 

 

 

 

 

o¢.aa oo.ooa o.s~emn noon Houoa oo.ooa o.sm~.m moan Houos
om.m oa.ma s.eaoau moon mo umom ms.ma n.5mm mung mo boom
om.~a oH.om m.~mmgn moan ma Hmuoa -.mm m.oow.v moan ma Hmuoa
om.w> on.H H.Hm~ mmsmnmm ¢~.H m.mw oooonoz
om.ma oa.H e.aom Hooan ~m.a a.os nossaaaassa
oe.HH oa.H e.Hom osausomue vo.a s.mm Hanoum

thalmmma Hmuou A.E av huouauuou no Hmuou A.E av muouauuou no

mmmmuo:« mo w muucsoo mcauuomxm «o m . knucsoo mcauuomxm
mmmucmoumm
mmmum>m Hmmccm msmd mama

 

 

 

loosssusooc

II.mH mamdh

 

174

TABLE l9.--Major trade flows between developing countries
and 21 DMECs in 1972

 

 

 

 

Major exporting countries Remaining
Export group 3 countries
Name No. of total of total
Clothing Hong Kong, Korea, 7 90.87 9.13
Yugoslavia, Israel,
Singapore, Philippi-
nes, India
Engineering Hong Kong, Mexico, 5 85.35 14.65
products Singapore, Yugoslavia, ‘
(excl. road Korea
vehicles)
Textiles India, Hong Kong, 7 85.43 14.57
Iran, Pakistan, Korea,
Brazil, Yugoslavia
Food products Brazil, Argentina, 12 77.30 22.70
Israel, Yugoslavia,
Morocco, Ivory Coast,
Mexico, Philippines,
Korea, Ghana,
Malaysia, Paraguay
Misc. light Hong Kong, Korea, 8 89.90 10.10
manufactures Mexico, Israel, Singa-
pore, Yugoslavia,
India, Lebanon
Wood products Malaysia, Korea, 8 78.48 21.52
and furniture Yugoslavia, Brazil,
Philippines, Singa-
pore, Mexico, Ivory
Coast
Leather and India, Brazil, Argen- 7 82.42 17.58

footwear

 

tina, Hong Kong,
Korea, Yugoslavia,
Pakistan

 

 

 

 

 

175

TABLE 19.--(Continued)

 

 

 

 

 

 

 

 

 

Major exporting countries Remaining
countries
Export group % %
Name NO' of total of total
Chemicals Mexico, Yugoslavia, 13 72.93 27.07
Brazil, Israel,
Bahamas, Argentina,
Trinidad and Tobago,
Tunisia, India, Korea,
Indonesia, Bermuda,
Neth. Antilles
Iron and New Caledonia, Yugos- 6 87.93 12.07
steel lavia, Korea, Mexico,
Brazil, Dominican
Republic
Worked non- Ghana, Yugoslavia, 5 90.91 9.09
ferrous Cameroon, Surinam,
metals Bahrain .
Drink and Jamaica, Algeria, 6 53.85 46.15
tobacco Mexico, Yugoslavia,
products Cuba, Bahamas
Non-metallic Mexico, Yugoslavia, 4 74.23 25.77
mineral Bahamas, Hong Kong
products
Pulp, paper Brazil, Yugoslavia, 5 87.95 22.05
and board Mexico, Morocco,
Hong Kong
Road motor Mexico, Yugoslavia, 3 89.10 10.90
vehicles Brazil
Rubber Israel, Yugoslavia, 4 81.58 18.42
products Korea, Malaysia
Source: Same as for Table 18.

 

176

Even if this trade strategy could be successfully
implemented, its impact on employment and consumption may
not be socially acceptable to some governments in the AAS.
To produce manufactures of the quality level demanded in
industrialized countries, it would be necessary to use
technologies that are more capital—intensive than they ought
to be given the relative factor endowments of the AAS '
countries.

The technologies needed to produce manufactures for
the European market are likely to be more capital-intensive
than the technologies needed to produce similar goods of a
lower quality for local consumption or intra-regional trade.
A trade strategy based on export expansion to industrialized
countries would generate relatively little employment, and
thus be ill-suited to development plans which place parti—
cular emphasis on employment creation.l Furthermore, policies
designed to implement this trade strategy may constrain the
production of consumer goods of the price and quality which
suit the economic level and tastes of the local population.
Thus, while this trade strategy might be favorable to
economic growth, it would not favor employment generation and
an adequate production of-consumer goods for the local popu-

lation.

 

1Concerning the relationship between choice‘of tech-
nologys product quality and employment, see A.S. Bhalla,
Ed. 1 .

177

Evaluation of a trade strategy

based on intra-African trade

A second trade strategy for AAS countries would be
to expand intra-regional trade within Africa. Policy
measures required for the promotion of this strategy are
described below.1

African countries tend to levy high duties on im-
ports, particularly imports of consumer goods, from neigh-
boring countries. Exceptions exist only among the few
countries which are associated in customs unions. Indeed,
as a result of the reverse preferences granted by AAS
countries to the EEC, duties and/or quotas imposed on EEC
imports are often much lower than those imposed on imports
from the majority of African countries. Expansion of intra-
regional trade requires that these barriers be removed or
lowered. This could be done through the creation of free
trade areas and/or customs unions. If such measures were
not politically feasible, bilateral and multilateral agree-
ments could be established to lower tariffs on imports from
African countries while maintaining higher tariffs on im-
ports from non-African countries.

The lack of convertibility of many African currencies
may impose a severe constraint on the expansion of their

mutual trade. If countries must pay convertible currencies

 

1For a detailed description of measures which may
be adopted by AAS countries in order to promote regional
trade, see UNCTAD (2).

178

for their imports from one another, the volume of trade will
be constrained by their over-all foreign exchange earnings.

To avoid this restriction on trade, governments could enter
into bilateral clearing agreements and/or agree to create a
payment union. Bilateral clearing agreements have been used
in the past by a number of developing countries, and have
been partially successful in promoting mutual trade. However,
for trade expansion to take place among all African countries,
a very large number of such agreements would be needed. It is
therefore doubtful that this approach would be successful in
the long run. An alternative approach would be the creation
of a payment union among African countries. A union would
present fewer problems and be more manageable administrative-
ly. The problems raised by possible trade surpluses and de-
ficits could be solved by establishing rules which set limits
on these deficits and surpluses.

Two additional factors which may hamper the expansiOn
of trade among African countries are the lack of efficient.
marketing channels and adequate transport facilities. The
expansion of marketing channels cannot be achieved overnight,
and, in the short run, trade among African countries may be
constrained by the lack of such channels. Thus, an urgent
task for African governments would be to set up marketing
organizations to seek outlets in African countries,organize
trade fairs, establish contacts between producers and importers,

and facilitate administrative matters. Creation of intra—

African marketing channels is especially urgent in African

179

countries where existing channels are dominated by foreign
firms concerned solely with the expansion of trade with
their home countries.

Lack of transport facilities constitutes a severe
contraint to intra—regional trade expansion for landlocked
countries. Expansion of these facilities would become a
profitable investment once the volume of trade among African
countries has reached a sufficiently high level. In the
meantime, African governments would need to take an active
role in the deve10pment of transport facilities for intra-
regional trade (e.g., creation of new shipping companies,
expansion of the rail and road networks). Agreements would
need to be reached among African countries for the alloca-
tion of the costs of transport facilities, with preferential
treatment extended to small landlocked countries which may
not be able to finance large transport infrastructure pro-
jects.

The chances for an intra-African trade strategy to
succeed are probably better than those associated with the
previous trade strategy. First, the technological basis
needed for an intra—African strategy already exists to a
large extent. African countries usually have the capacity
to produce low-quality, low-priced consumer goods. Large
numbers of small, cottage-type industries already produce
for local consumption and for export to neighboring countries.

Governments need only to re-organize the small-scale industry

180

sector, improve the technical efficiency of small producers,
and provide financial help to small investors. Second, this
trade strategy does not require large amounts of foreign
exchange. For many small-scale industries, little equip-
ment needs to be imported, or simple types of equipment
could be produced locally. Another alternative would be to
import low-priced second-hand equipment from industrialized
countries, and thereby minimize foreign exchange costs.l
Third, a number of trade agreements already exist among
African countries and, in general, there should be no major
obstacle to expanding them. The only opposition would
probably come from very small and poor countries which fear
that a process of economic "polarization" would take place,
favoring producers in the large African countries, and
slowing down their own industrial expansion. The establish-
ment of a payment union and the development of adequate
marketing channels and transport facilities may, on the other
hand, encounter major difficulties. Political will may be
lacking, and financial constraints could limit investment in
transport projects.

Despite the above shortcomings, chances for a sub—
stantial expansion of intra—African trade should be fairly
good. As shown in previous chapters, trade in consumer
goods is already relatively intense among African countries.

In particular, AAS exports of Group 1, 5, 7 and 8 commodities

F

1For example, see 0. Cooper and R. Kaplinski (3).

181

tend to be directed primarily toward African countries

(see chapter III). Moreover, trade within Africa does not
constitute an isolated example of high intra—regional trade
intensity. As shown in the last section of chapter IV,
trade within the various developing regions tends to be

much more intense than inter-regional trade. An outstanding
example of successful economic integration among developing
countries is provided by the Central American Common Market.
.African countries could learn from this example in order to
achieve a similar level of success.

A final argument for the expansion of intra-African
trade is that the tariff preferences granted to the AAS
have lost some of their value since the EEC established its
GSP scheme. It is suspected that a small number of the
countries included in the GSP scheme (in particular those
listed in Tables 18 and 19) will become or remain the major
LDC exporters to the EEC.

An intra—African trade strategy may appeal to AAS
governments concerned about employment and the consumption
needs of their populations. The technologies needed for
the production of low-price, low-quality manufactures are,
in general, relatively labor-intensive. They should thus
favor employment generation and lead to an improvement of
the current income distribution. Furthermore, the price
and the type of commodities produced should be well suited

to the consumption needs of the local population and the

 

1See W.T. Wilford (4)

182

populations in neighboring countries.

One potential drawback of this strategy is that its
over-all impact on economic growth might be lower than that
of the first trade strategy. The choice between the two
strategies involves, in part, a choice between two different
development patterns: a development pattern focussed on
employment generation and the satisfaction of the basic needs
of low—income groups, or a development pattern focussed on
maximization of over-all economic growth with less concern
for employment generation and improvement of income distri-
bution.

In choosing between these two strategies, AAS govern-
mentswould need to consider their respective chances of
success and their socio—economic impacts. Some relatively
developped AAS countries, such as Cameroon, Ivory Coast, and
Senegal, may opt for a mixed strategy since their manufac-
turing sectors are fairly dynamic and they may be able to
specialize successfully in the production of certain commodi-
ties (e.g., tropical food products, wood manufactures) for
export to industrialized countries. On the other hand, small
and relatively poor AAS countries may find it more advantageous
to adopt a strategy emphasizing intra—regional trade.

Detailed investigation of a country's economy would
be needed in order to make meaningful recommendations re—
garding the trade strategy, or mix of trade strategies, which

should be adopted. Although the analyses undertaken in the

185

present study do not permit specific recommendations re-
garding individual AAS countries, they do suggest that, for
the vast majority of these countries, it would be inappro—
priate to place major reliance on tariff preferences granted
by the EEC and other industrialized countries as a basis for

the planning of industrial growth and economic development.

184

LIST OF REFERENCES

. A.S. Bhalla, Ed., Technology and Employment in industry,
Geneva: International Labour Organization, 1975.

. UNCTAD, Trade expansion and economic integration among
develpping countries, Nedeork: United Nations, 1967.

C. Cooper and R. Kaplinski, Second-hand equipment in a
developing country - A study of jute processing in Kenya,
Geneva: International Labour Organization, 1974.

. W.T. Wilford, "The Central American Common Market:

Trade patterns after a decade of union", Nebraska Journal
of Economics and Business, 1973, 13, 3.

 

CHAPTER VI

SUMMARY AND CONCLUSIONS

1. Summary of findings

This study was designed to meet two main objectives.
The first objective was to determine whether the AAS res-
ponded positively to EEC tariff preferences during the
1962—1969 period. The second objective was to identify the
main variables which determine the geographical intensity
of AAS manufactured exports.

A number of time series analyses were undertaken in
order to evaluate the AAS response to EEC tariff preferences.
These analyses were conducted at various degrees of commodity
aggregation, from total manufactured exports to commodities
at the 5-digit level of the CST Commodity Classification.

In general, the analyses indicated a failure of the AAS to
respond to EEC preferences by expanding manufactured exports
to EEC countries at a higher rate than would have prevailed
in the absence of such preferences.

The lack of AAS response applied to over-all exports
as well as to manufactured exports. With respect to indivi-
dual manufactured commodities, the analyses indicated a
positive AAS response for only a small fraction of the total
number of commodities investigated. Most cases of positive
response involved raw materials—intensive semi—manufactures

rather than consumer goods. Moreover, newly produced consumer

185

186

goods (i.e., goods first produced after the granting of EEC
preferences) were exported primarily to developing countries
in Africa.

The analyses showed that the lack of AAS response
could not be attributed to a low growth rate of AAS exports,
or to a lack of price competitiveness on EEC markets. It
was also shown that, for most manufactured commodities, the
AAS held a substantial preferential margin, and the lack of
response could not therefore be attributed to low EEC tariffs.
Indeed, the lack of reSponse was more pronounced for heavily
protected consumer goods than for semi—manufactures which are
subject to a lower level of protection.

Although it was outside the scope of this study to
evaluate the factors responsible for the lack of AAS response,
three hypotheses were considered. First, the AAS may lack
the technological level and managerial skills needed to pro-
duce and successfully market manufactured goods of the type
and quality demanded in EEC markets. Second, the AAS may
have failed to attract foreign investors who could have pro-
vided advanced technology, managerial skills and extensive
marketing channels. Third, product differentiation may be
Inore important than price differentiation as an explanation
of export competitiveness. Thus, the additional competitive-
ness afforded by tariff preferences would fail to have the
expected positive impact .

A further explanation for the lack of AAS response

187

can be derived from the Linder similarity of preferences
theory. This theory implies that tariff preferences should
have a limited impact on the expansion of exports between
countries with highly dissimilar structures. In order to

test the validity of this prOposition, a trade model based

on the Linder theory was developed and applied to AAS ex-
ports to 140 countries. This analysis showed that the geo—
graphical intensity of AAS exports - as expressed by average
propensity to import (API) - was negatively correlated with
the absolute difference of per capita income ( IAPCII) and

the distance (in miles) between the exporting and importing
countries. It was also shown that API was positively corre-
lated with the independent variable "language" (L). As im-
plied by Linder, the model applied better to consumer goods
than to semi—manufactures. Furthermore, as already demon-
strated in the evaluation of the AAS response, the independent
variable "tariff preferences" (P) was shown to be non-signifi-
cant. Thus, findings from the cross—sectional analysis
supported those of the longitudinal analyses.

On the basis of the above findings, two contrasting
trade strategies for the AAS were examined: a trade strategy
based on an expansion of manufactured exports to industria-
lized countries, and a strategy based on an expansion of
intra-regional trade. It was argued that, given the
characteristics of the AAS, the latter strategy would have

more chance to succeed than would the former strategy.

188

2. Potential impact of current General System of Preferences
(GSP) schemes on manufactured exports from developing
countries

It is difficult to draw general conclusions on the
basis of the findings of this study with respect to the
potential impact of current General System of Preferences
(GSP) schemes on the exports of manufactures from developing
countries.

Conditions characterizing the AAS-EEC trade arrange-
ment differ from those prevailing under the GSP schemes in
two contrasting ways. The majority of the countries included
in the GSP schemes are at a much higher level of development
than the AAS countries (most of which are included in the
U.N. list of least deve10ped countries). More developed
LDCs, in contrast to the AAS, may possess a technological
level and managerial skills which will enable them to expand
their manufactured exports so as to take advantage of the GSP
schemes. On the other hand, two aspects of these schemes
make them less favorable than the EEC-AAS trade arrangement.
First, tariff preferences granted by the EEC to the AAS are
less restrictive than the GSP schemes with respect to commo-
dity coverage. Second, AAS exports to the EEC are not sub-
ject to various quotas as are beneficiary LDCs'exports
under the GSP schemes. Furthermore, during the 1962—1969
period, the AAS was the only group of developing countries
to enjoy tariff preferences from the EEC. Thus, these pre-

ferences should have given rise to greater trade diversion

189

from other develOping countries,in favor of the AAS, than
would be the case under current GSP schemes which apply to
nearly all developing countries.

No general prediction can be made regarding the
impact of the GSP schemes on the basis of the findings of
this study. It may only be suggested that these schemes
are not likely to benefit countries at the same general
level of development as the AAS (e.g., Latin American
countries such as Guatemala, or Asian countries such as
Bangla Desh). It also seems probable that the approximately
20 developing countries which are major exporters of manu-
factures to industrialized countries (see chapter V) will
be the main beneficiaries of the GSP schemes.

It should be noted, however, that even these
countries should not be overly dependent on the GSP schemes
as a basis for expanding their industrial sector. It is
possible that these schemes will be maintained in their pre-
sent form only as long as the economic growth of the pre—
ference-granting countries is not on the decline. Otherwise,
protectionist forces may lead governments to introduce more
restrictive schemes than is currently the case. These
forces are already having an effect in a number of industria-
lized countries. France, for example, has decided to set
curbs on imports of textile products in order to protect its
textile industry. On June 20, 1977, the French Foreign

Trade Minister invoked article 19 of the General Agreement

190

on Tariffs and Trade (GATT) in order to limit certain im-
ports, and thereby maintain the current employment level

in the textile industry. Import limitations would apply
mainly to products such as men's shirts, women's blouses,
T-shirts, and cotton thread which are imported largely

from developing countries. It was stated that France would
have recourse to special provisions in the preferential
agreements granted to countries associated with the EEC in
order to limit imports from these countries.1 If similar
measures were to be applied by other industrialized countries,
the potential effectiveness of the existing GSP schemes
could be greatly reduced. It therefore seems important for
developing countries to follow a cautious trade strategy
which gives as much importance to regional trade as to trade

with industrialized countries.

 

1The above measures were reported in the June 21, 1977
issue of the International Herald Tribune.

