,~' .'
, ' ,
' r ' ; ' ,
MACRO ECONOMIC POLICY &STABILIZATION PROGRAM IN A SMALL OPEN ECONOMY:
AN EMPIRICAL ANALYSIS OF THE FOREIGN TRADE PATTERN OF THE
, IVORY COAST (1963 - 1980)
A dissertation submitted to the
Division of Graduate Studies and Research
of the University of Cincinnati
DOCTOR OF PHILOSOPHY
Laisse A. Lahouani
;-
, , :
Maitrise Es Sc. Eco., University of Abidjan, 1976
M.A., Boston University. 1980
TABLE OF CONTENTS
ABSTRACT • • • • •
ACKNOWLEDGEMENTS •
i i i
LIST OF TABLES ••
v
LIST OF FIGURES
vi
CHAPTER I:
INTRODUCTION . . . . · . . . . · . · · · · ·
1
A.
An Overview of the Study's Objective
and Analytical Framework ·
· · · · ·
1
B.
Objectives of the Study
· . . . . · . · · · · ·
5
C.
Pl an of the Study
. . . ·
· · · ·
8
CHAPTER 11:
REVIEW OF LITERATURE ON THE MONETARY-
APPROACH TO THE BALANCE OF PAYMENTS.
la
CHAPTER Ill:
THE ANATOMY OF THE IVORY-COAST'S
ECONOMY
• • . . . . . . . ._~. • . . •
23
,-iF RI C~--I'
.~\\.
'11
A.
General Perspective of the l(:on.omT~. C:~I'." •
23
1.
Economi c St ructure & Res-ou~ces ..""\\. %..~.
23
2.
Forei gn Trade & Econo~tc (po~~es,' ~ ~l •
27
3. ,The Balance ?f.P~yment:S:o/.the r~~-t/
Coast:
A Crltlcal Evailuat~on
• ~/" I);' •
29
B.
The Monetary Sector of the 'lV,Or,y~coas':',e';'~' •
32
1.
An Overvi ew of the Banki'1,'g?~~~~~~mV••
32
2.
Further Development of the Ba'n-!<.:.i'fi'g
Sys tern • • • • • • • • • • • • · .'
34
3.
Interest Rates Policy
• • • •
38
4.
The Development of Specialized
Financial Institutions • • • •
41
CHAPTER IV:
THE MODEL • • •
45
A.
Analytical Framework
• • • • • • • •
47
B.
Implementation of Domestic Credit-Approach to
Stabilization-Program in a Regime of
Fixed Exchange Rate
• • • • • •
51
C.
Balance of Payments & Capital Mobility . • • • •
,53
D.
Specification of the Model • • • • • • • •
57
E.
Comparative Dynamic Results
• • • • • • •
64
CHAPTER V:
STATISTICAL METHODOLOGY &EMPIRICAL RESULTS
69
A.
Statistical Methodology
69
B.
Empirical Results
83
C.
Dynamic Analysis &Stability
Conditions • • • • • • • • • •
97
CHAPTER VI:
SUMMARY, POLICY RECOMMENDATIONS
&CONCLUDING REMARKS
• • • • •
101
APPENDICES:
APPENDIX A:
Stability Conditions of
Equation (4.7)
• • • •
106
APPENDIX B:Computations of the Changes in
the Predetermi:ned Va ri ab 1es 0 f
Equat ion (4.8)
110
APPENDIX C:
Reduced Form Equation of Income
Determination • • • • • • • • •
III
BIBLIOGRAPHY • • • • • • • • • • • • • . . . . . . . . .
113
.-'.
' .....
.ABSTRft.CT
This studyad.~resses the problem of economic in.~~abi.,nty in a small
opendevelopi I:1g economy whi ch has comparative advantage ~,n agri-
.
. .
.
cultural commodities:
The I~ory-Coast.