APPENDIX

191

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No mosaosom N No.soo

No No No 1.3.so sosooo musoao Hoosoo a Ho.seo

No No No No so No No No No No macaw noose N Ho.ooo
No No No No No so No No No No No mono moss“: N ON.Neo
No No so No mooooopmsso oossoo N No.Nmo

No No No No so No No No No No swam oomsoo N Ho.NNo

No No No so No No No No No nsoooonmsso ooaomosd sausao N om.amo

No No No No No No No No No No No No No swam oosuo .oooaoo N 0N.Hmo
No No No No No No No No No No No No ammo sonosm N oa.amo
No No No No No No No ooooso N oo.oNo

No No No No No No No No No No sounsm N oo.mNo
No No No smoso .eooso sass oso mass N om.NNo

No No No so No No No No No mass monouosoosoo N 0H.NNo

No No so No so No No No No msoaooomaoum some sosuo N oo.mao
No No No No moonaom poo: N oo.mao

No houses home no uoosuxm N om.mao

No No No No No No No No No so .m.o.s .uoos moans sosoo N os.NHo
No No No No No No No No so ouoa ooonn .oouaom N oa.Nao
so No No No No food son» sofio .floooo masses N 848

No No w some omsos sonoso N om.HHo

No No No No No No ssuasoa sonoso N oo.aao

mm N0 N0 mumom + mmwwsm m0 uwwE CQNOHM N ONTHHO

No No No No No No No No No No No some osa>om N oa.aao
«a ma NH as as a o s o m o m N H oooo
GONumsuomoo Ammmmau Emu

mm H 2900
An A u mdd

 

 

 

 

oouumum manomxo gosaz um umow

moss . NoaN .ssossoo was sooo sh mousodxo mofiuaooesoo oossuooossoz n-.H-¢ mamas

 

192

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No No so so so No massam .muomsm oaooeosa N Ne.Noo
No No so No so No No so No No ooooo noosoas ..uooosoo soooo N Ho.Noo
No No so No No masssm .msoosm sosuo N oa.Noo
so No ooNHHoHoooc co>o .mmmmmaoz N om.amo
No No No No No No menu +.mooon mo .muomsn sosoo N ON.Noo
mm mm mm memo ocm muoon mo .ummmm 3mm N oa.aoo
No No so No No No No No No No No .sa> ooosoaz .aosa moanouomo> N Nm.mmo
No No No so No soaoss> sons .aosa noasouooo> N No.omo
No No so moosaoe N oo.mmo
No No No No No No No No ooNsoe oo osNHoEom.sson N oo.omo
No . No museum mo moods N No.moo
No No moanmoomo> souosm N No.eoo
No cosmos mo Hoods N oo.ooo
No No No No No No No No No moanouomo> moans N ON.ooo
No No No No emenuaoso oo>somosd + oosmdosa N oa.moo
No so smash suNs museum sososm. N No.Noo
No No No so No No No nooass oases one .ao> N om.mmo
so semen nsosuas .moooaoeuoz N Nm.mmo
No No No No No noose spas .moooaoenoz N Hm.mmo
No No iooauoo nanosom N No.Nmo
No No No mousse .mososos moans N Ho.Noo
No No No No Macao No .mooa summons N No.ooo
No No No No No No No suummm N No.ooo
No so No No No moons .nooom ssoxom N ao.ooo
No No so No so No No No Houosooam .moaoooz N om.ooo
«N NN NH AN as a o s o o o m N H o
coaumanommo mmmau woo
no sou
AnmeHuCSOU mg
Aconcaucoov un.Hum mqmme

 

193

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No No No No No No No so No No No No eone onH>om H 0H.HHN
No No No No so No No oooesou oenouoeosnee nonuo N om.NNH
No No No so No No No No No No eouuenemHo N 0N.NNH
so so so enemHo N 0H.NNH
No No No No No No No No eoHanm N oo.NHH
No No No No so No No No No No noon N om.NHH
so No nasceno> N mH.NHH
No No No No No No No No oan N NH.NHH
No No so No so No No No No eanno noon .eeoenoneH N No.HHH
No No No so No No No No Hope; HenonHz N Ho.HHH
No No No so No so .e.e.n .menm snenHHso N ao.mao
No No No nemonH> N so.amo
so No No No No so .ueees HeHoHoHune one Hennuez N oo.aao
No so so xooueuueos .edsom N mo.amo
No No No so No No manoeeoe .eoooem N oo.aao
No No .e.e.n .eHeane son oooo .mena H ma.Hoo
so No No noosod neHo one nee: H oo.Hoo
No No No No No No No No No No No eoooHeen neHHsHe one exeouHHo H om.Hoo
No No so No No nonoHeon seHHeHe nonoo + nenm H ON.Hoo
No No No so No No .aena eoooo one eeeHooono N oo.mso
No No No new .HHo .nopusn eoooo H Nm.Nso
No No oueed eoooo H Hm.Nso
mm mm Hmmsm uoonus3 Hoo3om mooou H om.mso
No No No enoenoxo oeoooo N om.Hso
No No No No No No No No No .uense oooooo + eoooooo sense N aH.Hso
eH MH NH HH 0H m o s o o o m N H eooo
coHumHuomon mmmau
He smo
AQmeHuGSOU mg
Hooanunoov uu.Hsm mHmms

 

194

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No No No No HHo ooee nooooo H om.HNo

No HHo neonuesom H ON.HNo

No No HHo one see neHo H OH.HHo

No oeee3.neHn H oa.HaN

No No No No No ee>oon .enoHone .eonnom H NH.HaN

No No No No eonoonnon one eenom H HH.HaN

No No oHee noenoo N om.osN

No No so No No He>eno N oo.NsN

No No nouuoo oeneoo H oo.NoN

No No No No No enoonHH noouoo H 0N.moN

No No No No No No No No No No No No nouuoo oeoneo ooz H OH.NoN
No so No oeneezuoooeHo .Hoos e.aeenm H OH.NoN

No No No No No No oonneHm .oooz esonooHnoo-noz H NN.NoN

No No No No No so No No No No nzee .ooos esonooHnoonnoz H HN.NoN

No No No No enomooHeusesHHen H OH.NoN

No No No No No so No .neooe.n .ooos moonooHnoounoz H NN.NoN

No so No No Heooneno N ON.HoN

so senses ooueneneoon H om.HmN

No No No No No No No uHsno.meoHo so eoeee no nooHn H oa.HNN

No No No No No No No No No No eoeoe nooooo H oo.HNN
No No No No No No No No No No No No ooHHone .eeononnono H NH.HNN
No No No eHeeHne noneo no eone sen H oa.HHN

No No No No No No No No No No Hoos usonuHs eone noose H os.HHN
No No No No so No No No No eone oeoo H oo.HHN
No No No eone oHeo sen H 0N.HHN

«H AH NH HH oH a o s o m o. N N H eooo
coHumHHomoa AmmmmHU emu

An

moHHumsoo mmd

 

 

 

 

AQmSCHuGOUv II.HI< mqm<9

 

195

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

so No No so No No eHHo HeHoneeem H OH.Hmm
so No so .mena HeoHeneernend neneo N ma.Hoo
No No No No memeones .eeoeo .onHooez N Ha.Hoo
No No No No No No eHeeHne + nee non eenHoeoez N os.Hoo
No No No No eennee HeeHne .enHooe> N mo.Heo
No No so No No No No eenHed .eeneHnne> N Nm.mmm
mm mmmam .oHEmuoo Mow emucofimsm N Hm.mmm
No No eHeHneuee oanoHoo sense N 0H.mmo
No nHe oHooHH N Ha.oHo
No eeeeoHHHm N mm.on
No . eooe oHueoeo N No.mHm
No No No eHnoese N Ho.NHo
No oHoe oHnnoHnm N mm.mHm
No so No No neosxo N HH.mHo
No aneosHo H oN.NHo
No so No No No xes eem N No.Hmo
No HHo HeHneeoonH N HN.Hmo
No eHHo eHeeHne no .me> oeHoHooz N OH.Hmo
No No No No No No No .e.e.n .eHHo eHneeeme> oean H oa.NNo
No No No No HHo oeoeeHea H oo.NNo
No No No No HHo unnoooo H om.NNo
No No No No No No No HHo eHea H 0N.NNe
No No No HHo oeeenHH H OH.NNo
No HHo e>HHo H om.HNo
No No No No No No so No No HHo unseen H oo.HNo
«H NH NH HH OH a o s o m o m N H eooo
coHumHHomoQ mmmHu
He soc
Aflmwfiuucfloo m:
Hoeanunooo un.H:< mHmms

 

196

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No No oeneonenunon .eueene season H oo.HNo
so No eeaene senses sense H No.HNo
No No oeenneoHn> non .eeeene nennnn H Ho.HNo
No so No No so eeHoHune nenueeH sense N oa.NHo
No so so so No No so eeone No memes N 0N.NHo
so moHOHuHm mssmmocumm N o~.NHm
No No No No No No No No eHeeHne.nuo «0 eone oenedena H NN.HHo
No No No No one eeom oeneaena H NN.HHo
No No No No No one meene oeneaena H HN.HHo
so No No No No No No eenHeon nenoo oo nenueeH H oo.HHo
No No No nenueeHnoHeo H ON.HHo
mm muonmssmssuxo musm How .monm m ms.mmm
No No No No No No enHo N NN.NNN
No No No noneem N HN.NNN
No No No No No eeneeoeonHeHo .eeoHoHeoeenH N 0N.NNN
No .e.e.n .eneNHHHenen H oa.Hoo
so eneNHHHeneo oHeendeond nenuo H NN.Hoo
No .meHe oHeem H HN.Hom
so No eneNHHHuneo enonemoneHz H 0H.Hoo
No No mwO£m MOM mEmeU + mwnmfldom N om.vmm
No No No No eunemneueo N ON.eNm
No No so No so No No No No No deco N OH.ooo
so No No No No No No No No No eoHueEeoo one eeenonea N oo.NNo
No noses oeHHHoeHo oHeeEonn N oN.HNN
No No No so so eennust unenmenn H NN.HoN

oH NH NH HH oH N o s o m o N N H e
ooo
coaumsuomoo Ammmmao Emu

Anmwduucfioo mg
Hoeanunooo u-.Hun mHmns

 

197

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No No oocHH.omHmH.oumOQHommd + momma N om.Hvo
so No .mmm moons ommoum no ucosnoumm N Hm.Hvo
No No oumonuodmm + momma oomfiuocmm N os.Hvo
No No oneon onHoHHnn enan N oo.Hoo
No oumonuommm ocm momma ummux N om.Hvo
No momma mcHuaHum oomEIoanomz N HN.Hvo
No No No No No so No No No No No No .m.m.c.o003 mo mmHoHuHm Honuo N mm.Nmo
No No No No No No so No No No No No @003 .ousuchmo mo moHOHuum N ms.Nmo
so No eeneno enoeoHa neoooz N Hs.NNo
No No No so No so so No No No No snnHos one snuneaneo .eneoHHnm N oo.NNo
No No No ooo3 no..eeHoHune e.neaooo N ON.NNo
No No No so No No so No No No moumuo .moxon cooooz N 0H.Nmo
No Hson @003 ocm H003 oooz H om.Hmo
No monouma How .monm @003 H mo.Hmo
No moHocmc Hoou How meHum nooooz H oo.Hmo
No No mumom oxmum .mmom .oH0dumom H mm.Hmo
No No ommuomooo no“ @003 H NN.HNo
No No No @003 HmHOHoHuum H Nv.Hmo
No so No eHennea oooz H NN.HNo
No No No No No ooossHa H HN.HNo
No No No No No mmoH Ho EEm .Hooco> H 0H.Hmo
No .oumn..0Hm> ..uum Hooosu.nuo N mm.mNo
No No uHon cmm N oo.mNo
No so No No ens» nennH .ensu oHoenoena N NH.NNo
No Hobnsu omnooumm H oo.HNo
«H mH NH HH 0H m m s o m o m Ni H oooo
:oHumHuomon HmmmmHU emu
AnmoHHusHHoo mod
HoeonHunooV nu.Hu< mHmme

 

198

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No No enenHo.une.nHunoo no eoHnnen H Ho.NNo
so No No .o:.oeHo oHoenunse «0 eoHnnen N NN.Noo
No so No No .nHo.enoo oHeenense no eoHnnen N HN.Noo
No No No No ee>He>.ne nenuo .eoHnneH eons N oo.NNo
No eoHnseo NHen enHo no Hooz N HN.Noo
No No No No No No No No No No No No wenxoenoeeHnnn .eoHnneN nouooo N NN.Noo
so No so No No .seo eHHHneno one eHHa nouooo N NN.NNo
No No No oenoeeHn .eenem nouuoo N HN.NNo
No No No No No so so No No No .noeeHnnn.eoHnneo noouoo nenuo N NH.Noo
No No enenHo eHneueme> senoo oo nee» N NN.HNo
so No so .nHo.nHonoo oHuenense Ho nnes H No.Hoo
No HHMUOH HON sauna mac—MM HO xMHh H Nm..nmo
No .umH.m HOG spunk mHEwH HO xMHm H aminmo
No No No No so No so No No HHeeen non .nnes nouooo N No.Hoo
No No so so No so so No No HHeuen now non .nnes nouuoo N H¢.Hoo
No No No No HHeoen non NHen .H003 no see» N NN.HNo
No .uwH MOM .HOO3 Qmmvhmo MC Show N HN.Hmo
No No HHenen noo .nnes xHHm H oH.Hoo
No No No semen no eeHoHene sense N NN.Noo
No No No No ommmo.am How .mnommm poo Honuo N mm.Noo
No No No muommm mcHNmoo Honuo + conumo N Nm.Noo
No No No No exooHn.exoon euon.eneueHmem N ON.Neo
No No No No eeHoHene snenoHeeuo N 0N.Noo
No No No No No .nneaea + semen No .eoen.eexom N HH.Noo
No No ooumoo .ooumcmonEH .Hommm N mm.Hvo
«H NH NH HH oH N o s o N o N N H eooo
soHumHHomoo HmmmmHU Emu
AQmOHHuCSOO mg
HoennHunooV nu.Hn< mHmme

 

199

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No No ommn Momma .mmcHHo>oo Hoon N Hv.smo

No No so No No No No No .e.e.n ..nne .uxeu doueoez N Na.omo

No No No No No so No No No nenHH eHneu .oem N HN.ooo

No No so No No No No enexneHn noeeoo N No.ooo

No No No so No No No openneHn neHoos N Ho.oNo

No No No No No No No No .uxeo oo mane» .emanze .eHHem N 0N.ooo

so No so No No No so No No No No No No No omen one exoee .uxes N OH.oNo

No No sneanoen non eoHnneo.nxes N No.omo

No No .oee eHHexee ne>o3 oo eonz N No.Noo

No so mnHooes no eeHoHenn N Ho.ooo

No No No No No No eane no eoen euez N No.ooo

No No No No No No No No meson .eoeonoo .eans N Ho.ooo

so No so emananu + eoHnneH oHueeHm N oo.mmo

so No so. .e.e.n .eoHnseo .nmnH.oeoeoo N oo.ooo

so No No No No .neH..oenanH .eoHnneo eHHexes N No.ooo

No so eHeo Ho eeHoHonn N OH.oNo

so No so eoeHm en» nH .sneoHonnem N oo.omo

so so so No so eoHnneo pen nenuo one eHHns N mo.omo

so Heuneeenno .nnes eHHHneno N No.eoo

No so eemoes.eHeneH ne>oz N No.eoo

No No iosoHooo eoHnneo ne>o3 sonnez N Ho.ooo

so No enenHo.me>.neo none eoHnnen N eN.Noo

No NHen eenon no .eoHnnen N NN.NNo

No so No oHeeeHe eon..nnee oeoeHnn N os.Noo

so so No enenHo.ene.oeHo no eoHnnen N No.NNo

oH NH NH HH 0H s o s o o o N N H eooo

coHumHuomoo AmmmmHU emu

HomoHHucsoo wmd

Hoeanunooo un.Hum mHmns

 

ZOO

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No Hommoo o0 .mNoHHm Hoummz H NH.Nmo

No No nemaoo oenHoen H NH.NNo

No so No so No Heeue .noNH «0 emnHoeeo N No.Nso

. so exoeno sesHHen no muses N HN.oso

No No so eneene Heeen one nosH N Na.oso

ho mCOHuowm Hmwum HO GOHH N Nm.m5o

so ones Heeue one nonH N Ho.Nso

so eneHm no .enes .eeHeuom N HH.Noo

No eoHe no eeHoHuno N No.Noo

No No eoenonoo .oneeeo no eeHoHenn N No.Noo

No No eenoeeoanm.eenooeHHHz N HH.Noo

No No so No No No No No euoee oHneneo nenuo N No.Noo

No so so No No No No No No euoee oHeeneo oeNeHmne N oo.Noo

No No onHaHm oHneneo N No.Noo

so No No eonnn mnHoHHnm N Hs.Noo

No No No enoHnn snouoenoem N NN.Noo

No No No oneeeoueoeeeoee no eeHoHenn N NN.Hoo

No No so enooe Heenennnoe + mnHoHHsm N NN.Hoo

so eeoee mnH>ea one oeom N HN.Hoo

so No No No No oenoHoo ne>e .oneeeo oneHenon N NN.Hoo

No No No esHH N OH.Hoo

No No No No No No No eHeHneuee mnHuHeHm N oo.smo

so No No so No snueeaee eoee-onen N os.smo

No No No No No .uesnnHHex .eoeaueo sense N oo.soo

No No eeegneo oeouonx .neHooz N Hm.sNo

oH NH NH HH OH a N s o N o N N H eooo

coHumHuomoQ AmmmmHo emu

AQmOHHHSDOU mg

HooocHucouv 1|.Hum mqm<e

 

201

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No No No No so No No No Noraneeoe\nonH oo .one oHoneenon N HN.sNo
No COHH NO mw>0um..uuow.nw GOZ N HH.Nmo
so No No No eHoou eHneeNnenoneunH N oN.NNo
so No No No eHoou onen sense N NN.NNo
No No No meccmmm Cam wmfiocmhso N NN.mmm
No No sneeenooenme non .eHoon ones N NH.NNo
No so nemooo.eeHaeue.exoeu.eHHez N NH.NNo
No No No No Heeue.nonH.eeHaeue.exoee.eHHez N HH.eNo
No .ee.noNH oo .NnHueen eoeHd eno N Hs.NNo
No ean Heeue no noNH oennem N NN.NNo
No No No No No No No Nee .nanoo non enenHeenoo N HN.NNo
No No No No Heeee.nonH Ho .NnHoneo.NnHueez N HN.NNo
No No noHannHe no eneo .eennn N NN.NNo
No No No No No No No No No Heeue .nosH oo .eEsnn N HN.NNo
No No HzxsananHenHe so .enenHeunoo N NH.NNo
No No No HHooNo neaaoo no .enenHeenoo N NH.NNo
No No No No so No No No No Heeoe no noNH oo .enenHeonou N HH.NNo
No No No No anannHe so eenooonnuo N NN.HNo
No No No No No so No No No No No Heeee one noNH No menoeosneo N NH.HNo
No HanoH. us eoeene eanHenH< H NN.Noo
No HEENH. +o eeeene eoHnHean H NN.Noo
No unmnonsnn nanHenH< H NH.NNo
No Hoodoo mo .monHm .monoe H mN.Nwo
No EEmH. uo>o .muoosm Hoodoo H NN.Noo
No No nomdoo mo .mcoHuoom .mHmm H HN.Nmo
NH NH NH HH NH N o s o o o N N H eooo
. ooHumHuomoo mmmHo
He soc
AQmOHMUCSOO mg
Hoeanunoov u-.H-< mHmns

 

202

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

so .oee .enHeen nee: .eanN N NN.NHN
No No so eennuoonue mnHueon N NN.NNs
No No No No noHueNH>en oneHnH non eueom N NN.NNs
No No eneHHene neneo N NN.NNs
No No no so No No No eeHoNo No eases N NH.NNs
No so No No oeNHnoeon eon .eeHoso N HH.NNs
No so No No No eeHosonouoz N HN.NNs
oo No No No No eeHoHne> noeon no names N NN.NNs
No No . eeHoon eHoHne> N HN.NNs
No eeHoHne> nouoe non eHeeeno N es.NNs
No no eeHoHne> HeHoeam N NN.NNs
No No No anodenenu now eeHoHne> nouoz N NN.NNs
No No eeHoHne> neeneeeea oHHnoa N NN.NNs
oo No No No No eneo ene>Hna N NH.NNs
No No .e.e.n.ee>HuosoooH no names N os.HNs
No No No No No No No No enenHeenoo N No.HNs
No mcommz .msm> mooom mmzHHmm N No.Hms
No No eeHoHne> sesHHen eoH>nem N Ho.HNs
No No No No No No eoHoem N NN.st
No .e.e.n .oeeH oo eeHoHene N oN.oNo
No No .e.e.n .nanHnoHe no eeHoHenn N NN.NNo
No No .e.e.n .neamoo oo eeHoHnun N NN.NNo
No No exooHoea one exooH N HH.NNo
No No so No No No nanHsoHe .eeHoHene oHoneenom N NN.sNo
No so neamoo No .eeHoHene oHoneeoon N NN.sNo
NH NH NH HH oH N N s o N o N N H . e
ooo
coHumHHomoa AmmmmHU Emu
Anmmflhucsco mg
HoennHenooc a-.H-n enema

 

203

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No HeHnenen oeonnnn .exoom N HH.NNN_

No No so .35 nenno no eeHoe nnHz eeonm N oo.HNN

No No No No No No No No No No flxoonotzononnmn :qu moonm N No.Hmm
No so No No so No No No No No eeone nennnm N HN.HNo

No No No ee>oHN nennnn N No.Hoo

so No No No No No No neeNoeen nenno N NN.HNo

No so No No No No No neemoeen nenno one enen N NN.Hoo

No No No HeHnenen oenHeHd no open N NN.Hoo

No No No No so so No No No No .e.n.oenanx .ennenneo nenno N NN.Hoo

so No No No No No No No No .e.n.oenanx .ennennem.neonn N NN.Hoo
so No No No oHneeHe non .eNonoonm N NN.Hoo

No No No No mannoHo nenneeH N NN.Hoo

so No No .m.o.anonmmmm now mmHHOmmoood N NN.Hoo

so No exooe .eNonoone .ee>oHo N oN.Hoo

No so so eeneHeeenn .eneenoo N NN.Hoo

No No No No so No No so eHHe> .ee>neoe .eHseno N NN.Hoo

so No No No No e.neeo3 .ennenneN neon: N «H.Hoo

so No No No No No so No No No No No No .eson + e.nee .ennennem neon: N NH.HoN
No so No so No No No No No No mflnHm + e.nenos.ennennem nenno N NH.Hoo

No No No No No No No No No No No No No , .eson + e.nen.ennennee nenno N HH.Heo
No No No No No No No so No No so No moooo He>ene N NN.HNo
No No No No No No No No No No No No No .e.e.n .eennoHnnnn nenno N No.HNN
No No No No No oooz .NnHooen no eeHoHnnn N No.HNN

No No No ennoHnnnn HeoHoez N No.HNo

No No No so No No No No No eneee nenno one enHeno N HN.HNo

NH NH NH HH NH N o s o N o N N H e on

coHumHnomoo AmmmmHo Mmu

Aflmmflhfiflfioo mg
HoenannooV -n.Hum mHmns

 

204

.moHuHoOEEoo ooachHm on mnooon N mmmHU
oHHn3 moHuHoossoo coschHMIHsom on mnomon H mmmHU

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hm
No No mcoonom cam: 0cm mcmh N om.mmo
so eeoenenne one eeHHennns N Hs.NNo
No No No No No No menonez N NN.NNN
No No eeHoneo N HN.NNN
No No eeneHe onen N sN.NNo
No No No No eenennn one enoonn nenno N NN.NNN
No No No so eoHenm .eoHeHN N HN.NNN
No No No No No sno>H no eeHoHnnn N NH.NNN
No . No No No No sneHHezeN noHneanH N NN.sNo
No No so sneHHezes ne>HHe no oHoo N NH.sNN
No No .neEoea iosneHHezes no 3332 N HH . sNo
No No No No No No No No eennnanoe .snenoeno N No.oNN
No No No No Aeoes oneno eNannHem N Ho.oNo
No eoneon one eeoeHm N NN.NNN
No eHHon N NN.NNN
No No so .No No No No No menHoHneeHn no eoen eeHoHnnn N No.NNN
No mocHNmmmE ocm mnommmm3oz N 0N.Nmm
oH NH NH HH NH N o s o N o N N H
cOHumHnomoo mmmHU ooou
. ie one

AamoHnusmoo odd

 

 

 

 

HommcHuoouv I|.H|« mHmma

 

205

.HHH ennHo> one H eenHo>o
.osNH .NoNH I soNH noonneesleeneHooeenleoens nNHenon
ocm .HmoHnucooo odd vH may no sumo now oEdHo> ocov
momH .moHuHcoEEou :momonsm on» no oOHmwO HmowumHumum "mHommsnm
.oomH I mmmH xoonnmowlmoumHoommdloomne cmHonom
.moHuHsafisoo cmomonsm on» no oOHmmo HmoHumHumum “monsom

nomHz u «H UHHQomom cmOHnom Hmnusou n s

muHo> nouns u mH :oomo u o
HHez u NH omos u N
chmuonsmz u HH nmommmmomz u v
omcou H OH coonosmu u m

omsoa u m Hmmosom u N
soEonmo u m ammoo Nno>H u H
":oHumsmHmoo snuasoo now Nox

an

Afimncwucouv II.HI< mHm<B

206

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NN NoH NN NN NN Ho NH H . o

sNo.oH NNs.NH sso.o NHs.N NoN.oH NoH.o NeN.N Hoo.N N

oHo.HN ooH.NN NNo.oo NNo.Ho HHH.NN ooo.oN NNN.NN NsN.oN o

oNN.N ooN.N NsH.N ooN.H Nso HNe oNo oHo.H N

oeN.os NNN.No NNN.No NNo.NN ssN.sN sso.oN oNN.NN ssN.sN N

NsH.soH NNo.NNH ooo.ooH NNN.NN NNo.ss oso.Ns ooH.NN NoN.oo H AeN eeeHo
ooo.N NoN.o NNo.H NoN NoN ooN.o NsH Nso.o o

NNN.oN oNN.HN NHN.Ho st.os oNe.oo NNN.sN NNo.oN NNN.oN N

NNo.ooN soH.ssN ooH.NNN seo.sNN HNHNNNN NNN.HHN NNN.ooN HNo.oNH e

oNo.N NsN.N NNN.H HNN.N sNN.N NHo.N «NN.H see N

ooo.oN eHN.eN NHN.oH NHs.oN ooo.sH oNH.NH NNo.oH NNN.N N

NNH.osN NNN.NNN Noo.oNN NNN.sNN ooN.NsN NNN.NNN NNN.oNN soo.oNN H AeH eeeHo
NNo.N NNN.o NNo.H NsN NNN Hoo.o oNH Nso.o o

soo.oo NNo.oo sNN.No NNo.oN NoN.NN NNo.NN NNN.NN oHo.NN N

HNo.NNN NNN.NNN Nso.NsN ooo.ooN NoN.NoN NNN.soN NNN.sNN «NN.NNN o

NNo.o NNN.o Hoo.N sNo.N NoN.N Noo.N oNs.H NNN.H N

oos.sN NNN.NN NNN.No moo.eN ooH.NN Noo.Ho HoN.NN NHN.oN N

NNN.sHN NHN.oNo NNH.sNN NNo.NHo HNN.HNN ooN.NNN NNH.HNN NNo.oNN H Hones

soHumc moono

NoNH ooNH soNH ooNH NoNH ooNH NoNH NoNH imHeeen soHoonsoo
Hoeo.H No

NoNH I NomH .moHnunsoo mo masons HmsoH>HocH
on mmHuHoossoo mo museum HmsoH>HocH mo unomxo odd mo osHm> II.mNI< MHm<B

 

207

 

 

 

 

 

moo mom vsH.H mmo Nmo HmN.o HN som.m o

HNm.vN HHs.HN HansH sms.oN oso.HN oHv.NH mmN.s NNNno m

HmH.NOH mHH.m0H oom.mm mms.OHH oHN.mm mam.Hm omm.Hm smv.mo o

mNN.H Hov.H on sHv Ham mmH.N moH me m

Hos.v mdm.o vmm.o ONH.m Hmo.s Nms.m mHm.v omN.m N

ovv.mmH mom.mmH mam.0NH OHmnovH mNm.oHH Hoo.oHH on.moH vmo.NOH H N moono
o o o o_ o o o o o

H o o N H o H o m

omv HNN own mom on sHm osm mNm v

Ho mo NN o o oN ON 0 m

mmmnm 0Hm.m moo.H mmo.N mmm.H ooo.H va.H ooH.H N

Hsv.v Noo.m oHo.N ooo.N moo.H smm.H Nms.H moo.H H. H moonw

mNo.m smm.m wow so ONH oHH mHH msH o

smm.mN osm.oN voNsNN oHo.mN mmm.HN omo.mH Hss.HH Noo.sH m

vsm.moH va.NNH woo.NOH NmN.ms Hanmo Hmo.Ns moN.ms mNo.oo v

omv.H mHo.H Hvo.H HHo.N ooH.H NHm ooN.H Hmm.H m

sNH.mm mos.om mov.oN mmmnHN mom.HN HoH.0N Nooan msm.0N N

www.mNN Hms.mmH ovmnmmH ooH.NNH ONo.mHH HNm.moH omm.o0H msH.ooH H o moonu

H coHumc moono

momH momH soNH oomH momH vomH momH NoNH nIHnmoo suHoossou

 

 

 

 

 

 

 

 

 

 

Hoenannoo. II.eNI4 mHmns

 

208

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

omH ON mN «H H m ow mm o

Nos.m HmN.o mmv.m oNo.m ooo.0H vsm.m mms.m mom.v m

mmm.sm «No.Hm smo.mH oHo.vN Nmo.NN mmm.oN NNm.mN Nso.oN v

va.H mom.H mHm msH.H mav.H moo smN HsH m

mmH.oN Nmo.0N HNo.oH on.HH omm.m mHm.s mmN.o svm.m N

osN.Hs svm.mm mom.Hm NHv.ov HmH.mv mmo.oo mmH.mm ooo.mm H o dsonu
o o o o o o o o o

mom.N ONH.N moo.H NHN.H NHH.H moN.H mmo.H mmN.H m

Hmonm omH.m mmN.N NNN.H ems.H NNNNN Hoo.H omo.N o

Hm mm no sm mm mm HN HN m

Nom.m omo.m ooH.N omH.N mmm.H com oom HsN N

msm.HH st.o omv.m moo.m 0Hm.v oov.o mom.m moo.m H m moono
o o o as o o o o o

Nom.N smN.m Nmm.N mmo.mN mHo moN mm HNN m

vHo.mo Hom.mm oNonmo mHo.sm NOHnHo NoN.Nm smm.N¢ osm.mo v

NH m NH om ms NH vN sN m

Hmo.N NNN.m oom.N mHm.m oovnN oNo.H Nom NHN.m N

moo.mv ooo.mo Nam.so omo.vm moN.oo omHnmm Nom.mo mHH.mv H o moonw

:oHums moonu
NoNH ooNH soNH ooNH momH vomH NoNH NoNH IHumoa NuHoossoo
HoomnHunoov II.mNI¢ mqmma

 

209

.HmoEdHo> oumnmmom 039v

 

momH I somH I moumHUOmmm I moHumHumum oomna cmHonom Noam «Hanucsoo odd nomo
non eno .eennHo> oHc NNNH one soNH .omm “eHeeennm .ooNH I NNNH I eeneHooeen
I moHumHumum momma cmHonom .moHuHmsEEou smomonsm on» no oOHmmo HmoHumHumum «monsom

 

.moHnusmoo umHHmHoom o
.omm u o .geoaH nennoe u N .eoHnnn u N .oHnoz u H
”mcHsoHHOM on» mH moHumcHummo mo moHnusmoo now oooo

one .=eom=o nenuo: u N
no

.mooom nmsdmcoo on mnomon N mmmHU
onm .monmnommmcmEIHEom onm mooom oumHooEnoucH on mnomon H mmmHu

He

 

 

 

 

o o o o o o o o o

om sm mm mH m mH NN m m

ONo woN HmN HmN st sHN NoH moH o

om om m m HH NN o mm m

Hos.m mNs.s Nom.N mom.m msmnN mmo.N mHonH How N

Hmm.oH moo.o soonN mmo.m mootN soN.N mos.H moo.H H m moono
s H m o N Hm m o o

svH smo No oNs NN Ho ov o m

mHm.H NHN.H omH.H OHs mNH.H oss HmN.H oos v

o mH o s mm mo ov N m

mmo.m NNN.m momnH Nom.H omo.H mmN.H mNm.H Hms N

mss.m ssm.v mmo.m mos.N mos.N NNH.N NON.m movsH H s moono

coHum: moonw

NoNH momH somH oomH momH vomH momH NoNH HoIHumma suHoosaou

 

 

 

 

 

 

 

 

 

 

Hoenannoov II.eNIm mHmne

 

210

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ms.m mo.N Nm. mo. NH. 0H. HH. sH. o

ss.NH HN.vH mm.oH mo.0N Ho.NH NN.oH HN.HH wm.oH m

mo.so Nm.oo oo.oo NN.NN om.oo oo.oo ms.mo Nm.oo o

mo. mo. mo.H mo.H Ho.H mo. oH.H NN.H m

Hs.mH NN.NH om.mH NN.sH NN.NH Hm.mH ss.sH om.0N N o moonw
No. mH. no. mo. No. mo. mo. oo. o

oH.HH mm.m vo.m NN.HH mH.oH oo.HH mm.m No.NH m

«moon oo.mm mo.oo NN.sv HN.No No.No NN.om so.mo o

mo.H so.H oH.N sm.H mH.H mm. oo. NH.N m

NN.Nm mm.om mo.mo NN.NM Ho.mm mo.om oo.oo HH.Ho N HmN mmmHU
mo.N NN.H om. oH. mm. os.H so. ss.H o

sm.mH oo.VH NN.NH ov.mN oN.oH mm.0H mo.o Nm.m m

Hm.ss Ho.ms om.ms NN.No wo.os NN.Hm mo.om NN.mm o

mm. oo. oo. ms. mm. oH.H om. oH. m

vo.m No.m sH.o Nm.o mo.o oo.m NN.v Ho.v N HmH mmmHo
ms.H Nm. mo. oH. oN. NN.H so. NN.H o

No.NH vo.mH Hs.NH mm.ON ms.mH mo.0H mo.m Hm.m m

Ho.mo mo.so sN.os Nm.vo NN.No mo.os ms.ms om.ms o

om. Nm. mm. mm. mm. vo.H wm., Ho. m

om.mH mo.oH Ho.oH mH.mH Ns.mH om.NH om.HH om.NH N Hosea

H :OHumc moono

NoNH NoNH soNH oomH moaH oomH momH NoNH nIHnmmo NuHoossou
Hnneo nenv

momH I NomH .monmsm umxnmfi I moHnussoo mo masonm
HmooH>HocH on moHUHoOEEoo mo masonm HmooH>HocH mo unomxo mod II.nNI< mHm<B

211

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

oo. co. co. oo. oo. oo. oo. oo. o
NN.0N Ns.NN No.NH oN.NN NH.NN NN.oN oN.HN oN.NN N
sN.HN oN.NN oN.HN NN.oN NN.sN sN.oN os.NN NN.oN e
sN. oo._ os. oo. oH.H Ns. oo. oN. N
oo.oo Ns.Ne No.NN os.NN No.NN NN.oN NN.oH oe.s N N moons
oo. oo. oo. NH. oo. oo. oo. oo. o
HN.N ss.s HN.N NN.sN oN. No. NH. oo. N
No.No NN.so os.NN NN.so sH.NN NN.oN NN.sN NN.NN o
No. He. No. oH. NH. No. No. No. N
sH.N ss.s ss.N NN.o Ns.N NN.N NN.N NN.o N o anono
oN. He. sN. NN. Ns. oo.N No. NN.N o
so.NH os.NH sH.oH NN.NH NN.NH oo.oH No.s sH.o N
NN.os sH.Ns NN.ss NN.Ns NN.os NN.os NN.oN Ns.oo o
NN. Ho.H sN. NN. NN. oN.H oH. NH. N
NN.N Hs.o sN.s NN.N so.o oN.o oo.o NN.N N N moons
oo. oo. oo. oo. oo. oo. oo. oo. o
No. oo. so. No. oo. oo. so. oe. N
oN.NH sN.s NH.NH oN.NH NN.NH NN.NH oN.HN oN.HN o
oN.H NN.H NN.H oo. oo. NN.H NH.H co. N
NN.NN Hs.NN os.Ns sN.so oN.NN NH.NN NN.ss NN.oo N H anono
soHumc moonw
NoNH ooNH soNH ooNH NoNH ooNH NoNH NoNH inIHneen anoonnoo
Hoenannooo II.nNI< mHmme

 

212

moHnucoOU umHHmHoom u

.mNIm oHomB :H oooH>onm mumo Eonm oochuno moumEHumm “monsom

W ”cm stmomzo HQSUO: "

Numm u e

Nsmgfl Hwfipnv: N m
«OCH3OHHOM m3“ mfi COfiuMGHUmwfl MO m@HH#GSOO MON @000

.eoHnnn I N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hn
.mooom nmsmmcoo on mnemon N mmmHo
ocm .monouommonmEIHEom onm mooom oumHoosnounH on mnemon H mmmHU Hm
oo. mo. oo. oo. oo. oo. oo. oo. o
om. oo. so.N No. mm. Hm. oN.H so. m
oo.o NN.N sm.m Hm.o om.m om.s oo.s No.NH o
no. No. NH. mo. No. ms. oo. so.o m
HH.om ms.mN NN.so NN.NN HN.NN oN.HN oo.om sm.oo N o moonw a
NH. mo. oH. oo. so. oo.H mN. oo. o
oo.N NH.N Ns.N No.oN oo.H NN.H sm.H Ho. m
HN.Nm NN.oN oo.sm HN.NN No.oo NN.oN oN.ov os.No o
oH. om. oN. NN. NN.N NH.N NN.H «H. N
NN.No oH.oo NH.NN NN.oo om.vm oN.Nm No.om oo.om N s moono
NH. mo. so. mo. Ho. Ho. HH. oH. o
oo.N NN.oH oH.sH ms.oN NH.NN No.oH oo.NH No.NH m
HH.Nm oN.NN HN.Nv ms.Hm oo.Hm HN.No Hs.so os.os o
oo.N HN.N sm.N om.N so.m oo.H ms. me. n
os.om ms.mm sN.HN sm.vN NN.NN No.oH NN.oH om.o N o moono
A coHumm moonw
NoNH ooNH soNH ooNH mooH ooNH moNH NoNH oIHumoo NuHoossou
HooscHunooo II.oNI< mamma

2113

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

OOs.Nm «oN.HOH NoN.Ns Oso.Os oHO.No NNN.No soO.Ns NNN.sN NHN.Oo OHo.HN mNm.oN NoN.sN HNN.No HoH.NN NNv.Nv NNNNNH eonn.nuo
sH Noe H moH s NNo N Noo o Nov ON osN N sNN sH NNN men HHo
HHH.vN NoO.s Noo.NN NNN.N mos.NN ONN.s oNN.HN ooo.o NON.NN NON.v «oo.NN oNo.o Noo.NN NOs.s sOs.ON sNO.o eonn.nno
Os ooN ssH ooN oN NON H NNH Ho non HON ooN eN oNN oN . NON one NNN
Nso.oN HNN.sH NoO.HN «oN.NH sss.oN oso.NH oHN.NN NNO.NH Nos.NN NH0.0N NHO.NN NNH.oH NON.NN Hoo.sH NNN.NN oov.sH eonH.noo
NNN.N NON.N HHN.O oHN.H mom.N ooo.H NOH.N NNN.H oOO.N sNN.H NNO.N oNO.H NNO.N oNO.H ooN.N ooN on: HNN
NOO.oo vNN.oN ONN.No Hoo.oN NNo.No oNN.NN NoN.NN HNs.oH NoH.NN Noo.NH NNo.NN ONO.oH NNN.NN NsN.sH NOO.oH oNo.NH monH.nno
N NNH so oN so HOH oH mm s ON sH HN HH oN O oN men NHN
Hss.NH HOO.N HoN.NH HON.o oHN.NH NNN.N NON.NH HoN.N NoN.NH soN.v NNn.NH NoN.N oNo.NH oNO.o HNN.NH NsN.v eoaH.nno
sH ooN H NNo so NON o NOO ON NNN HH ooN NN sNN NN NNN men HNo
oNN.NoH NNO.NHH Hso.HNH oom.oo NNN.NNH osN.Os oNO.HNH HNN.Ns NNO.sNH ssN.Os sHN.oNH oNH.Hs HNo.OHH oNO.No oNN.NOH NsN.NN eooH.nno
OsN.HH sso.HH oHo.N «NH.NH sNN.N oNo.OH HOs.N oNs.o Ooo.N sNO.o NNN.N HNO.s OoN.N ooO.o NNs.H NNH.o men NoN
NHN.ON oHs.NN oOs.ON NNO.oH soo.oN NNN.oH NoH.NN Noo.N NoH.oN NoH.N oNo.mN Noo.o vNN.NN NNN.N NNN.NN NNN.N eunH.nno
ooo.N NoO.HN NsH.o oHo.NN HNN.e Nss.NH oNN.N NON.s ONN.N NNN.o NoN onN.o NoN.H oNN.v ONo OoN.N men N.NsO
sOo.oN ONN.eo NNN.NN NHH.eo mNo.Ne NsN.No oON.oN HoN.sN oNN.NN HNo.He NoN.sN NHO.NN oNN.NN moO.No sON.HN oos.sN eonH.nno
sNH HoN.H oH OoN.H NH NoN.H NN oNO.H VH NNH.H HN NNN.H oN NHo.H NH HsN.H men NNO
ooN.oNH NNN.sN NNo.ONH ooN.oo oHN.NHH NoH.oo NHN.HNH ooN.so NNN.NOH HON.oe oHs.sN soH.Ho Nos.Ns eOs.eN eNo.No NNN.eN eooH.nno
NoH sNH.OH NoN NNN.N sON NNH.OH OHH. oHO.o NH sNN.o mN HHN.o O NNH.N O eNe.N one NNO
=eomzn one eeomzn one =eomzn can, geomzn can =eomzn one eeumzn one =eomsn one =eumzn one ”new“
nenno. nenno= nenno= nennoe nenno= nennoe nenno: nenno: eon eooo
lxm
use oeHm
NoNH ooNH soNH ooNH NoNH ooNH NoNH NoNH TnnomaH
HOOO.H No