In order to prescribe an
efficient solution to such instability, the study underwent a deep
review of the ~tructural characteristics of this agricultural-oriented
economy. In addition,
.
some earlier policy recommendations have been
.
.
reviewed.
Putting together the existing elements in a consistent
manner, the study has resulted in designing a stabilization program
suitable to the primary producing country under study.
This stabili-
zation program is a monetary macrodynamic model of the balance of
payments.
In other words, a variant of the traditional monetary
'~ . '
;;.
~
approach to the balance of payments has been designed and. recommended
for the ~i~stability" generated through the prices of primary com-
moditi~s in world markets.
Therefore ·the balance of payments becomes
t.he.foc·ai.p~oint of:theanalytical framework •. MoreoveT, th~ main
..... -.:.. ':'. ",:-;:>.:':.,:,~~~:: ..• r
,,:,,.•
':
•
,
i::_~":"':"
"
."!,
''':
' " . . "
f~.atlil7;~. ;~f. th,~;:r:()d~l ... develOP~d J.nt.he stU.9/i~ ·consi.~~\\ of U~11P9j}'~~!;::;;;~'<
dqin~~t1'~:'C:?I'lPo~:,en~f'Aqurceof~he. ~pney Base as the' POli.cy":_cb'ntr~lle~··'·'
" :; .~..
•
•
: .,~" c
'. :-' ,,"
variable.
That ,is th~ balance of payments. ;s;.brolig~t...into e.quilibrium
~
> '-. - . ~:.~ "'~~:'7t;~ :; -.
.-;
..i·-~r~ .
¥ , "
throughca model that determines the amount of domestic credit to~;be
,'.; ,... " .~i),"1:'
.
injected. i n'to the e~onomy by t~e local moneta ry autho.rit iecs~
~:
" . . . . : :
.
This model was prov~d to be efficient and consistent with the
overall methodological framework.
In effect, using the estimated
'.'.
:. \\.~
coefficients of the structural equations, the system as a whole was
found to be stable.
this dissertation is not the result of th~ author1s efforts alone.
"-
It is ~lsd the result of the help and advice of many other ~eople,
especially the professors on my committee.
In particular, Dr.
Wolfgang Mayer (chairperson) provided me'with considerable'insights
and helpful guidance concerning many aspects of the project which were
not obvious during the process of study formalization and concep-
tualization.
Dr. Ronald Wellington was very cooperative and positi-
vely critical during the many stages of the project.
He, especially,
showed concern about what I am expecting from the project in the near
future.
To him, I owe my sincere gratitude.
I also express myappre-
ciation to Dr. Tsain Fen-Lin and Dr. John Powers for their helpful
advice and remarks.
I owe my special gratitude to the Director of the Institution I
work for in the Ivory-Coast:
"Le Centre Ivoirien de Recherches
Economi~~es et Sociales" (CIRES).
Dr. Achi Atsain made it possible
forme to obtai~ a grant from the Institution to finance all the costs
incurred at all stages of the project.
The accomplishment of this Ph.D. degree has not involved only
,
"
people in the academic arena.
Many friends did bring me support' of
all kinds either directly or indirectly related to the project comple-,.
tion.
extend my sincere appreciation to Cynthia Curtis who, despite
her busy professional schedul~, has been able to find time to type the
i i i
different manuscripts ef the project and more importantly meet ail the
deadlines as promised.
Above all, I owe my greatest debt to my family, relatives and
parents.
My daughter Nadia, missed many years of her father's close-
ness and attention.
Many thoughts go to "Chief" T. Soumanou who pro-
vided. me much needed support at many crucial moments of my isolation
abroad.
My father. Lahouani, Oke i nst ill ed in me a sense of ri gor and
discipline which helped me positively on various occasions and
endeavors.
Most of all, I owe thi s degree to my 1ate mother. whose
lifetime of love and affection was indispensable toward the completion
of my first degree and the Master's degree.