HmomH I NomHo

:mooq nocuos ocm m<< ecu sonw emumzo nocuo: ocm 0mm man so monmuomoscmEIHEom ocm monouomoscms mo unomeH II.mI< mHmda

 

214

.mumo o: .munqucHnm nouamsou

.QflBUZD Hm>wcw0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.monsuomwssmEIHEom cum I u uo manomEH .Q<BUZD noonnom
Oms.sm mso.mm st.os mNm.VN voo.mH Hmm.sH NHv.s mHs.HH NHo.v soo.H Nov.m mNH.m NoN.N «HH.N vsv.N sHs mUQH.nuo
mmH Hmo.mN omH moonsH mmN moN.mN som.s omo.mH omo.m Nom.mH Nmm.H mmN.mH Nso.H sHm.0N mm NNN.ON mad vmo
omN.vs vmo.NoH oss.mo oms.vo was.vo omm.os omv.mo moo.ms mss.mo Nvm.mo mmonoo Nsm.vm smv.oo ooo.vv ommnNm mmm.Hv mooa.nuo
mo va Hm oov vv osm om vow mm va mo smv mm smv sm mHv m<< smo
moo.mm omm.m Nms.0m Nos.¢ svs.HN oHo.v omm.mH on.v mos.mH mssnv omH.HH moo.v oNH.NH vms.m me.HH msm.N muoq.£uo
vH HNm NH NMH sN Hm sN mo m mm s me v NNH v mN m<< Nmo
som.st osm.mH NHmnmmH HNo.m msmnsmH smm.m mom.oNH ONm.m mms.moH vmm.v vov.ooH oms.m mms.Hs omm.m mom.oo oos.N mUQH.:uo
Nsv.v Hmm.NH mON.m moonm smo.v omm.s owo.m moN.m VHN.m mom.v mmH.m moo.m Nwo.m mom.m smH.m NNO.N m<< Hmo
"BORN
:momzo 0mm :momza 0mm =mumzo 0mm :momzo 0mm :mumzo 0mm =mumzo 0mm :mumzo 0mm :mumzo 0mm munom
necuO: nonuo: nonuO: nonu0: nmzuo: nonuon nonuo= nonu0= xm ooou
" so I oN.HN
momH momH soNH ooNH momH vomH momH NomH munOQBH
HooscHucouv II.mI¢ mqm<9

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NN.oo Nov.No NN.No sNO.No ss.No oHo.NN HH.so OON.NN NN.Os HNN.so NN.oo NON.sN sN.Oo HNH.NN sN.Ho HNs.Nv o
NO.N sNo.o ss.N NNO.N eo.o Hes.N HN.N oON.v ON.O HNO.O oo.N NsO.N NN.N HNN.N NN.N NON.N N
NN.NH ONs.VH No.oH HNO.oH NN.sH NON.OH Oo.ON NNN.NH Ns.NH ONN.NH oo.NN ooN.NH NN.oN sHN.NN OH.sN NNN.NN N
NN.N oHH.N NN.N sHo.o NN.s «NN.o HO.s NNN.o No.N NNO.N Ns.o nos.N NN.N sHo.s os.o soN.N H HO.NNO
oo.NN soO.NN ON.HN oOO.oN NN.oN NON.sN oN.NN NNN.VN No.oN OOo.NN NN.oN NoO.ON No.NN soN.NH sO.Ne NHo.OH e
vo.sN HoN.sH sN.NN Nvo.NH NO.NN «NN.sH Hs.ON NOO.NH NN.oN HNN.NH Ns.ON NNN.HH OO.ON mos.o NN.NN ovs.N N
oo.HH NHN.s oN.NH ONN.s NN.N OON.N NH.N NOO.O sN.HH Hoo.e ON.NH Nos.e NN.oH st.o oN.NH sNO.v N
oo.N HoO.N NN.o NNN.N ON.N NNN.N NN.o NNN.H NN.N sNo.H NO.N st ON.H ONN NO.H NNN H oN.HNO
oo.No ONN.oN Hs.so oNo.NN NN.so NHO.NN NN.os NNN.NN sN.Os ONO.oN NN.Hs NoO.NN NO.Ns sNO.NN No.oo NNO.NN e
NH.NN OON.eH NN.ON oOo.NH eo.ON ssO.oH oN.NN NNN.NH ss.sN NNO.NH OH.sN HNH.NH sN.oN oNN.HH No.oN ONN.NH N
so.H NoN oN.H NNo oH.H woo so. one oH.H NNN oN.H NNN oo.H oNs NN.N NHN.H N
NO. On sO. sN oO. NN NO. NN NO. on NN. oNH OO. O OO. O H ON.HNO
NN.oN NNH.oNH NN.oo sON.oNH ss.No Noo.NHH NN.No NNO.NHH ON.oo HsN.eHH NN.No NNN.oN No.oN oON.NN NN.oo NsN.ss v
NO.sN OHN.NN NN.NN NNN.Os Ns.HN ONN.Oo oo.ON NsH.NN NN.sN NNs.Ne No.oN oNo.oN oN.NN «Ns.NN NN.oN NsN.ON N
No.N NHs.NH sN. NoN.HH so.v NsN.o NH.N oNO.s sN.N ONo.N NO.N NON.e ss.N Nso.o OH.e Noo.v N
No. sHv.H Oo. HNN.H so. NoN.H oo. sNH.H HO.H ONo.H NH.H NNN.H NN.H ONN.N vs. Hem H OH.HNO
no
my NH.NN HNH.NN oo.NN NsN.HN NN.NN NNo.NN NO.Nv ONN.NN NH.oO OHN.sN NO.ov HNN.HN NH.HN oNo.VH NN.oo NON.OH O
NN.NN NNo.Ho sN.NN Noo.sO NN.os NNN.NN Ho.oN NON.sN NN.oN sHN.NN ON.Oe NNN.NH oN.HO HsN.oH oN.HN ONN.HH N
HN.N NNH.OH NN.s seO.s NN.s NNN.N OO.OH oNN.s os.HH NOH.s oN.N NOO.O ss.NH st.e NN.NH Noe.N N.
HN.N NNN.N NN.N «NN.N oo.N esN.H oN.N NNO.N oo.N HNO.N NO.N NsN.H HH.O Nov.H NN.o sNN.H H ON.NHO
NN.oN oON.NoH NN.ON ovo.ONH HO.se NNN.NNH NN.sN oNO.NNH HN.Oe NON.eoH NN.oN one.NOH sN.Oo NON.NOH Os.NN NON.NN v
oN.NN NNN.NON oN.NN HON.osN No.ON NNN.NNH NO.NN NNH.HOH Ns.ON eNs.OeH HN.HN NON.NNH NN.oN oNs.NN NN.NO NNN.No N
NN.NH NNO.NOH NN.NH NNs.No sN.HN NON.oOH HN.NN ONN.NHH oN.NN NNN.ONH ss.NN oOo.NOH NN.NN NOO.NN NN.NN NHN.ee N
NN. NoN.H NN. oNO.H NN. HoN.H NN. . ONo.H Oo. sos.N on. oNO.N on. NsN.H so. NNo H OH.HHO
W: > w: > MS > m: > W2 > W2 > m: > m: >
enHNHno eooo
NoNH ooNH soNH ooNH NoNH ooNH NoNH NoNH sou
fineu new :5 enmnm uexne: u m:
HOOO.H N nHo enHe> n >

HmomH I Nomi wnmnm umxnme Him msHm> I moHuucsou mo museum msoHnm> Bonn mmHuHUOEEOU nouns mN mo munONHEH 9mm II.mVI< HANSH.

 

216

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NO.H svH NN. OHH Ns.« «oH No.o ON ««.H ON HN.N sH ON.O oN NN.N NN v
No.HH osN.H HN.N soO.H NN.NN NNH.H No.oN NNN Hs.O« NoN Ns.Ho NNN N«.NN OHN «o.sH NNH N
NN.N H«N NN. oOH HN.H N« o«.H NN NN.N NN so.VH Hs NN.HH Hs sH.«o Hs« N
No.«o NH«.HH NN.oN NNN.OH Ho.Oo NOH.N H«.«N HNN HN.NN Nos «O.ON sN NN.HN ooH ««.«H oOH H HN.NsO
HN.sN oNN.N oN.NN NoN.o NN.sN NNN.o ss.NN NNN.o HN.NN NNN.N Ho.«N HNN.s NN.oN NNo.s HN.«N NoN.N «
OO.NH osN.N No.s NNo.H HN.NH «NN.« o«.NN sHN.N Os.oH HsN.N os.sH «Os.N NN.oH oNs.« NH.NH «sN.N N
NN.ON ooN.N Ns.NN NNN.OH Os.o« oHN.HH NN.N« sN«.N ON.NN ONN.s «H.o« NNo.N «N.NN oNN.oH «N.NN oON.s N
sH.N Noe NN.N oNH.H OH.N NsH.H NN.H NN« N«.N «NN N«.H OHN «o. oeN sN. os H ON.HoO
ON.H NoH.H NH.N ooH.N oN.N N«N.N oo.N oHN.N sH.o NNs.s HN.N NoN.NH NN.N «N«.N oo.N HHH.N «
«N.N oON.N NO.HH NON.s NH.N N«N.H Nn.N sNN.H ss.N ooN.N NN.N NNN.« NN.s sON.o ON.H NsN N
HN.so oos.«N No.Ho «Ns.NN oo.oN NN«.Ho NN.«N Nos.ss so.«o NHN.Oo ss.No o«N.sNH No.os NON.No «O.No NoN.oo N
No.o H«H.« HN.N Noo.N oo.N NHO.« No. ooN NH.N HNO.N O«.« NOs.o Hs.N «s«.o oo.N sNH.N H OH.HoO
NN.H NN NN.H HN oN.« ON NN.N No oo.H «N OO. O NH. N OO. O «
NO.NN NH« sN.NH N«N «N.NH NHN OH.s NN NN.N ss NN.« «o NN.N oo NH.N o« N
Ns.N NNH N0.0H NsH os.«H N«N NN.NH NNH NH.HH ooH «O.NH NsH No.NH HHN «s.HH NsH N
Ho.oo OHN.H No.oo NoH.H NH.so NOH.H «o.Ns «oN os.Ho OsH.H ON.NN NNN.H NN.NN HNN.H NN.NN NNN.H H N«.NNO
HN.NN NNH.NN N«.NN Nos.NN H«.NN N««.HN s«.NN «oN.oH N«.NN «N«.«H NN.NN «Ns.NH NN.oN oNO.NH NH.No NNN.oH «
HN.NN OHo.HN Hs.N« Nos.oN Ns.NN Noo.NN NN.oN soH.ON sN.NN oNN.«H «N.ON moo.NH oN.NN ssN.HH NH.HN O«N.OH N
oN.NN «HN.NN No.NN HNN.NH NN.NN N«H.sH No.ON NNN.oH No.ON Hoo.NH NN.NN Ho«.NH OH.sN «N«.OH oH.sH ONO.o N
NH.N soo.H oH.N O«s.H NN.N NON.H NN.N HNo.H NN.« NNN.H N«.« Nsm.H «N.« NNs.H NN.N N«N.H H oN.NNO
NN.HN sNO.s NN.oN «HN.N NN.«N OO«.N oN.NN «NN.N ss.NH oHN.N NN.NN NN«.o NN.HN sNO.N oN. moN «
sN.o« NNN.NH NN.N« N«N.NH NN.oN NNo.s H«.oH NoO.« No.ON NNN.o No.N ONN.N N«.N NoH.N NN.s OON.N N
HN.ON ooN.o N«.sH ooH.o «o.«N NOs.o N«.s« N«s.HH NN.N« HsN.N NN.ON N«N.NH sN.N« ooH.HH NN.Ns OHN.NN N
NN.N ooo.N ON.OH ONo.N «N.HH NNN.N NN.NH NoN.N NN.NH N«O.N sN.sH N«N.« Hs.NH NNN.« sm.oH sNN.N H oN.N«O
ml > m: > m: > m: > m: > m: > m: > w: > 0000
nHNHno
He emu
NoNH NoNH soNH ooNH NoNH «oNH NoNH NoNH
Hoenannouo II.e«I< unmnn

217

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

so. vHH om. oMH oo.H osH mo. os mm. mm oN.m mom mo.o omv mm.N moH v
vv.mv vom.m No.oN owm.m oo.NH oNO.H Os.NN MNm.N ss.HN mmm.N MN.vM msm.N oH.vm Omv.N om.oN mms.H m
oH.mm smo.m om.mm sss.s No.oo oHv.m No.oN Nmo.v oo.sv Omo.m No.mm oom.m sH.sm HOH.v Ns.wo omv.v N
Ns.sH smo.m NN.oN mom.N mm.mN mNm.N Ns.om sms.m ov.Om ooN.m om.v mmm oo.H NvH oo.N sMH H oo.NNo
OH. em 00. mH HN. omH NH. Nm ms. Gem oo. mmN Ho.mH mNH.HH Hm.HN HNm.OH v
mm.o mHm.m ON.m omm.N mm.v Hmm.N oN.v mmm.N sv.m oNv.N Om.m moN.N oo.v ovo.N om.m vvm.N m
mN.sm soo.ov NN.sm mmo.sv mm.sm omm.vm oo.om omo.mm so.vm ooH.Nm mv.mw Hom.oo sv.so me.mm vN.¢o OHo.Nm N
so.o stsm mv.s mmonv vo.s oms.v No.0 Nov.o NH.HH mss.s mm.s mHv.m oN.m Nmo.v mv.m smN.v H oN.NNv
oH.m oom.m mo.m omH.m Nm.v smm.m so.s OMH.o Ho.m vsm.m oN.NH mmm.m om.MH omsnm oo.N vmm v
mm.VH moo.m No.vH NHH.® om.o «No.m oH.v mom.m vo.v vam.m mo.sH vmo.v sH.s HHo.N Ho.v OON.N m
oo.NN mom.vH mm.oH Nom.m HN.mH Nmm.oH oo.mN oom.mH oo.HN msm.vH Ns.Nm mom.m mm.sv moo.MH so.oH vos.m N
mv.mm va.sm om.mo NQH.Hv sm.oo mmo.vm mv.vo sHm.Hm sv.mo moo.vv mm.sm vmm.oH mm.Hm sms.m Hm.ss Nos.ov H oo.HNv
sm.Nm msv.Ho Nm.vm som.wv mo.om Hmm.Hv NN.Nm mmo.mm No.Hm omm.sm No.Hm omN.0m mv.mv ssN.mN Ho.mv mos.vN v
vm.mm mNN.mm mm.om NOH.sN om.sN Nms.oN mm.mN NHs.0N Nm.om HNm.NN ms.sN Hmo.oH om.sN Nmm.oH om.mN ovm.VH m
sm.MH Hmo.mH oo.mH ooo.MH os.mH oms.HH ON.mH mmN.MH om.sH on.NH mm.oN sHN.NH mH.mN mom.MH oN.HN QNH.HH N
mm. oHo sN. st Ho. Hem mm. omm om. osm No. oom mm. Hmm so. ovm H oo.NHH
om.m Nmm.o mw.mH oom.oH mH.oH HsH.o Hm.MH mmm.o mH.sN MHonHH No.oN NvN.m mm.vm mmo.OH ms.oN Hmo.NH v
vH.mH voo.NH mo.VH wmo.OH mo.mH HmH.m mo.sH moo.m vo.mH moo.m mm.sH moH.s VH.m oom.m mm.HH va.m m
mm.mo sss.mv Hv.oo NHm.mo No.Hs soo.mo mN.so mmo.mm NN.Nm moo.NN No.oo QmN.vN mo.mo mos.sN No.oo oso.mN N
Ho.N wov.H mo.m mHH.N sH.N Hmm.H ms.H mom om. mmm so. ovm mN.H Nmm oo.H NHm H oN.Hmo
NH.N vmmsH oo.N mmv.H oN.s on.m ms.NH omm.m mm.oH mHs.N om.HH omo.N mo.m vmm.H HH.m mmv v
mo.mv omNtov mo.oo omH.mN NN.mm omm.sH mH.mv Hoo.mH oo.mm vHv.mH sm.oo oNN.oH Ho.Hm mmo.oH om.mm ooN.m m
mH.oN Nmm.MN vs.mN mNm.mH 0m.Hm Hom.vH mo.mH oom.m No.NH soH.m Ho.ON omm.v oo.oH NmN.m mm.mH osm.N N
oo.NN vmo.mH HN.Hm sHm.mH oN.mN ooonoH No.oN smm.o MH.mN mms.m oo.NN mom.v oN.NN mom.v oo.mN msm.m H Nm.Nso
m: > m: > m: > m: > m2 > m: > m: > m: >
ll 53 no eooo
Hm Emu
momH momH somH ooNH momH vomH momH NomH
HomscHucouv II.va< mqmdfi

218

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NN.Ne ooN.NN oN.O« ONN.«N NN.O« H«N.sH NH.N« s«N.sH HN.N« NNN.«H HN.N« NNN.NH ss.N« ONH.N N«.N« NNs.s «
NH.o« ONN.NN HN.N« N«N.NN NN.s« NsO.HN sH.«« NoN.sH O«.N« NN«.«H ON.O« NN«.HH N«.O« NNo.N NN.«N oNN.N N
NN.N Oso.N sN.N NN«.H NN.N Nos.H os.« sNN.H N«.N NsN.H NN.N «ON.H HN.N H«H.H NN.N NoN N
NN.s oNo.o No.s NNN.« NN.s sNH.N NN.s NHN.N HN.N NON.N N0.0H s«N.N ON.HH Hs«.N oH.NH NHH.N H HN.HNo
«H.NN NNN.sH sN.NN NNO.NH NH.oN NNN.NH «N.sN NNH.NH NH.«N N«O.NH NN.HN NNN.NH NN.ON HNs.OH O«.NN NoN_HH «
NN.No Nsv.oN NN.NN OHN.«N «N.NN Nso.NN NN.NN ooN.NN oN.NN ONH.HN NN.ON «NN.oN NO.«o sNN.NN NN.Ho HoO.HN N
os.NH NsN.NH Ns.N «OO.o NN.N OHo.N sN.N NNN.N NN.N soN.H Ns.H «Hs HN.N NON NN.N HNN N
NH.N ONs.o Ns.N sN«.N Hs.N NHN.« OH.N ONO.N No.« NNH.N oN.N sNO.H NN.N ONN Ns.N ONN H OH.HNo
sN.OH NNH.N HN.HH «sN.s No.NH NOs.o NN.oH NOH.N oN.sH NsO.s NN.ON ONH.N NO.HN NsN.s HN.NN Nso.s «
NN.so No«.NN oH.«s NOO.«« ON.Hs ««O.NN ss.No HNs.NN Ns.Ns HNs.NN ON.Hs NsN.sN «H.Ns «sN.NN N«.Hs NON.HN N
HN.HN HNo.oH 0s.NH HHN.N sO.«H «NN.o No.NH N«O.s NN.s sNO.N NN.o «N«.N OH.N Hos.H ss.N «NH.H N
s«. NsN «N. N«H os. NsN NH.H sNo HH.H N«« oN. NsN Hs. s«N oN.H ONN H O«.HHo
NN.N «so sN.N ONH.H NN.NH NNN.N NN.NH soN.H NN.NH HN«.H os.NH NsN.H NN.HH «NO.H NH.NH NHN.H «
HN.HN HNo.NH HN.oN oNN.NH sN.os HH«.OH HN.«N NNN.OH oo.ON sNN.N sN.Ns NNo.s HN.Ns «ON.s Ns.«s sNo.N N
sO.N NNN oN.H oNN NN.N NoN NH.N NoN «N. sN HH.« NH« ««.N sNN Os.s HNN N
NH.N NoN No.N «HN oN.H «NH Ns. NN s«.« «o« No.o Noo NN.o NNN O«.« NON H HN.NNN
HN.NN NNs.s s0.0N oos.o Ns.ON NoN.« sN.N NNN.H NN.oH N««.N «s.ON NHN.« oo.NH «N«.N oN.NH sNN.N «
oo.ON «sN.NH NN.NN NNH_NH ON.Ho sNN.NH oN.Ns ONs.NH NN.«s NNH.NH sN.«o NON.«H oN.No NON.NH N«.NN oNH.OH N
ON.NH NNH.N No.s HNs.H HN.NH sNN.N Ns.N oO«.H «N.N oHN.H NN.NH OOO.N sN.NH OHs.N No.NN No«.« N
NN.N NN«.H NN.N HNN NO.« s«N sN.N oNo NN.N «oo oo.N Ns« Hs.H NON NN.N s«« H NN.HoN
«s.NH NNN.N NH.NH NNN.N «N.sN NNN.N NN.oN sNo.N ON.sH sH«.« N«.s NNN.« NN.o «NN.N NN.N N«N «
NN.ON ooH.NH Hs.N« NsN.o «o.N« sON.N NH.H« «NN.N H«.oH ONO.« «s.« HHs.N sO.« NHN.H NN.«H sN«.N N
NN.ON s«N.« NN.NN NNo.N NN.NH soO.N «O.NN oNN.N Ns.NH NsN.N NN.« NNN.N N«.OH NNN.« No.N« NNN.s N
NN.NH ooo.N NH.N «NN NO.N ONN ON.« «sN sN.N« ONN.NH Ns.NN ONN.s« NN.Ns NNN.sN NN.NN oN«.o H ON.NN«
m: > m: > m: > m: > m: > o: > m: > m: >
nHNHno eooo
He soc
NoNH NoNH soNH ooNH NoNH «oNH NoNH NoNH

 

 

 

 

 

 

 

 

 

 

 

 

 

HomscHucouv II.m¢I< NH.NH;

219

.momH Du NomH mumms .mmHuHcsasou smomousm unu mo mOHNNO HmUHumHumum NmHowmsnm

 

.munomEH mmu I moHnmu HmoHuNngc I momnu cmHonom .moHuHasanu :mmmonsm on» No monuo HmuHumHumum "mounom

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.moHnucaou .00m + =mum=n nonuo: n v mam .umm I m .=muoq nonuos u N .mmd n H
"manomaH uo :HmHno ozu now new: mm: 0000 oaHsoHHou use Hm
NN.om smo.va No.om ovo.HmH NH.NN HNv.omH NN.NN amo.mmH om.mm Oss.mHH ov.oo Hss.0NH NN.Nm VNo.vm so.mo vso.om v
oN.NN mom.ovH so.NN ooH.NHH ov.sN smm.ms os.NN omo.Nm No.oN vmv.ms NN.NN omv.mm vo.sN www.mv No.NN Hsv.Hm m
NO.N NNO.Nv mo.m Hmm.sH oo.N mHo.mH oH.v omv~0H NN. mmv ms. som.H sH.H osm.H oN. oom N
cs.v www.mN NN.o som.sH mo.m OHN.NN mN.o mmo.mH Nv.s Nom.mH No.N NNN.mH oN.NH oHo.ON vo.VH mHm.0N H 0H.vmo
om.mm Hos.N Ns.Nm ooo.N sN.No momnN OH.No HoN.N OH.No Nmo.N Ho.mo vmm.N os.No mHH.N No.om Hov.H v
Nv.sH oom No.NH Nmm oN.HH vmv mm.HH vom sN.oH vmo NH.mH mam No.HH son 0N.sH sNo m
N«.NH on on.0N mom NN.NH mvo oo.VH moo v_.s cam ms.o ooN mm.m NsH so.m ovN N
oo.N st oo.N Nov sN.N msm Ns.OH mom No.HH omv oN.NH omv Ns.NH NNo Hs.oH mHv H om.smo
No.NN mms.s so.NN NNan OH.NN Osm.o sm.VN mmm.s oH.VN NNO.N os.NN NNN.m Hm.sH vms.v oo.oH mso.m v
NN.sN Hov.NH mv.mm «oo.NH No.HN omm.m om.om vvm.HH Nv.mm ONH.HH NN.sN ovs.m OH.NN mNm.0H NN.Hv mmH.m m
NH.sn HON.NH NN.sN NNv.NH os.vv HHH.¢H NN.oN moo.HH No.ov OHm.mH sm.om omo.m sH.Hv Hmo.HH Ho.Hv mmHnm N
sm.H mvv Nm.H moo No.H on oo.H oom ms.H vmm oo.N Nms NN.H sHm om. mNH H 0H.omo
oo.HN va.v oN.HN Hmm.m os.NH vms.H oN.NH ooH.N oo.oN HNs.N sH.NN omm.m sN.sH vNN.N oo.NH NNH.H v
vo.ms smm.vH No.os NHo.NH sm.os omm.o NN.ss ooH.m NH.Hs NoHNOH mH.vs mNo.NH No.ss oos.m ov.vm Nom.s m
NN.N Hvo NH.N mom NN.N omN oo.N omm Ns.H ovN NN.N mHv mv.v com NN.H omH N
mo. mH NN. sN oo. om oo.H NNH oo.H moH OH.H NsH mm. NHH ms. mo H Ho.mmo
mm.mm How.o No.oN Nav.m NN.NN va.v oN.NN NNN.v om.ov mv0.m HH.Hv vvv.v NN.ov mom.m om.mv svm.N v
NN.No st.HH vo.oo moo.m NN.No va.s No.oN NmN.s Hs.sm NNN.s NN.oN Hmm.m so.mm Hmv.v No.Nm Nam.m m
om. mm Hm. ms NN.H moH NN.H oVN ms.H sHN NN.N sHN os.o NNN NN.H NNH N
mm. msH om. Nm so. om vv. mm NN. sN sm.H osH NN.H sOH mm. mN H NN.Nmo
m: > m: > m: > m: > m: > m: > m: > m: > woou
HmcHoHnO emu
NoNH mooH soNH ooNH momH ooNH NoNH NoNH
Homchucou. II.va< mqm<9

220

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NNN O«o.os ooN oNN.Hs HNN NNN.Ns Nss NN«.os oss «No.oN Nso NNs.«N ONo NNs.NN NOo N«s.ON «
Oso NNN.NH N«o sO«.NH «NN ONN.N NNN ONN.N oN« NoH.N No« HNo.o NN« ««N.s NH« HHN.o N
N«s NNo.NH st osH.HN N«s NNN.NH st NHN.«N NNs ««s.NN NNo «Hs.sN «so NH«.NN NNo H«N.HN N
OsN NNN.N NNN «NN.N NoN NNN.s N«N NNN.s «NN oHN.o NON N«N.o NNN N«s.s NNN.H oNO.« H HO.NNO
st Ns«.N« NHo NON.o« «NN NNo.s« oNN NoN.H« NoN NHs.NN NNN NNo.sN «N« NsN.HN oH« O«N.NN «
NNN NNs.Ns «ON NNo.os NNH NNN.Ns «sH HsN.os NoH NNo.ss N«H ONH.ON N«H ONO.Ho «NH NNs.Ns N
sNN.H NNs.« ON«.H oNH.N NNN.H sN«.N NHN.H «NO.N sHN.H NNN.N HNO.H N«N.« «oH.H ONN.« oNs NN«.N N
HNN.H H«N.N NNN.H «NO.N NON.H sH«.H sNN.H NNN HNN.H «ON Nso.H HN« oNN.H HNN oo«.H NsH H ON.HNO
«ON ONN.Ns sNN Noo.so NNN NHN.«o «NN «Ns.Ns oNN NNN.so «HN N«N.No Os« NNN.No NN« sNo.No «
No« oHN.NN «O« NsH.NN «NN NNN.O« NsN NNH.N« NNN ms0.0« OON Noo.o« «sN NOH.N« NsN N«H.o« N
sNN «ss.H sN« «O«.H o«N oOH.H ON« NsO.H oNN NON.H OsN «NN.H Oo« N«o.H «NN NsN.« N
«««.« N osH.N sH HsO.N «H HsO.N «H NNN.« N oN« ooN I O I O H ON.HNO
NN« Os«.NsN NNN ooN.NHN oNN ONN.HON OO« Ns0.0NN ssN NNN.NON sNN NON.oNN «NN «sN.NON oNN NNN.NoN «
««N Noo.«NH «HN OsN.oNH NON OsH.ONH NN« NNo.NHH oNN O«N.NHH NNN HoN.NN HNN HON.NN ONN N«N.ss N
NsN sNN.NN s«N Noo.ON Noo NNs.«H NNN NNs.NH NN« NNO.NH oo« ONN.N NN« N«N.HH ON« NON.OH N
HN« sNN.N NON sN«.N OH« NNO.N Ns« «HN.N s«« ONs.N NNN sOO.« NO« N«N.o «N« st.H H OH.HNO
ooN NNN.NN NNN NNN.sN NsN sNN.NN HN sON.NH« NNN OHs.NN «Ns NNN.NN «Ns HON.ON NNs «NN.«H «
NON.H «NN.s« s«N.H «NN.NN oNN.H «sN.NN OoN.H NNN.HN NsH.H NNN.NH NNN.H oN«.NH NNN NNH.oH NNN NNH.NH N
«ON NNN.NH NoN oOH.N «ON N«N.o «oH ONo.«« N«s sNN.N NNo OOO.s HNN oNN.s ««o NNN.N N
N«N ONs.N N«N NN«.N oNN «oo.N oso HNO.N «NN NNN.N ssN NHo.H sHN sNN.H HON NN«.H H ON.NHO
NON Nss.NsH NNN NOO.osH NoN NHO.«NN oNN HH0.0NH HNN ONN.sNH NNN ONo.sNH Nos osN.O«H NNs Noo.s« «
OON.H NNN.NNN NNO.H ONN.HNN oNO.H Nss.NsH sNO.H NNH.oNH NOO.H NNN.NNH NHO.H NON.ONH oNs s«N.HNH NNo ONs.sN N
sHo NNN.NsH sHo NHO.NHH N«o NON.NoH oOs oHH.NoH NNs HHN.NsH NON HNN.oNN NN« NNs.NNH oN« NNN.HOH N
N«N Hs«.H NON «NH.H «NN sNN.H sNO.H Ns«.H NHH.H oN«.N ««N sNH.N NsN s««.H NNN N«N H OH.HHO
n a n o a O a O a O a a a o n o
eooo
chHnO
. He emu
NoNH NoNH soNH ooNH NoNH «oNH NoNH NoNH
Hnonxo nHo eoHnn u n
Hence nHo sonnneno u o
HmomH I NomHv I coHnQ Ucm quochU I moHnuczoO o0 museum msoHnm> EOno moHuHoOEEoo nonE 0N No monOQEH umm II.bVI4 mqm<e

 

221

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

os« NHm «ON mmm OON ONN NNN Nmm Nm NON No HoN NNH NON omm Hm «
Nms ONO.N smv NNN.N Nmm NHN.N NoN «NN.N NHN Nos.H NNN HOO.H «NH NNo.H «NH ONN m
sNH Ons.H NHN sN« moH NNN NoN NN NoH NON sNN NHN No HNs oHH NoO.« N
NNs smm.«H st m«m.sH «NN mom.o NNN N«o.m s«N smo.m HNN N«N N«N oss HNH «mm H HN.NsO
oN smm.mON Nm «NN.ONN mm NNO.NNH mm oHo.OmH NN NNN.N«H N« moH.HsH m« «NN.NNH NN ONN.oHH «
on Nso.Nm sm Ooo.m« m« HOH.NOH H« «Nm.«mH mm «N«.Hs ON mom.«s om «NN.«N om mHN.Hs m
sN oN«.«mm Om NNN.Nom NN ooo.Nmm NN «NH.HNN NN «NH.sHm om NHH.soN N« Omm.st oN msm.NNN N
oN Nmo.mH Hm m«o.om «m NNN.«N sN ooo.mH NN oON.NH mm smN.N mm soN.o mN O«O.m H ON.HoO
Om ms«.mN mm NHo.O« s« osN.NN NN oNH.Nm oN N«s.mN NNH OHm.oo «NH mom.os os NNN.sN «
osH Hmm.NH «s HHN.HOH sN NHN.sH NN mNm.«N NN s««.o« NON st.mN m«H NNH.NN mo Hmm.«H m
msH s«m.on NNH Hs«.Omm oNH ooo.NNm NNH Nos.NOm N«H NHo.smm NNH NNN.Nso NoH H«o.s«m OsH OHs.«mm N
«N sNN.N« NN NNN.HN Nm mso.m« om OO0.0H NoH NNm.sH «NH O««.om HNH sos.mm «sH smo.mN H OH.HoO
NNH ONH NNH NNH oNH «Hm «oH NN« H«H osH I O «NH NH I O «
NNH «HN.N NNH Hms.H NON «sO.H NNN O«N Nmm NHN OH« omH NO« NoH NHN ««H m
sNH N«N.H NNH «NN.H N«H NHo.H O«H NOm.H ONH HNN.H NNH «N«.H NNH msm.H N«H NsH.H N
NON ONN.N «HN omm.m NON NHm.m NNH «mm.« ONH smH.o sNH «oo.o ONH st.s NNH NHm.o H N«.mmo
ooN st.sN msN Nos.oN «NN Hmm.Ns «NN HHs.N« OHm «Hs.o« sOm «Ns.«« H«m smN.«« HNN sNH.Oo «
NNN «NH.«NH NNN NNN.N«H HmN «ON.NOH NNN «oo.sN oNH HON.«s NNH NHN.mo «oH OON.No soH st.mo m
NoN NH«.NN omN oNN.Hs NNN ooo.ms NHN sH0.0N NNN Nss.NN HoN Nmo.HN ONN O«m.m« NNH NHN.NN N
NNH ss«.N mHN NNO.N «NN NNH.N NNN moo.s NNN Nmo.s HsN som.o mHm «HN.N NNN ooo.m H ON.NNO
O«N msN.mN oNH OHm.N« NNH Nos.m« «NH msN.NN moH Oso.sH N«H ONH.N« ONH NHO.mm NNH «NN.H «
N«N NNs.mo oNN mHm.Os NNH sso.mm OsH OsN.mN OsH oNO.H« HNH Nss.NH «NH HNO.«H oNH ONs.«H N
NNH NHm.Nm omH NH«.mm N«H NNm.m« NOH HHN.NHH oN ooH.NHH oHH NON.NHH NHH o«N.Nm oNH HNN.HNH N
NoN mHo.HH NoN ««m.mH NNN Nss.OH NoN NoN.NH HNN NoN.OH NNN HsN.oH NsN ON«.oH NoN omm.NN H ON.N«O
n a n a o a m a n a n a a a n a eooo
ienHNHno see
NoNH NoNH soNH ooNH NoNH «oNH NoNH NoNH
HooDCHucouo II.QVI¢ mHmda

 

222

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

«NN HO« NNN NNN «HN OsN NN« ««H «sN NNH NNN NNN.H NNN «NN.H HsN «No «
mNN NOH.NN mmm mss.OH NNN NNs.m sNN oNO.N HNN NNN.s osN mNo.N ««N «NN.OH NNN mmo.s m
HNN NON.ON smm NOH.NN NmN NNN.NN HNN soo.«H HHN mHN.oH NNN «so.mH N«N msm.oH NNN NNN.NH N
NoN NNN.HH NNN NHs.N NoN NNs.N NNN NNN.NH NNN NH0.0H oNN HNH.H NsN OHN N«N ONN H O«.NN«
NNH NH« om ooH HN o«o.H NNN N«m NNH «oo.N ooN «NN OHN NHm.Nm NHN mNmsm« «
sNN «ON.NH oHN NNH.mH NNN sNH.OH «NN ««o.m sNN sH«.s NsN onnm ««N mHN.oH N«N «sN.HH m
coH nmN.mNN HNH omH.moN oNN HHN.NNN «NN moo.N«N NoN NO«.mNN HNN soo.moN mHN Hmm.NsH mHN ooN.NVH N
OsH ooH.NH «NH oNO.HN NNN oNH.0N N«N ssH.oN osN HNH.NN NNN NNO.NN NNN m««.ON sNN NNO.NH H ON.NNv
NNN oNs.oH H«N NNO.NH moN mso.NH HON «sm.ON Nmm Hos.oH mmN oo«.HH osN smo.«H omN N«m.N «
NNN sNH.mN NNN on.Nm NHm Nmo.oH NNN mON.oH osm omm.N «mm NNH.NH sHN Nmm.o mNm Hms.o m
NHm ommnsv N«N N«m.om soN smm.mm NNN st.mo mam mom.s« sNN HmH.Hm moN «ms.om OsN on.Nm N
NHN NoN.sHH oNN «NN.ooH «NN NNN.HoH N«N «NN.ONH ONN NNN.sHH OH« NNN.NN «O« «ss.HN NNN NNO.NOH H N«.HN«
NNN «N«.NN NNN NNN.HN NNO.H HNN.O« N«O.H NNN.NN NNO.H HNH.NN NNO.H «HN.NN sNO.H NNN.NN s«H.H sNo.HN «
HNN.H NoN.HN osN.H ««N.HN NNN.H sNo.NH NHN.H Hss.NH HON.H NNH.sH NNN.H NNO.NH O«N.H NsN.NH NNN.H soo.HH N
mHm ovmsom mmm mHN.mN mmm NNO.NN ««m m««.«N Nmm MNo.«N NHm moN.mN mNm sso.oN Now ooH.NN N
NN« NNN.H ONN Ns« NN« N«o NN« NNN Ns« NNs NN« NNs oN« NNs ON« NNs H N«.NHH
NN HHN.«NH oN oNs.NNH so «H«.NN No NNN.NN N«H NHN.HN oN NNN.N«H NN NHN.HNH so ONN.NNH «
oo soo.NmH oo NoN.NmH mo «NO.«mH Os sovnvNH so oHH.ONH sm mmo.mNH om mms.mo «o HoN.mN m
NN ooN.oNN sN Ns«.NNs NN NNN.N«s Ho NOs.H«N oo sNN.osN HN HNN.«s« NN sN«.NON «N oNO.HNN N
N« NNN.NN NN NOH.sN oo NNH.NN oo NNH.«H oN NNN.N N« Hoo.s ON O«o.OH NN NON.N H ON.HNO
s«N.H NsN NHN.H «NN «sH.H «NN.N ONN HNN.« oHN NNN.N NNN.H NNN.N NHN.H N«N.H ONO.H oo« «
osN.H «H«.NN NN«.H H«N.oH oNH.H HNs.«H N«O.H NoN.«H NOH.H OsH.NH «NH.H NNN.N HsH.H OsN.N HNH.H OHN.s N
NNN.H HNN.NH NN«.H NNN.OH «NN.H «Ns.HH oHN HNN.o N«N Nso.N ONO.H ONO.« ONH.H NsN.N ONO.H NNN.N N
N«O.N NNN.N NNN.H ooo.NH sNN.H NNN.N oNO.H N«O.o sNN ««N.o OHN.H sOO.« HNN.H Ns«.N NNO.H NsN.N H NN.NsO
n a a a a a n o n o a a n a n a eooo
:HmHHO
He emu
NoNH NoNH soNH ooNH NoNH «oNH NoNH NoNH

 

 

 

 

 

 

 

 

 

 

 

How=CHucouo II.n«I¢ mqm<a

 