To her memory, I dedicate
this dissertation.
iv
In Memory of My Mother
v
LIST OF iABLES
TABLE Ill-I:
Development of the Ivory-Coast's
GDP:
1960-1980 • • • . • • . • .
24
TAB LE I Il - 2:
Historical Development of Major
Export-Products:
Selected Years
26
TABLE Il 1-3:
Development of the Ivory-Coast's
Trade Balance:
Selected Years
28
TABLE 111-4:
Statistics on BOP &Domestic Credit
31
TABLE I Il -5:
Interest Rates Structure in
Ivory-Coast (%)
.
39
TABLE V-I:
Regression Data ••
87
TABLE V-2:
Required vs. Actual Changes in
Domestic Credit • . • • • .
92
vi
LIST OF FIGURES
FIGURE I:
Policy-Mix Model • • • • • • • • •
55
FIGURE 11:
Simultaneous Adjustment Process in a
Policy-Mix Model • • • • . • . . . •
56
FIGURE 111: Sterilization Policy and Oomestic
Credit Multiplier
• • . • •
89A
CHAPTER I
INTRODUCTION
A.
An Overview of the Study's Objective and Analytical Framework
- - -
The development strategy adopted by most developing countries
,
of West Africa as they achieved independence in the 1960s was mainly
based on short-ruQ and long-run economic planning.
The Ivory-Coast
was not an exception to the practice of such economic strategy.
Indeed, in 1960 a series of ten year projections of the economy
through 1970 was prepared, which called for a continuation of the
policies followed during the final stage of the colonial era.
After
this, came a four-year plan for 1967-1970 and a five-year plan for
1971-75.
The last plan covered the 1976-1980 period.
These different
development plannings have achieved some success in various sectors
of the national economy1.
However, since they have been developed within the framework ~f
the outer-orientation of the domestic economy, their efficiency has
been subject to many questions over time.
Because as the country
becomes more and more engaged in international transactions, domestic
structural problems are more likely to be of external sources than
internal ones.
Hence there results an economic instability related to
the dynamics of the world markets.
This instability is generally
1For a discussion of the evaluation of these plannings see:
Ivory-Coast:
The Challenge of Success:
Bastiaan A. den Tuinder:
The
Johns Hopkins University Press (1978), Baltlmore. -----
2
defined in terms of "Variations in prices, quantum, incomes, export-
earnings and is ascertained by the amplitude of changes in these
vari abl es and the frequence of changes as well,,2.
Therefore a1ter-
native development strategies, mainly in terms of stabilization
programs, have been suggested and carried out by the regional central
bank:
The "Banque Centrale des Etats de l'Afrique de L'Ouest" (BCEAO)
which is an institution of the Monteary Union of \\~est Africa (UMOA)
3
established in 1962 •
The main objective of the stabilization program to be designed in
this study, is to eliminate the "instability" mentioned above.
Therefore it is defined as "a policy-package designed to eliminate
disequilihrium between aggregate demand and supply in the economy,
which typically manifests itself in the balance of payments".
The
attainment of such equilibrium would result in lower domestic infla-
tion rates and improve the country's external position.
In order to implement stabilization programs in each memBer
country, the BCEAO used quantitative rediscount ceilings on short-term
and medium-term assets and liquidity ratios to control the level of
monetary base.
The efficiency of these two policy tools is very much
in doubt as instruments of monetary control in developing countries
2Brown, C.P.:
Primary Commodity Control, Ch. 1., Oxford
University Press (1975).
3The Monetary Union of West Africa (UMOA) was signed in 1Qfi2 by
Benin (then nahomey), the Ivory-Coast, Mauritania, Niger, Senegal.
Upper-Volta.
Togo joined in lq~3 while Mauritania withdrew in 197~.
3
bec~use of excess liquidity of banks.