223

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

nmN mmN.NVH MMN wmm.m0H NmN ovv~nn mNN mmo.mh NNN mom.mw hHN avo.om ooN ooH.vv MHN cam.wm v
mmm wVH.VHH omm vvm.om NNm Nnm.vw 5mm HNm.mm mvn mvv.Hv Hmn Ham.Nm Nun nnH.wN wNn oVH.hH n
NOn mmm.m va mow.m va vvv.o NmN wNo.o hmH mmv.o th wov.w th me.v ovN Nmm.N N
mam 5mm.mH mmN oov.vH HHm mNN~0H th vvw.m mmm Npo.m mvm NmN.o mNm nnm.h th th.w H HN.Hnw
oHv NoH.Hv ovv wvv.mn MHv mmv.Hn mNm Nmm.mN on mmn.mm mmv oNN.wN mac NNN.HN mom hva.mH v
mvo.H rpm.vm NHH.H Nmm.om me.H NHm.mN HON~H Nmm.wN owH.H Nmm.wN NNo.H www.mN Ono.H Nmm.HN moo.H mNm.oN n
avo.H hmm.MH mHh Nov.m omm «Nv.v mom mNo.n och Nmo.N MNm va.H hwn mmH.N NmN hwo.N N
aNN omo.mN NNN mvv.VN NON www.0N NMN mom.HN wNN mvm.m wHN mmm.¢ omH veh.v NNN mmH.¢ H OH.Hno
omN.N Hmm.m 0mm.H mmw.m wvo.N th.m mmv~N www.m ooH~N HMN.m wHH.N 0mm.n vmw.H va.n me.N on.m v
OHM m NmH.mH Ham.m HhN.MH mam.m HNm.OH mov.m ONH.HH mmm.N mvo.0H mNo.m NON.m vmo.N mHm.m moh.N mmm.h m
hmo.H www.mH mhw NHm.NH wa mOH.OH wmo~H oom.m mmn vMH.v wmw www.N mmH.H va.H mmm.H hmm N
th.H mmN wm¢.H ooH mVM.H mNN me.H vmv Hmo Non mNm Hov HHm HNN me va H ov.HHw
mOH mmH.o va NNN.m VHH mmm.NN VHH omn.mH vOH nom.mH HoH whw.NH ooH th.OH NOH 0mm.VH v
mHH mov.wMH mNH mmm.vm HMH om¢.mh NNH mwm.Nm wNH mov.ww oNH mNo.Hw mNH www.mm mNH v0m.mw m
QMH ovm.N VNH voo.N wHH Nno.m MHH Hmm.N oNH mom om mnm.v mHH mHm.N m5 hmH.NH N
mHH OOH.m NNH hvo.v NNH mv¢.H mHH wmm ONH mom.m mNH wov.m VMH ovv.¢ mHH MHm.v H Hm.mmm
ow hom.th mv Hmm.mMH mv wHo.hm hm nVH.hm hm NHv.ow Hm hvo.mm ov HwN.Ho He th.wm v
«V mvo.HmN mo ooo.me Hv mmo‘va mv mNN.mmN vv 0Hh.nvm mm Non.Hmm mm HNN.mvm vm www.mmN m
mm mmm.wm mm OHm.0m mm Hmo.mm hm CNm.vN vm mem.NN vm wa~mm mv th.wm vm 5mm.Nm N
mN mNo.Nm vN mNo.vm MN wNm~wm mN ovo.mN mN owH.VN vN on.mH NN NNF.MH MN vmv.mH H mN.Hom
mmN Nmo.HH mom mvv.m me va.m Nov Hvo.m new Nmm.m HNm wNm.MH mmm oom.m moN HH@.N v
VHm Nom.mm mmN mmm‘mN mmN moo.mH mmm MNm.oH mHm mon.NH NNN mmh.m com mwm.o omm NMH.N m
mmN NNv.mH mmm Nmm.wH ONN mwo.n mvm th.HH own mNN.0H mmv vmw.m hum va.m mum oom‘NH N
mam oom.HH ohm OHN.H mnm NmH.H hmm now.H wwm mom.mm Nhn Nam.p~H th vmo.ooH Nov moo.wH H om.NNv
m o a a m a A a m o m a a a a a uuoo
3535 Eu
mme momH bme womH mme vomH mon Nme
Hucanuzoov nu.nq-¢ wands

224

.mvu¢ «Hank how mm 035m "wuudom

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

meuucsou .00m + zmuuzn umnuo= u v was .uam u m .=munq nunu0= u N .m¢¢ u H
"manomaa mo :HoHuo man How 00m: mu: 0600 mcHonH0m one An
hnm va.mNm Hom me.Nmm qu mmm.v0m Nom rpm.awN Hom NHa.OMN va www.mvN mhv omo‘mmH omv hHh.mmH v
me omo.HwN mNm nmm.hHN HNm mHm.HvH mom vvo.NNH mHm mom.QVH on vvw.vHH hhv nhH.Hm mhv ch.ww m
vmm mhn.vm NHm HNo.mm mHm HmN.0m NNm NOO.oN mwv hvo.H mmv NHm.m nmv mmn.v mmv mom N
mvm mmH.mv mHm va.vn an th.mv Hhv nmN.mm mow mNH.mm mmv ONN.Nv wmv www.mv mmv ONH.wv H 0H.vmw
va 00H.m mmN NH>.m HmN me.oH mam mum.m omN oHH.m mmN mvo.0H NVN mwm~m NNN me.w v
mNH wmw.v NNH nmv.m NmH mmw.m mMH moo.v mmH mmo.m mmH vmv‘v mmH mmm.N MVH Nmm.N n
on.H hhv mmm.H MNm va.H HNm NmN.H 0mm mmm.H HmN NNo.H mmN Hmm ahH Nmm mNo N
th.N moH mmm.H MON oHo.N mmH eno.N N«N NHH.N NmN moo.N mmN vvo.N owN va.N omH H om.hmw
HVm nmv.MN NoN won.VN th mom.mN mNm www.mN 0Nm mmo.mN mom Now.mH Nom mmm.mH N0n NmH.NH v
va NVN.mN HNv hvm.aN mmm mnm.VN «Hv mHm.NN vmm mmm.mN mmm NHw.NN Nwm NNm.NN 0mm hmo.MN m
0mm mmN.vm NNm Hmm.mm mvm Hom.ov hwm MHm.Hm vmm mov.ov me oow‘Nm mNm an.mm VVm wmw.wN N
VNH NHw.m NMH N>n~m vMH mom‘m NmH mmm.m wHH mmo.m omH OOH.w wVH va.m mNH mmm H 0H.wmw
New NN«.m Now NoH.m mHo 0mm.N mob mo>.N vv> moo.m mww Nem.m mmw mON.m wmh wmm.H v
won hwv.ov vvm Nwm.vm mmm OMN.mH NNm onw.VN mwv No>.HN ovm mNN.NN HHm Hmo.mH MHv mvo.mH m
me oom.N omN MHo.N mmN Noo.H Nmm MHo.H rpm mmw MHv VHo.H Nmm bmv.H onv mwm N
mmN Nm mwN mMH chm mvH mHv va nnw on ohm men wwv owN mmm HmH H Hm.mmw
0mm mmm.wH mam Nmo.mH va HHo.0H mmv mmo~HH mov m¢v.NH 0mm wmo.NH mmm ovm.m mHv Nvo~h v
mmm mom.mN VNv va.0N mHv NmN.NH NNv HmH.NH va HNN.mH oov moo.VH NHw mm>.0H onv mom.n m
mmN mNm mHv oNH hmv mmN NNm an an mmN NHm mmm ‘mom omw 0mm mON N
vm ovm~H ONH mmv mm mmm ooH Hmm mNH mON VNH ohm.H mNH mmm wMH mmH H mm.wa
m o m a m a a o a a m a a a m a 300
AmchHno emu
mme mme Nme ome mme vomH mme Nme
GwDCHucouv I|.nvu¢ mama;

225

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Nv. Om Nv.m va NN.m OVN mv.N Nmm ON. N Ov.H OO OO.N OHH OH.O ONO 30m
OH.NO OO0.0 OH.OO OOH.O Oo.Om OmN.v OH.ON NOO ON.O OON O0.0H NOH.H NN.ON OH«.H OO.vm mmo.m moHuO<
mv.Nm Hmm.m Nv.Om OON.m VO.ov mOO.m Om.NN OMN.m vo.OO NOv.N O¢.ON VOO.v ON.NN «Oo.v mH.Om OOO.m 00m ON.NvO
NO.H HO mv. mm vm. NN HO. O OH. O OH. OH mm. ON OH. O 30m
ON. HN Nv. mm «N. OH vv. ON me. ON mm. Om vv. Om HN. Om moHuwd
O0.00 OON.O OH.OO NON.N Nv.OO Omm.O mm.OO «mO.m Om.mm OHv.m ON.OO ONm.m ON.OO NN0.0 vH.OO NOm.m 0mm HO.NmO
ON.nv Onv.H HO.NN Nvm mN.MN va H0.0 OO OO.N Om ON.v NN Nm.m ON HH.mH OH 30m
O0.0 Om OO.NH NHm O0.0 ONH N0.0H NOH Hm.vH vOH OO.NH mN Nm.OH Om mm.ON ON MUHMOO
ON.Om OOO.H Om.vO HOm.H mm.NO Oom.H NN.NO HOH.H mO.NO OOO.H Om.NO OOv Hm.NN NON Om.OO HON 0mm Om.Hmo
HO. O NN. mm ON.v VNH OH.O Omv mv.H ON Hm. O OH. O OH. N 30m
O0.00 NOO.N O0.00 mNo.v HO.mO VNv.v vv.HO mmm.v NO.NO NHv.m Om.OO NOO.N Nm.OO HON.v vN.OO OHO.v MOHHMO
OH. O NO. m OH. O Ov. mN OO. Om OH.H mm mm. OH OH. O 0mm ON.HmO
ON.NH OOH mN.VN ONv HH.mv OOO HH.mH OvH mO.mH OmH Om.O vv vo.N Nv OO.mH Om 30m
Ov.mO OON ON.mN OOO O0.0H NHN NO.MN OON NO.mH OmH mm.HN OO NN.vm NNm NH.mv HHH OUHHOO
ON.vN OO Ov.Om OOO HO.vv ONO NO.NO NNO O0.00 OOm O0.00 NHm OH.Om ONN No.om OO 0mm OH.HmO
MN.ON HHN ON.OH Ovv Ov.vH HOv NH.NH HOO v0.0N mvm Om.O OOH OH.O NOH VN.N NOH 30m
HN.O OOm OO.HH NHm ON.HH NNm ON.MH Oov N0.0 VOH HO.HH mON mo.OH OOH vO.vH HNN moHumd
OO.HN va.N NO.HN mNO.H NO.mN mOm.N vo.OO Nmo.N VO.MN NHO.N O0.0N ONm.H ON.mO mOm.H NO.NN HOH.H 0mm OO.MHO
VO.H OHH ON. HH OO.H NO NO.N HmH OH.OH HON N0.0H NNm Nm.m OmH Hv.ON HOO.N 30m
NO.mO Hmm.m m0.00 Omm.v Om.ON NNN.N mv.Nm mmO.m ON.Nm HOO.H OO.mm mNH.H N0.00 HON.H Nm.ON HOH.H muHuO<
vN.vH OvO N0.0H ONO.H OO.HN VOO.H mm.vv OOO.N OO.HO OOv.N ON.mm OHO.H HO.mv NOO.H NN.H Om 0mm OH.HHO
m: > OS > m: > m: > m2 m: > m: > m: >
coHumc wwou
IHumwQ Emu
OOOH OOOH NOOH OOOH OOOH VOOH mOOH NOOH
Huch Hmm CHV mumsm umxumz u m:
Hooo.H » cHO msHm> u >
AOOOH I NOOHO oHuoz may mo umm» mnu 0cm

.OUHHOd .Umm mcu Ou mmHuHCOEEOU uoflma ON mo muuomxm Odd mo wumcm umxnme Dam wsHm> In.ovl¢ mqm<fi

226

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NH.H N mm. O OO. O OO.H N nm. N OO. O OO. H OO. O 30m
NN.NN HNH OO.Nv OON OO.NM VOH NH.NN OO OH.vH mm ON.NH mm Ov.v OH OO.m OH MUHMM<
HO.HN Ovv NO.NO NON N0.00 Omm O0.0N HON NN.OO HNm OO.NO NON N0.00 va O0.00 vVN 0mm ov.NHH
mv.Ov mmO mm.mN VNm ON.ON Omm OH.Ov ONO vo.HN OOO.H H0.00 HNO ON.Nv OHm ON.OO NNO sou
OO.N OOH mO.m Ov ON.v ON vo.m mO Om.v mO mN.N vHH Hv.N 0O H0.0 NO 60Huw¢
H0.0v OvO V0.0N OOO OO.mN mvN.H ON.Ov OHO Om.¢N Ovm vN.ON OOO HO.qv VOV OO.mn VOO ufin ON.HOO
O0.0n OOO.HH HO.mH vOm.m ON.NN OOO.v NN.Ov MHv~v mO.mv NOn.v mm.mH NOO O0.0N NOO.H OO.VH VOm 30m
OO. O OO. O OO. o OO. O OO. O OO. O OO. O OO. O MUHuud
NH.mO NNv.ON Om.vO Hmm.OH ON.NN mOv.OH ON.Om OOO.m mm.vm OON.m NO.vO mmO.v OH.HN OON.m mo.OO OON.O Dam NO.NNO
HN.O NON mm.m va Ov.VH OOO N0.0N HNH OO.vv HON v0.00 ONO vv.vv OOH H0.00 «HN 30m
NO. ON OO. Om OO. VH HN. H OO. O OO. O OO. O OO. O MUHuut
NN.OO «HO.HH NN.OO OO0.0 OO.VO HHO.H NO.mN Own HO.mm OOm Ov.OH OO Om.mm OHN OH.HO NO URN Hm.NNO
VO.vN NON Nm.O mv OO. O m0.0H vm Nv.N OH OO. O OO. O OO. O 30m
OO. O OO. O OO. O OO. O ON.O HH OO. O OO. o OO. O moHuwt
Om.mN OO OO.HO HNv O0.00H OOO OH.OO OHN ON.OO HOH O0.00H OOH O0.00H OOH OO. O Una OO.HOO
vO.mN VOO vN.ON OOO.H OO.NN OHO.H ON.MN Omv~N OO.HN OON vv.ON ONm.H N0.0H OOH.H Oo. O 30m
HH.vv mmO.H vv.vm NON.N O0.0H ONH.H OO. O NO.v OOH OH.N NOH NO.N ONH HH.m va OUHHMO
ON.NO Omm.H N0.0H OOO NO.HO va.v NN.ON ONO OO.mN OmO~N Om.NN NNN.O Ov.OO mNm.m O0.00 ONH.v Hum OH.HOO
ON. m ON. O ON. N OO.m Hm ON.H NH Om.H OH OO.N ON Ov.H OH 30m
Om. v Ov. O OH. H ON. N HH. H NN. O Nv. m NH.m cm MUHHMO
mm.OO an.H VN.OO OHO.H ON.OO NHO.H HN.OO ONO H0.00 HNO NO.NO HOO.H ON.NO OOH.H N«.OO HvO.H UGO mv.mmo
OO.m VO oo.m Ov OO. OH HN.H ON OO. O OO.N Hv ov.N Nm Om.v mv 30m
ON.N NNH OO.N ONH OO.N OHH O0.0 OmH Om.O NO mo.m Nv ,mO.m ON NN.N ON MUHHOO
OH.OO HOm.H Hm.OO VNv.H ON.NO Hmv.H NN.OO OOm.H VO.mO ON¢.H Om.vO mOv.H ON.HO ONN.H OO.NO ONO 0mm om.mmO
m: > m: > O: > m: > m: > m: > m: > m: > :oHum: mvou
IH m0
OOOH OOOH NOOH OOOH OOOH vOOH OOOH NOOH .u O emu
Aumquucouv ||.uvI< mqm<a

 

227

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

OH.NO VH0.0 O0.00 ovm.O OH.mO OHN.v OO.mO OHO.m vN.VO ONO‘m HO.mN VOO.N v0.00 NNO.H Nv.NO vvo.H son
on. ON NN. OH ON.N OOH NO.N OOH mN.N ONH OO.H OO OO.H em OO. O mUHNMO
HO.Nm vOO.N Oo.mm OOO~N NO.Vm VOm.N Nv.mm NOO.N No.mm OVO.H OO.VN MOO NO.Nm OOO OO.Nv OON 0am OH.HOO
O0.0N VON O0.00 OHH HN.OH VO O0.0 NO ON.m OH mm. H NO.N v OO.H v 30%
NO.VN NOH O0.0 Om ON.O ON N0.0 OO ov.O ON O0.0 OH OO.N OH N0.0 OH MUHHMO
ON.Ov NOM HO.HO NOH O0.0N Omm Hv.HO OOO O0.00 OOO OO.NO vON ov.OO ONH OO.NO OON 0mm ov.HHO
O0.00 OOH OO.NN OOH OO.NN OO O0.0m NN OH.mH OO ON.OH OO O0.0 HO ON.mH OO 30x
N0.0 ON OO.m OH ON.O NH NN.NN VN NO.N NH ov.H O NO.N OH OO.m OH MUHHOO
NN.OO OOH HO.vN Omm ON.NN mHN OO.Nv NM ON.VO OOm mm.mO OOO OO.NO OOv Om.mO ONv 0mm H0.000
VN. O OO. O HO.N OH Ov.N vH OH.H H OO.N NH V0.0H OO OO.vm OON 30m
O0.0 ONH NO.NH NO O0.0 vv OO. O OO.N N OO.N Nm Ov.m O O0.0 ON MUHHOO
O0.00 OHH.H OO.NO VOV OO.NO Oov O0.00 OOO HN.HO mO O0.00 ONv O0.00 OON O0.00 OOv Dam ON.HOO
OO.vm ONv HO.Hm NNN OO.mH ON O0.0N OHN OO.VN NON OO.N NOH HO.H OH OO.NO OON 30m
NO.v OO Ov.O OO vm.OH OO NH.OH OO NO.NH ONH ON.mH ONH HN.OH H«H H0.0H OHH MUHHMO
O0.00 ONN vN.OO OHv Ov.ON vOm vN.VO Ovm Nv.mO OmO Hm.ON ooo.H Ov.vO OmO HO.Nv OOm 0mm OO.NNV
OO.Hv NOO.N ON.vO ONO.v Nm.mO OHm.N N0.0m OOO mv.MH mOO VO.H O vN.N HH OO. O 30x
OO. mm OO. MO vH. O O0.0 OOH NO.m OOH NN.O Nm O0.0 OH OO.m OH MOHHMO
HO.NO VO0.0 v0.0m OVN.N v0.0v ONO.N NN.OO OVN.H ON.NO HO0.0 O0.00 NNO OH.OO HOH ON.OO Omv 0mm ov.NNv
NO.H Hv OO. O OO. O mm.OH mVN mO.H ON OO. O OO. N NN.H Ov 30m
OO.NH HMO HO.v OvH HN.O HO ON.v NHH Nv.N vom ON.O NHH NO.m NOH M0.0 NOH MUHuu<
OO.HO OOO.N Ov.OO OOo.m ON.VO ONv.H HO.VO OOOOH ON.OO NHN.m O0.00 mvv.m O0.00 OHO.N vN.HO OON.N 0mm ON.NNv
NN. OO Nv. OVN mo. HN OH. OO OO. N Hm. OOH OO. vm Nm.O HNv.N 30m
NN.v OOO.H OO.¢ NHO.N OO.m vov.N O0.0 OOO_O HH.m OOO.H VO.N OON.H NO.H NNO ON.H OOO MUHHOO
O¢.OO OOv.Om NO.vO OOv.NO H0.00 Ovm‘OO N0.00 OHH.MO O0.00 VOO.NO vH.NO HON.Ov Om.OO ON0.0m V0.00 Ovo.Nv 0mm ov.HNv
m: > m: > m: > m: > m: > m: > m: > m: >
:oHum: wnou
lH mm
OOOH OOOH NOOH OOOH OOOH vOOH OOOH NOOH .u O BOO
835208 nu.ovu< manta.

Hut anma now uccu mm wamw "monsom

 

 

 

 

228

 

 

 

 

 

 

 

 

 

 

 

 

 

vo. OOO HO.H ONN NN.v OOH OH.¢N ONO.¢ NO.NN HO0.0 ON.O OON.H Nm.m OOO.H oo. 0 30m
00. 0 OO. O Oo. N om.oH ONNON OO.N OO¢.H ON. NV 00. o No. O MUHHN<
O0.00 Ooo.HN O0.00 OHm.OH ON.OO NO0.0 OO.VO OON.OH OO.VO OOO.NH O0.0m NO0.0H OO.HO ONH.ON O0.00 OOv.HN Uflm OH.vOO
Nv.O ov N0.0 Ov OO.NH OO OO.N Ov O0.0 vv ON.O OO NH.O Ov HN.N NO 30m
OO.NN OVH O0.0 OO vo.HH HO Nv.oH NO Oo.mH Om ON.HH NN ON.OH mm OO.NH Om OUHde
N0.0N Omv ON.HO NOV N0.0N HOO OO.HO OOv NN.ON OHO Nv.mN HOV OO.NN 0mm HN.OO OOO Uflm OO.NOO
OO.NH NOH N0.0 OO V0.0H HO O0.0 OO OO.vH Om O0.0N OOH HO.vH NO ON.O ON 30m
ON.ON NOO OO.HN OON N0.00 OVH OO.VN COO N0.00 own O0.0H VN OH.NH NO OO.N HO mUHHu<
O0.0H ONH ON.HN OON ON.OO ONH O0.0H OOH O0.0N vNH ON.OO ONO HO.NN Hov NN.VO ONN 0mm OH.OOO
OO. H OO.H O Ov.N O OO.N O N0.0 HH OO.H O oo. o NO.N w 30m
OO.HO .OOH ON.OO HOH NH.OO vOH VO.NO ONH Om.om OVH O0.00 OOH ON.OV mm VN.VO ON MUHHM¢
O0.0 OH OH.N v OO.NH Nv HH.Ov OOH O0.0v NOH ON.vq OOH ON.OO ONH vm.Nv mm USN H0.000
VO.NN NN ov.O v O0.0 O O0.0 O V0.00 NH O0.0N OH M0.0 m OO.Nv O 30m
om.ON OH O¢.Om NN H0.00 ON Ov.OO HH H0.0V OH OO.¢v NN O0.00 VH Ov.HN O OUHNM‘
Oo.Ov ON HH.OO mv O0.00 OH vw.vO NH Ov.ON O N0.0N VH ON.NO OH HN.OO m 0mm OO.NOO
OO.NO HO0.0 O0.0v OON.O OH.Ov NOO.N Oo.Ov OOO.N O0.00 OO0.0 V0.00 OO0.0 OH.OO HO0.0 OH.HO OO0.0 30m
NO.NH OVN.H NN.NH OHO~H HO.VH OOO N0.0H NOO O0.0H OOO.H N0.0H NHN.H ,ON.OH OVH.H V0.0H NOO.H MUHHM<
Om.Ov Oom.v OO.Hv va.m oo.Nv OON.N O0.00 OOv.N NH.HO NvH.N V0.0N ONH.N O0.0N NHo.N HO.NN ONv.H Dam HN.HOO
m: > m: > w: > mz > m2 > MS > m: > m: >
:oHum: wuoo
:H mm
OOOH OOOH NOOH OOOH OOOH VOOH OOOH NOOH .u D Emu

 

 

 

 

 

 

 

 

 

 

 

HawscHucouv ||.0¢|¢ manta

229

 

 

 

N Noe ONN~N cocmnmN N no NNN.N MNmoNnum
N NNN.N NNN.N uNm3ox N NNN NNN.m unamm
N NNN NON.N magmx N NNN ONN.N Nocmsom
N NON «em eunuch N NNN «ON.N oNanmmm :moNcNeoo
N NON.N moo.0NN comma N aNm.N NNN.NN xumE:mo
N Nmm NNO.N monamn N mm moN ameonmo
N NNN NNN.N ammoo Nuo>N N omm.N NNN.NN mem>ononomNu
N ONm~N NNO.N NmemN N NNN Nee magnum
N mmo.N Nmo~m ocmNmuN N moN NNN omcou
N com NNN.N cmNN N mNN mN mw:MNmN ouoeoo
N NNN NNN.N meN N NNN mNN.m MNneoNoo
N Na NNN.NN «NocN N ooN coo.0N maNgo
N NNN.N mwm nchmoN N 0N N«N cane
N NoN.N oom~NN manned: N NNN NNN .mmm =m0NNN4 Nmuuamu
N Now NNN.N mcom mac: N mmm.N NNN.Nm mwmcmo
N va Nmm mausucom N NNN NNN cooumemu
N om NNN Nuamm N oNN mNN «NconEmu
N pm NNN mchso N NNN NmN.m «NummNsm
N mum NNN mmstomsu N NNN nom.Nm NNNmNm
N NNN oom.m mommuu N moo.N vmo.oN stmamm
N mmN mmo~N mcmsu N mmv.N NNN.oN mNuumsa
N NNN.N ooo.eNN Nameumu N mmN~N ooN.NN mNNmNNmsN
N NNH Nw MHQEMU N mvm NHo.OH mcaucmmud
N «on NNN . conmw N ONN NNN mNoch
N 0Nm.N NNN mNmmcaNom .Nm N mmN NNN.N «NummNN
N oNN‘N NNN.N ccmNcNm N moo «m mammN-mumN4
ANV A.s NV ANV N.s NV
Na Non mzw auucsoo Am Non mzo Nuucsou

 

 

 

 

 

 

 

 

ANOOHV

m4¢ ms» sNN; chwmnu mmNNucsoo No mmmam:MN new .Nom .mzw -u.m-< mNmaa

 

230

 

 

 

N «N oNN muNo> Noam: N Nqo NNN soNcsmm
N on.N NNN.NoN souchx omuNco N oN hm mchsw .uuom
N mm NNN ‘ macaw: N NN« «Nm.w Nmmsuuom
N Nmm NNm.NN - amxuaa N Nmo.N Nom.¢m oGMNom
N oNN NNo.N mNmNcaa N omN Nam.m summ
N oNN «NN ommnoeunmwNaNua N NNN mnm.mN cmumemm
N NNN NNN omoe N Now Now msmcmm
N mmN «No.m vamNNasa N NmN.N ovm.m Nmauoz
N NN «NN «Namucme N NN NNN.N «NummNz
N NNN Nam.m cmsza N mm omm NmmNz
N NNN NNN.N mNuNm N NNO.N Nmm.m c:MwaN 3mz
N omm.N NN«.mN ncmNumNuN3m N NNN.N mmN mNcoomNmu 3mz
N mmo.m NNN.NN cmcmsm N mom.N mmN.NN ocmNumaumz
N mOH mom.H cocam N mOH HMH.H msOHnEwuoz
N OOH m MUHumd nuuoz amacmmm H omH mmO.N oooouoz
N OON NN mchsu nmNammm N mmm omm.¢N ooNxmz
N NNN ONN.oN :Nmmm N NmN ooN msNuNusmz
N mmN NNN.N emz umN> nusom N NoN NNN mNamumNsmz
N NNN on.mN moNNNN ausom N NNm NNN. mszcNNNmz
N No mmN mNNMEOm N «cm mNN muNmz
N NmN mam mcooq muumNm N mm NN« . NNmz
N NNN NNN Nmmmcmm N NNN NmN.N MNmmmNmz
N va NON.N «Nnmu< Nusmm N no «NN szNmz
N we NmN mucm3m N oNN NNN umommmmomz
N NNN «NN.mN mNcmasm N oNN~N NNN.N «NNNN
N NNN mNo.N mNmowonm N NNN omm «NumnNN
NNV N.s NV . Nay N.a NV
Am Nun mzo auucsou Nu Nun mzu Nuucsou

 

 

 

 

 

 

 

 

NewscNucoov -u.m-¢ mqm¢e

 

231

 

000N .mN NNsn ..0.N.¢ “.o.0 couchnmmg .NNN Nonesz Nouucou manomum
~GOHmN>Ho unommm can moHumHumum .ucmemoH0>mo HmcoHumcumucH How mocmmd NNO

 

HNmH .mcoaumz UmuHCD “xuow 3mz .ONmH I mmmH uwm>usm UHEOCOUO GHNOS
.mNHmwmd HMHoom can UHEOGOON mo unwauummmo .mcoHumz UmuNcD NHV "mousom

.nocmum can» umnuo
mumsOQMH m NCO magnum N umnfisc moon ms» mHN£3 nocmum Now mocmum H NmQEdc «woo was

 

 

 

AM
N 0Nm NNN.N mNnamN N 000 «NN.0 mNmstcm>
N m0 Nmm.N mNNmN N «NN.N 000.0NN .m.m.m.0
N 0N0 0N0.0 mN>mNmomsw N 0m0.0 000.000 .¢.m.o
:0 7e 3 N E 3: 2
Am Num mzu uucsou Am Non mzu Nuucsou

 

 

 

 

 

 

 

 

NewsaNucoov un.mu¢ mNm<a

 

2232

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000.0 “cannon
000.0 .mmm .300
000.0 000.0 000.0 000.0 000.0 000.0 00N.0 000.0 000.0 000.0 000.N Nuoaama
000 00N 000 . 00N 000.N 0N0 000 000.N 0030000
000.0 000.0 000.0 .Nmosoouoi
000.0 000.0 00N.0 mammao
0N0 000.N 000.0 000.0 00N.N 000 000.N oocoo
000 . 000.0 .N ouoaou
000.0 mNnaoNoo
000.0N 000.NN 000.NN 000.0N 000.0 000.0N chsu
000.N 00N 00N.N 000.0 000.N 000.N 000.N 000.N 00:0
00N.N 00N 00N.N 000.0 000.N 00N 000.N .m.0.uuamu
00N.0 000.0 000.0 00N.0. 000.0 000.0 000.0 000.0 000.N 000000
00N 00N 000 000 000 000.N 000.N 000 000.0 000 000.N cooumanoi
00N.0. 00N.0 ”Noonamo
000.0 mNummNsm
000.N NNNmum
000.0 00N.0 00N.0 000.0 00N.0 00N.N 00N.0 000.0 005.0 000.0 000.0 000.0 000.N stmcmm
000.0 000.0 000.0 000.0 mNuumsN
000.0 000.0 000.0 000.0 0NNmuums<
000.0 000.0 0N0.0 chucwmuc
000.N oNN 000 000 000.N 0N0000
0NN.0 000.0 0N0.0 0NN.0 00N.N 000.N 000.0 000 000 000.0 000.0 000.N 000.0 00N.N mNummNN
000.N 000.N 000.0 000.0 000.0 .Numumm0
gnu m 0 u 000 0009 coco Nam: o> m0 NNmz 0N2 oocou on: o >N sou cmm mmazooo

Nmyzpoo uzNamomxu ostmomzN

AmwNsz

mmHuucsoo OGHuNomBH HmsoH>chH can meuucsoo mid HMSUH>HOQH cmm3umn wocmumHa II.O|¢ mqm¢8

1233

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

com .0 «0.2082,.
0N0 0N0 0N0 000.N 00N 000 000 00N.N 00N.0 000 000 ammoo.>N
000.0 000.0 000.0 000.0 00N.N 000.N 000.0 000.N 00N.0 000.0 00N.0 00N.0 000.0 000.N NNmuN
000.0 000.0 000.0 000.0 000.0 NmmumN
000.0 000.0 00N.0 000.0 000.N ocmNmuN
000.0 000.0 00N.N 00N.0 cmuN
000.0 000.0 xmuN

000.0 000.N 000.0 000.0 oNch

000.0 000.0 00N.0 chNmoN

000.0 000.N 000.0 000.0 000.0 Numaaam
000.0N 00N.0N 000.0N 000.0N 000.0 000.0 000.0 000.0 oaox mcom

000.0 mousccom

00N.0 NuNum
000.N 00N.N 000.N 00N 000 000.N 00N mchsw
000.0 000.N 000.0 000.0 000.N wmaNmomso
000.0 000.N 000.0 000.0 000.N 000.N 000.0 000.0 000.0 mommuo
000 00N 00N 000.N 00N 000 000 00N 00N.N 0:000
000.0 000.0 000.0 000.0 000.0 000.N 000.0 000.0 000.0 00N.0 000.0 000.0 000.0 000.N acuaumu
000.N 000.N 000 00N.0 00m 00 mNnaou
000 000 00N.N 000.N 000.0 00N.N 00N 000.N gonna

00N.0 000.0 000.0 00N.0 .NNom.nm
000.0 000.0 000.0 000.0 000.N 000.N 000.0 000.N 000.0 000.0 000.0 00N.0 000.0 00N.N magnum
000.0 00N.0 000.0 000.0 000.0 ncmNcNm
000.N 00N.0 000.0 mNmoNnum

005.0 000.0 000.0 00N.0 000.0 0N0.0 00N.0 umaom

gnu m 4 u :00 0009 0000 N502 o> m0 NN«: 0N2 omcoo on: o >N ado com wmazaou
0092000 ozNamomxm uzNamomzN

NoascNucou0 -u.0-< mqmaa

 

234

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000.0 000.0 «enema
000.0 000.0 000.0 00N.0 00N.0 00N.0 Nazuoz
0N0 0N0 0N0 00N 000.N 00N.N 000 00N 000 000.0 000 0NN 000.N uNuuoNz
00N.N 000.N 000 000 000 00N 00N 000.N 000.0 000 000.N 00N.N uuoNz
000.0 00N.0N 000.0 000.0N 000.0 000.0 .NnuN am:
000.0N 000.0N 000.0 000.0N 000.NN 000.0N .0oNuu 302
000.0 000.0 00N.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.N ocoNumnuuz
000.0 000.N 000 000.0 maanauNoz
000.0 000.0 000.N 000.N 000.N 000.N 000.N 000.0 00N.0 000.N 00N.0 00N.N oooouo:
000.0 oowaz
000.0 000 000.0 000.0 msNuNNsu:
000.N 00N.N 000 000.0 000.N T000 «Ncouousmx
000.0 00N.0 000.0 000.0 00N.0 000.N cstcNuun:
000.0 000.0 00N.N 00N.0 000.0 auNo:
00N.N 000 0N0 00N 000 000.N 00N.0 00N 000.N 00N NN«:
00N.0 00N.0 000.0 «NmNmNn:
000 000.0 Nanmz
000.0 00N.0 00N.0 000.0 0NN.0 Noommmmua:
000.0 000 00N.N 000 000.0 000.0 000.N ;uNaNN
00N.N 000 . 000 00N 0NN.N 000 oNuonNN;
000.0 0N0.N 000.N 000.0 000.0 000.0 000.0 000.0 000.0 cocmnmq
000.N 000.0 000.0 uNwsnx
000.0 000 00N.0 00N.0 magma
000.0 cannon
000.NN 000.NN 000.NN 000.NN 000.NN 000.NN 000.NN 00N.NN 000.NN 000.0 000.NN 000.NN 000.NN cumon
nae m 4 u :00 cups cane Nam: o> 0: NN«: 0N2 ooaoo loo: 0 >N :00 com amszpoo
amazaou uzNamomxm uzNauomzN
NooanNucoo. -s.0-¢ mqm<a

 

235

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000.0 000.0 000.0 oanNmna
00N.0 000.0 000 000.0 «Naoucoa
00N.0 000.0 cmzNoa

V 000.0 000.0 000.0 000.0 «Numm

000.0 000.N 000.0 000.0 00N.0 000.0 000.0 000.0 000.0 000.N .NumuuNsm

00N.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 000.0 00N.0 00N.0 000030
000.N 000 000.0 00N.0 00N.0 00000

000.N 00N.N 000.N 000.0 00N 000.N 000.N 000.N 000 .00 .2 .mm
00N 000.N 000.N 00 mchao .mm
000.0 000.0 000.0 000.0 000.N 00N.N 00N.0 000.0 000.0 000.0 000.N 000.0 000.N :Nmmm
000.0 00N.0 annumN> .m
000.N 000.N 00N.N 000.0 000.N 0N0 00N.N 000.0 moNumc .m
000 000.0 000.0 0NNoaom

000.N 000 000 000 000 .N auumNm
000.N 000.N 000.N 000.N 000.N 000 000.N 00N 000.N 00N.N 00N.0 N00 000.N Naomcmm
00N.0 00N.0 00N.0 00N.0 .N0 Noamm
00N.0 000.0 000030
000.0 mNcuasm

000.0 000.0 uNmmoogm
000.0 000.0 $000 000.0 000.0 coNasmm

. 00N mchau .0

000.0 000.0 00N.0 000.0 000.0 000.N 000.N Nmmzuuom
000.0 000.0 ncoNom

000.0 summ
000.N :uumeum

saw a N o :00 once 0000 Han: o> m0 NNuz 0N2 cocoa on: o >N ago 000 amazooo
0092000 ozNamomxm uzNamomzN

NumscNuaoov -u.0-0 mgmas

 

2365

 

OOOH .mmmum couocmHuo "000mx0 .moHumd I mmHu< oHaocoom Hucon m cuomxo

.mmmum couvcmHmo 0:» mo ucmauuummo 0Hnmmuoouuoo 0:0 0:0 wa¢.m.m

000N .moNumo maNucNum ucoacum>o0 .0.0 0.0.0 couchnmmz

 

NNO

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.OOOH I muuomIcmwzuwn 00cmumHa .00Humo 003mmuvocmmoo N>wz 0:0 mo acmauummmo NHV "mousom

000.N 00N.N 000.N oNnamN

000 00N 000.N 000.N 00N.0 00N 000.0 000.N 000 000.N muNmN

000.0 000.0 00N.N 000.0 000.0 00N.0 000.0 00N.0 00N.0 0N>0N00000

ONO.N 0Hwanwcm>

00N.0 00N.0 000.0 000.0 000.0 000.0 000.0 000.0 .m.m.m.0

000.0 000.0 00N.0 000.0 00N.0 00N.0 000.0 000.0 00N.0 000.0 000.0 00N.0 000.0 000.0 .¢.m.0

. 000 000 00N 000.N 00N 00N 00N.N 000.0 00N 00N.N 000.N .> 00000

000.0 000.0 000.0 000.0 000.0 000.0 000.N 00N.0 000.0 00N.0 00N.0 000.0 000.N .0.0

000.0 000.0 000000

00N.N 000.0 000.0 000.0 00N.0 000009

00N.0 00N.0 0N0.0 000.0 000.0 0N0.N mNmN009

000.0 000.N .009+.0N09

000 00N 00N 000 000 00N.N 0N0 0N0 000.N 00O9
. 0 . . . . a 000 0 .05 .o .m .0 00:0 . 0 . .> .80 .00

a 0 m 0 0 0 0 9 0 00 a: > 0 NN«: N2 0 0 z 0 N 0 0 0092000

0092000 02N9momxm 02N9momzN

20000N00o00 II.0I< mNm¢9

 

TABLE A7.-- Results of the test of the Linder model for

237

individual AAS countries

1. Cameroon

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Overall
CSTa) Regression coefficients regression Dggr.
group free-
b) Value wiiggt t p R2 p~ dom
Inter- C 9.021
mediate PCI - .193 - .115 - .728 .471
goods D - .526 - .243 -1.744 .088 .5028 .0005 43
(C1) L 1.145 .207 1.437 .158
P 2.958 .432 3.509 .001
Consumer C 17.073
goods PCI - .339 - .146 -1.283 .207
(C2) D -1.702 - .562 -4.993 .0005 .6126 .0005 42
L 1.577 .209 2.073 .044
P 2.137 .223 2.092 .042
GO-Cl C -9.675
‘ PCI - .404 - .320 -2.000 .058
D 1.792 .778 4.665 .0005 .6288 .0005 23
L .977 .240 1.430 .166
P 3.609 .822 5.294 .0005
62 C 8.030
PCI - .020 - .016 - .071 .944 ‘
D - .584 - .356 -1.900 .067 .2050 .109 32
L .267 .062 .303 .764
P .876 .174 .933 .358
G6—C1 C 9.100
PCI - .604 - .254 -l.305 .204
D - .551 - .193 -1.209 .239 .6304 .0005 24
L 3.118 .396 2.368 .026
P 2.591 .306 2.054 .051
GG-CZ C 13.957
PCI - .702 - .279 -l.303 .206 .
D -1.305 - .463 -2.450 .023 .6920 .0005 22
L 2.424 .293 2.033 .054
P 1.200 .142 1.108 .280

 

 

238

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE.A7.-- (Continued)
1. (Continued)
Overall
Regression coefficients Degr.
CSTa) regression of
group free-
Beta 2
b) Value weight t p R p dom
G7 C 14.370
PCI - .431 - .204 -1.936 .063
D -1.619 - .583 -6.180 .0005 .8034 .0005 27
L 2.598 .359 3.883 .001
P 2.099 .253 2.664 .013
68 C 15.162
PCI - .646 - .228 -1.142 .267
D -1.335 - .466 -2.401 .026 .5678 .002 20
L 1.683 .189 1.195 .246
P 2.187 .225 1.352 .192
2. Madagascar
Cl C 10.456
PCI .105 .071 .401 .691
D -1.062 - .333 -1.970 .056 .3259 .005 37
L 2.567 .471 3.414 .002
P .853 .104 .727 .472
C2 C 11.637
PCI .078 .044 .357 .722 -
D -l.165 - .285 -2.533 .014 .4339 .0005 61
L 4.129 .622 5.955 .0005
P .313 .025 .247 .805 ‘
GO-C2 C 11.824
PCI - .070 - .040 - .286 .776
D -1.056 - .262 -2.085 .042 .3721 .0005 54
L 3.509 .530 4.472 .0005
P .350 .030 .259 .797

 

 

 

 

 

 

 

 

 

 

 

 

239

 

 

 

 

 

 