Indeed, in the p3rticular case
of the West African countries, it has been observed that, using these
two policy tools to control the monetary base, the BCEAO is incapable
of conducting an efficient monetary policy [Onoh, J.K. 1982, Acquah
and Bathia 1977 and 1978].
In the context of the present study, our objective is not to revi-
sit the extent of the accuracy of these criticisms or how they can be
improved, but rather to suggest how the domestic source component of
the monetary base can be effectively used to prescribe an efficient
stabilization program and attain stable balance of payments positions.
In other words, Domestic Credit is our policy-controlled variable and
hence we shall consider a sound internal credit expansion (or contrac-
ticn) by the domestic monetary authorities in order to regulate
pressures on demand and particularly pressures on the balance of
payments.
The analytical framework is a variant of the traditional
monetary approac~ to the balance of payments, as recently developed by
M. Guitian (1973), Marcus Fleming and L. Boissonneault (1960-61),
J. J. Polak (1975), J. J. Polak and V. Argy (1971) for the developing
economies.
Using this variant of the traditional monetary analysis, the
model's dynamic adjustment process of the balance of payments, would
attempt to answer two formal questions:
1.
What are the effects on aggregate demand of changes in the
domestic component of the monetary base,
"
.' ..;--
2.
What are the effects of changes in aggregate dem~nd on
domestic income and the balance of payments.
Our macrodynamic model will explicitly answer these two questions in a
consistent analysis of the balance of payments of a small open deve-
loping economy operating in a regime of fixed exchange rate:
The
Ivory-Coast.
Another important policy prescription of the model,
4
according to some related empirical studies , is that a government must
increase (decrease) the domestic component of the country's monetary
base to finance hudget deficits (surpluses) resulting from balance of
payments deficits (surpluses)S.
From the comparative dynamic results of the model we shall obtain
a specific level of change in domestic credit to achieve such objec-
tive.
Furthermore we shall be able to determine empirically the
offset-coeffi ci ent (aF / aOC = a).
More importantly we expect (a) to be
between zero and minus unity for the assumption of imperfect capital
mobility to hold in the country under study.
The empirical determination of the relevant change in domestic
credit (~OC) becomes an effective monetary policy tool if applied to
support selective quantitative measures [Onoh, J. K. 1982l.
The
4See Chapter 11 for a review of such studies.
5For an argument on similar lines, refer to G. Myrdal: "An
International Economy", (London:
Routledge and Kegan Paul, 1956) and
~ Lirn, "Export instability and economic development:
The example of
Malaysia", Oxford Economic Papers, 25; 1 (1974), pp. 78-92.
5
instrument was first introduced in the context of a developing economy
in 1964 in Nigeria.
It was applied on and off until 1969 when the
Central Bank of Nigeria Credit guidelines were introduced.
Since
1969, it has been applied more frequently.
In applying the policy recommendations derived from ou~macrodyna-
mic model, to prescribe an efficient stabilization program for the
Ivory-Coast's economy, the present study will co~centrate on some spe-
cific objectives.
These objectives are spelled out in the following
section.
B.
Objectives ~ the Study
In looking at the economic structure of the Ivory-Coast, one
might question the revelance ef a stabilization program based on
"Domestic Credit Approach".
Indeed, as a member of many regional eco-
nomic and monetary organizations, such as the Economic Community of
West African States (ECOWAS)7
The Economic Community of West Africa
(CEAO) and the Monetary Union of West Africa (UMOA)6, the Ivory-Coast
maintains special bilateral aqreements with its regional trading part-
ners.
These regional economic agreements are likely to create some
impediments on the effectiveness of any independent economic policy.
However, these agreements have had only a marginal effect upon the
direction and content of the Ivory-Coast's trade, because approximately
6These three organizations gather almost the same member
countries:
ECOWAS has the largest number of participants it gathers
French and English countries in West Africa.
85% of its trade is with industrialized nations.
Moreover with th~
progressive development of the financial institutions in the
...