TABLE A7.-- (Continued)
2. (Continued)
a) Regression coefficients rgvgzgiion Degr.
CST g of
group free-
Beta 2
b) Value weight t p R p dom
G2 C 11.803
PCI - .152 - .110 - .517 .609
‘D -1.194 - .408 -2.091 .046 .3786 .008 28
L 1.546 .295 1.867 .072
P .480 .068 .408 .686
G6-C2 C 18.341 -
PCI .313 .221 1.249 .223
D -2.400 - .807 -4.693 .0005 .6588 .0005 25
L 2.027 .366 3.028 .006
P .665 .090 .734 .470
G7 C 14.439
PCI - .133 - .084 - .670 .512
D -1.900 - .579 -5.100 .0005 .8612 .0005 17
L 3.786 .586 6.279 .0005
P - .441 - .053 - .502 .622
G8 C 18.310
PCI - .206 - .115 - .752 .462
D -2.288 - .663 -4.330 .0005 .7937 .0005 19
L 3.041 .421 3.979 .001
P .520 .060 .509 .617

 

 

 

 

 

 

 

 

 

 

 

240

TABLE A7.-- (Continued)

 

 

 

 

 

 

 

 

3. Senegal
a) Regression Coefficients rovizziion Degr.
CST eg of
group free-
Beta 2
b) Value weight t p R p dom
Cl C 10.765
PCI .272 .144 1.088 .281
D -1.088 - .387 -3.056 .003 .2666 .001 57
L 2.432 .427 3.351 .001
P - .596 - .081 - .650 .519
C2 C 18.854
PCI - .575 - .229 -2.901 .005
D -1.715 - .444 -5.747 .0005 .6618 .0005 65
L 3.785 .494 6.298 .0005
P .602 .058 .733 .466
G0-C1 C 5.515
PCI .602 .315 1.392 .175 ‘
D - .628 - .214 -1.012 .321 .1201 .466 27
L 1.069 .178 .858 .399
P -2.091 - .319 -1.609 .119
GO-C2 C 13.960
PCI - .519 - .220 -1.759 .087
D -1.185 - .385 -3.173 .003 .5489 .0005 37
L 3.657 .533 4.652 .0005
P - .465 - .061 - .500 .620
G2 C 8.777
PCI .128 .069 .357 .724
D - .926 - .384 -2.143 .040 .2072 .115 31
L 1.075 .195 1.114 .274
P .759 .117 .663 .512
G4 C 8.079
PCI .532 .195 1.252 .219
D -1.174 - .409 -2.759 .009 .4132 .001 36
L 4.344 .642 4.690 .0005
P -1.843 - .248 -1.701 .098

 

 

 

 

 

 

 

 

 

 

 

241

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
3. (Continued)

) Regression coefficients rgvgzgiion Degr.
cs'ra . g of
group Beta 2 free-

b) Value weight t p R p dom
G5-C2 C 14.129
PCI -1.440 - .524 -4.347 .0005
D - .674 - .231 -2.010 .055 .7289 .0005 25
L 2.861 .426 3.270 .003
P .007 .001 .007 .994
G6-C1 C 5.432
PCI - .561 - .456 -2.385 .026
D - .136 - .074 - .457 .652 .4896 .003 23
L 1.335 .337 1.929 .066
P .627 .148 .876 .390
G6—C2 C 16.789
PCI - .692 - .282 -2.271 .028 .
D -1.497 - .383 -3.434 .001 .5868 .0005 42
L 2.467 .333 2.861 .007
P 1.669 .191 1.748 .088
G7 C 17.146
PCI - .571 - .250 -2.445 .018
D -1.645 - .489 -5.323 .0005 .6513 .0005 48
L 2.703 .406 4.113 .0005
P - .392 - .047 - .488 .628
GB C 17.288
PCI -1.572 - .572 -4.410 .0005
D - .878 - .247 -1.854 .073 .5932 .0005 32
L 1.980 .242 1.889 .068
P .382 .041 .333 .742

 

 

242

TABLE A7.-- (Continued)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4. Mali
a) Regression coefficients rgvizziion Degr.
CST 9 of
group free-
. Beta 2
b) Value weight t p R p dom
Cl C 10.610
PCI - .113 - .095 - .411 .686
D - .860 - .476 -1.698 .106 .4917 .009 19
L .728 .143 .611 .549
P .566 .111 .553 .586
C2 C 27.722
PCI - .530 .251 1.163 .261
D -4.022 -1.137 -4.828 .0005 .8205 .0005 17
L - .751 - .087 - .658 .520
P .394 .044 .351 .730
G2 C 9.589
PCI - .200 - .139 - .570 .576
a D - .667 - .347 - .846 .410 .3448 .128 16
L - .045 - .009 - .027 .979
P 1.181 .235 .912 .375
5. Dahomey
Cl C 5.313
PCI - .388 - .404 -1.718 .101
D .003 .002 .009 .993 .4754 .009 20
L 1.137 .311 1.601 .125
P 1.034 .272 1.553 .136
C2 C 17.526
PCI - .493 - .307 -1.798 .090
D -1.772 - .622 -3.831 .001 .7614 .0005 17
L .065 .009 .059 .954
P 1.054 .145 1.099 .287

 

 

 

 

 

 

 

 

 

 

 

243

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
5. (Continued)
Overall
Regression coefficients Degr.
CSTa) regression of
group free-
Beta 2
b) Value weight t p R p dam
GZ C - .289
PCI - .638 - .455 -1.584 .133
D .706 .381 1.345 .197 .4192 .056 16
L 2.558 .521 2.456 .026
P .455 .090 .444 .663
6. Ivory Coast
Cl C 6.026
PCI - .135 - .091 - .675 .503
D - .189 - .092 - .660 .512 .2553 .002 57
L .287 .061 .461 .647
P 2.541 .430 3.319 .002
C2 C 14.634
PCI - .255 - .116 -l.424 .158
D -1.260 - .387 -4.495 .0005 .5434 .0005 79
L 2.867 .404 4.879 .0005
P 1.453 .147 1.725 .089
GO-Cl C - .909
PCI - .847 - .524 —1.569 .133 ,
D 1.096 .605 1.827 .083 .2748 .171 19
L .940 .193 .758 .458
P 2.027 .437 1.687 .108
G0-C2 C 10.196
PCI - .310 - .158 -1.612 .111
D - .735 - .251 -2.411 .018 .3793 .0005 74
L 2.322 .364 3.652 .0005
P 1.294 .147 1.437 .155

 

 

 

 

 

 

 

 

 

 

 

 

244

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
6. (Continued)
Overall
Regression coefficients Degr.
CSTa) regression of
group Beta 2. free-
b) Value weight t p R p dom
G2 C 4.942
PCI - .207 - .152 -1.053 .297
D - .038 - .020 - .132 .895 .1876 .027 52
L .296 .069 .466 .643
P 1.858 .360 2.511 .015
G4 C 6.633
PCI - .247 - .176 - .900 .378
D - .610 - .358 -2.023 .056 .7316 .0005 21
L .540 .122 .869 .395
P 2.797 .616 4.500 .0005
G5-C2 C 12.730 '
PCI - .567 - .197 -1.287 .210
D -1.409 - .411 -2.938 .007 .6980 .0005 24
L 4.691 .557 4.574 .0005
P 1.150 .136 1.085 .289
G6-C1 C 7.785
PCI - .568 - .346 -1.623 .115
D - .319 - .141 - .768 .449 .4602 .001 31
L .547 .105 .618 .541
P 2.559 .440 2.787 .009
G6-C2 C 14.513
PCI - .414 - .179 -2.064 .043
D -1.334 - .415 -4.609 .0005 .6196 .0005 62
L 2.545 .358 4.211 .0005
P 2.129 .229 2.622 .011
G7 C 13.949
PCI - .415 - .213 -2.410 .020
D -1.351 - .503 -5.619 .0005 .7033 .0005 47
L 2.472 .404 4.702 .0005
P .734 .099 1.117 .270

 

 

 

 

 

 

 

 

 

 

245

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
6. (Continued)
Regression coefficients Overall Degr
CSTa) regression of
group free-
Beta 2
b) Value weight t p R p dom
G8 C 15.688
PCI - .436 - .189 -2.206 .033
D -1.588 - .527 -5.935 .0005 .7535 .0005 44
L 2.583 .363 4.323 .0005
P 1.104 .131 1.561 .126
7. Togo
Cl C 4.048
PCI - .368 - .270 - .718 .483
D - .016 - .010 - .035 .972 .l065 .478 16
L 1.028 .250 .923 .370
P 1.045 .205 .691 .499
C2 C 15.366
PCI - .650 - .337 -2.005 .057
D -1.377 - .487 -3.317 .003 .6496 .0005 23
L 1.624 .253 1.930 .066
P .217 .024 .143 .887
GG-C2 C 16.471
PCI -1.098 - .460 -3.281 .004
D -1.384 - .429 -3.247 .004 .7121 .0005 20
L 2.156 .298 2.284 .033
P .454 .025 .179 .860

 

 

 

 

 

 

 

 

 

 

 

 

246

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
8. Chad
Overall
Regression coefficients . Degr.
CSTa) regression of
group free-
Beta 2
b) Value weight t p R p (dom
Cl C .656
PCI - .568 - .448 -1.641 .113
D .780 .367 1.311 .201 .1290 .445 26
L .401 .094 .481 .634 '
P 1.401 .266 1.250 .223
C2 C 9.208
PCI -1.532 - .701 -3.468 .002
D .154 .041 .200 .843 .5371 .001 23
L 1.741 .201 1.327 .198
P 1.080 .108 .605 .551
GZ C - .325
PCI - .669 - .519 -1.946 .063 .
D p .969 .449 1.640 .113 .1685 .290 26
L .459 .106 .555 .584
P 1.611 .301 1.448 .160
9. Gabon
Cl C 9.956
PCI - .446 - .195 -1.715 .091
D - .444 - .190 -1.692 .095 .2615 .0005 65
L 1.721 .357 3.213 .002
P - .110 - .014 - .127 .900
C2 C 15.997
PCI - .296 - .096 - .418 .681
D -1.284 - .476 -2.168 .045 .2985 .174 17
L - .590 - .081 - .367 .718
P -1.317 - .153 - .661 .517

 

 

 

 

 

 

 

 

 

 

 

 

247

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
9. (Continued)
Overall
Regression coefficients Degr.
CSTa) regression of
group free-
Beta 2
b) Value weight t p R p dom
G2 C 7.635
PCI - .292 - .133 - .592 .561
D - .593 - .238 -1.061 .302 .1629 .470 19
L 1.362 .225 1.004 .328
P .315 .052 .223 .826
G6-C1 C 10.494
PCI - .495 - .206 -1.802 .076
D - .487 - .181 -1.567 .122 .2805 .0005 63
L 1.853 .364 3.208 .002
P - .504 - .062 - .548 .586
10. Upper Volta
Cl C 7.204
PCI - .412 - .213 - .993 .330
D - .280 - .121 - .544 .592 .3021 .062 24
L 1.658 .277 1.400 .174
P 1.165 .183 .926 .364
C2 C 18.342
PCI - .595 - .250 -1.785 .089
D -1.900 - .647 -4.672 .0005 .8122 .0005 21
L .430 .056 .471 .643
P .785 .101 .955 .351
G2 C 5.137
PCI - .411 - .212 -1.010 .324
D - .110 - .046 - .208 .837 .3707 .031 22
L 2.395 .401 2.041 .053
P 1.079 .173 .851 .404

 

 

 

 

 

 

 

 

 

 

 

 

248

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
10. (Continued)
a) Regression coefficients rgvizziion Degr.
CST g of
group free-
Beta 2
b) Value weight t p R p dom
G6-C2 C 15.088
PCI - .439 - .201 -1.062 .303
D -1.700 - .630 -3.237 .005 .7358 .0005 17
L 1.072 .149 .970 .346
P - .231 - .032 - .221 .828
11. Niger
Cl C 10.122
PCI .238 .179 A .741 .467
D -1.l32 - .513 -1.810 .084 .2425 .173 22
L - .032 - .006 - .022 .982
P .742 .137 .587 .563
G2 C 7.691
PCI .182 .129 .505 .619
D - .839 - .351 -1.189 .248 .1934 .317 21
L .636 .104 .364 .719
P .724 .125 .465 .647
12. Mauretania
C2 C 17.295
PCI .219 .103 .468 .645
D -2.052 - .408 -2.127 .047 .4063 .034 19
L 3.250 .441 2.184 .042
P -4.851 - .543 -2.535 .020
G0-C2 C 17.168
PCI .218 .103 .465 .647
D -2.036 - .405 -2.113 .048 .4053 .035 19
L 3.251 .442 2.187 .041
P -4.836 - .542 -2.529 .020

 

 

 

 

 

 

 

 

 

 

 

 

 

249

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
13. Central African Republic
Overall
Regression coefficients Degr.
CSTa) regression of
group free-
Beta 2
b) Value weight t p R p dom
Cl C 4.731
PCI - .435 - .334 -1.073 .295
D .092 .046 .155 .879 .3150 .070 22
L 1.032 .228 .960 .347
P 1.452 .298 1.349 .191
G2 C 4.679
PCI -1.132 - .755 -1.942 .068
D .685 .370 1.067 .300 .3858 .056 18
L .397 .090 .331 .745
P 1.899 .417 1.667 .113
14. Congo
Cl C 2.538
PCI - .018 - .012 - .048 .962
D .085 .033 .142 .888 .0223 .965 25
L .822 .155 .666 .512
P - .042 - .007 - .032 .975
C2 C 13.520
PCI - .499 - .233 -1.559 .131
D -1.282 - .399 -3.069 .005 .6823 .0005 27
L 2.425 .363 2.480 .020
P 1.139 .144 1.168 .253
G2 C 5.814
PCI - .379 - .242 -1.003 .327
D - .216 - .072 - .305 .763 .1350 .504 22
L .740 .138 .562 .580
P .127 .021 .098 .923

 

 

 

 

 

 

 

 

 

 

 

 

250

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE A7.-- (Continued)
14. (Continued)
a) Regression coefficients rgvizziion Degr.
CST g of
group free-
Beta 2
b) Value weight t p R p dom
G6-C2 C 9.348
CI - .420 - .319 -1.694 .105
D - .862 - .411 -2.665 .014 .6420 .0005 21
L .879 .193 1.025 .317
P 1.261 .259 1.656 .113
a) The CST code used refers to the following groups of
commodities:
C1 = Intermediate goods and semi-manufactures
C2 = Consumer goods
GO-Cl = Intermediate goods and semi-manufactures in CST
Group 0
G0-C2 = Consumer goods in CST Group 0
G5-C2 = Consumer goods in CST Group 5
G6-C1 = Intermediate goods and semi-manufactures in CST
Group 6
G6-C2 = Consumer goods in CST Group 6.

b) The regression coefficients are designated as follows:

C =
PCI =

Constant

Absolute difference between the per capita
income of the exporting country and that of the
importing country

Distance between the exporting country and the
importing country

Binary variable for identity or non-identity of
languages spoken in the importing and exporting
countries

Binary variable showing whether a trade agree-
ment exists between the importing and exporting
countries.

BIBLIOGRAPHY

 

BIBLIOGRAPHY

Aitken, N.D. and Obutelewicz, R.S. A cross—sectional study
of EEC trade with the association of African
countries. The Review of Economics and Statistics,
November 1976.

Balassa, B. Country size and trade patterns: Comment.
American Economic Review, March 1969.

Balassa, B. European Integration: Problems and Issues.

American Economic Review, Papers and Proceedings,
May 1963.

Basile, A. Commerce extérieur et déve10ppement de la petite
nation. Geneva: Librairie Droz, 1972.

Bhalla, A.S. (EdJ Technology and Employment in industry.
Geneva: InternatIOnal*Labor‘Crganization, 1975.

Blackhurst, R.G. Trade preferences for the less developed
countries: theoryand potential for Latin America.
The UniverSIty 0 Chicago, unpubIIShed Ph. D. thesis,
1968.

Brown, W.B. Market segmentation and international
competitiveness: Trade theory and practice re-examined.
Nebraska Journal of Economics and Business,
Summer 1972.

Cooper, C. and Kaplinski, R. Second—hand equipment in a
developing country — A study ofgjute processing—In
Kenya. Geneva: International Labor Organization,
740

EdiafriC-Service. L'industrie Africaine en 1970.
Bulletin de l'Afrique Noire (Numéro Spécial).
Paris:IEdiafric-ServICe, 1971.

EFTA. The Effects of EFTA on the Economies of Member States.
Geneva: United Nations, 1969.

Fortune, J.N. Some determinants of trade in finished
manufactures. Swedish Journal of Economics,
September 1971.

251

252

Green, R.H. and Seidman, A. Unity or Poverty?
Baltimore: Penguin Books, 1968.

Grubel, H.G. Intra-industry specialization and the
pattern of trade. Canadian Journal of Economics
and Political Science, August 1967.

Herman, B. The Optimal international division of labour.
Geneva: International Labor Organization, 1975.

Hufbauer, G.C. The impact of national characteristics
and technology on the commodity composition of
trade in manufactured goods. In R. Vernon (Ed.).
The Technology Factor in International Trade.
New‘YOrk: NBER, Columbia Universitnyress,Il97O.

International Monetary Fund, African Department Study Group.
Financial Arrangements of Countries using the
CFA Franc. IMF Staff Papers, volume XVI,‘N0.2,
Washington D.C.: IMF, 1969.

International Monetary Fund. Surveys of African Economies
(volumes 1 to 4). Washington BIC}: IMF, 1968.

Johnson, H.G. Economic Policies Toward Less Developed
Countries. WaShington‘D.C.: The Brookings
Institution, 1967.

Kenen, P.B. Nature, Capital and Trade. Journal of Politcal
Economy, October 1965.

Kreinin, M.E. Effects of the EEC on imports of manufactures.
The Economic Journal, September 1972.

Lary, H.B. Import of Manufactures from Less Developed
Countries. *NewfiYork: NBER, Columbia University
Pfess,'1968.

Linder, S.B. An Essay in Trade and Transformation.
New York: JOhnIWiley, 1961.

Lipsey, R.G. The theory of customs unions: a General Survey.
The Economic Journal, September 1960.

Lydall, H.F. Trade and employment. Geneva: International
Labor Organization, 1975.

Mahfuzur Rahman, A.H.M. Exports of manufactures from
developing countries. Rotterdam: Rotterdam University
Press,’l973.

 

253

McClelland, D.H. The Central American Common Market —
Economic policies, economic growth, and choices
for the future. New York: Praeger Publishers, 1972.

 

Meade, J.E. The Theory of Customs Union. Amsterdam:
North Holland Publishing Company, 1955.

Ouattara, A.D. Trade effects of the association of
African countries with the Common Market.
University of PennsyIVania, unpublished Ph. D. thesis,
1972.

Statistical Office of the European Communities. Forei n
Trade-Associates — Yearbook 1959-1966 (14 voiumes,
one for each AAS country), and Foreign Trade —
Associates - Yearbook 1967—1969 (volumes 1 and 2).
Brusseis: Statistical Office of the European
Communities, 1968-1969 and 1970.

 

Statistical Office of the European Communities. Tariff
Statistics — Import - 1964 - Tab.4—5. Brussels:
Statistical Office of the European Communities,
no date.

Tinbergen, J. Shaping the world economy. New York: The
Twentieth Century Fund, 1962.

UNCTAD. Trade in manufactures of developing countries —
1970 Review. New’York:‘Uniteanations,il97l.

 

 

UNCTAD. Trade egpansion and economic integration among
developing countries. New York: United Nations, 1967.

U.S. Department of the Navy, Oceanographic Office.
Distance between Ports. H.O. Pub. No.15l.
Washington D.C.: U.S. Government Printing Office,
1965.

Uri, P. Ed. Trade and investment policies for the seventies -
New challenges for the Atlantic Area and Japan.
New‘York:iPraeger‘PubliShers, 1971.

 

Vinay, B. L'Afrique commerce avec l'Afrique. Paris: Presses
Universitaires de France, 1968.

Viner, J. The Customs Union Issue. New York: Carnegie
Endowment for International Peace, 1950.

 

Wilford, W.T. The Central American Common Market: Trade
patterns after a decade of union. Nebraska Journal
of Economics and Business, Summer 1973.

254

Young, C. Association with the EEC: Economic aspects of
the trade relationship. Journal of Common Market
Studies, December 1972.