IvorY-Coast, paralleled by regular revisions of the treaty of the West
African Monetary Union, monetary policy in each member-country is
becoming more and more autonomously determined?
Therefore, any eco-
nomic policy aimed at manipulating a variable which is likely to
interact with some of the regional agreemen~s, would provide not only
the domestic economy but also the entire region with a policy tool
adequate to change or strengthen the regional economic activity.
Hence in order to assist the domestic policymakers to design a con-
sistent stabilization program for stable economic growth, this disser-
tation will focus on the following points:
1.
To provide a statistical analysis of the components of the
balance of payments within the context of the monetary approach:
especially the analysis of the relationship between domestic
credit as the policy-controlled variable on the one hand and
the other predetermined variables:
exports, ~apital flows and
net foreign assets on the other hand, within the constraints
of a balance of payments equilibrium.
In other words, the
.. ".. ~;
.•" >'~
.
effects on aggregate demand of variations in the domestic com-
ponent source of the country's monetary base will be examined.
Such analysis has been recently carried out in short-run sta-
bilization models constructed and utilized for many developing
economies, using the Quantity Theory of Money8.
By relating
f·- •
7See Chapter III for a detailed description of the monetary system
of the Ivory-Coast.
8See Chapter III on literature survey of empirical evidence of
.,-
such models.
.
l.~~.'.::,,~.!,;;:;,:,~.~,{.-.',:~.;:.:::::~:.::.;~,f~;:j~'?'(?C='""~···"""":Y' "'~7 ~"7~~"~'i1~,cl'-i'k'~~~~~~~~~
.•::.'.:•.
•.-,' • •..
..
:
.•..•..
••~. .c.
~- - ~;,>~_:~/~~';,)
. .. ~ .
: .' :::' ':-'
:.:-,~/. .'"
~'~.~"::;"
<:c.
:~:~:''''>'
:"':",
the determination of national income to money base, the model
~:~..
.':-
first seeks to obtai~ empirical evidence that in developing
T,r-
countries, the primary motive for holding money balances is for
"',,:'
transactions purposes and second that the money base (MS) is a
significant variable for income determination in those economies.
2.
To derive empirically "The Domestic Credit Multiplier" which
is also referred to as liThe Offset Coefficient" in the inter-
national economics literature [Dornbusch, 1980, SWOSODA,
A-1973J.
The determination of the offset-coefficient is relevant
to our study, because of its policy-induced impact on the
balance of payments.
In general, it represents the fraction
of any policy-induced change in bank reserves which is offset
through the capital account.
Under such circumstances, if
capital mobility were perfect, then the offset-coefficient
will equal minus one and any increase in bank reserves will
leak out through the capital account.
Consequently the
.-'
<.
:r,
Cl.uthorities lose control over bank reserves.
And with less>
i!;~.:"~'_.
~'.. .
.
than perfect,capit~lmobi1ity, the offset-coef.ficient is bet-'
ween zero and minus one.
In any case, the coeff;cient
r,efl ects a negative ccrre 1at j on between Jor'e,i gn exchiing~ reserves
and the do~estic securities held by the cent~al bank.
Indeed,
the empirical work on offset-coeffici~nt conducted by Kouri
:':." .-
;·~......... l ..
and Porter [1974J supports the view that asset substitution is
.- .
~. ".,
less than perfect.
Therefore, their conclusion is that monetary
policy is efficient whenever used to stabilize ~n economy9 •
. 9Kouri. P. and M. Porter 1974: "International Capital Flows and
Portfolio Equilibrium" Journal of Political Economy Rn.
8
3.
The third objective is an empirical application of the results
obtained in the second step above.
In other words the study
will show that there exists a rate of change in domestic cre-
dit that brings about and maintains equil ibrium in the balance
of payments irrespective of the quantity of foreign reserves
held by the country, the level of its net capital inflows and
its exports.
That is, variations in domestic credit expansion
(or contraction) will be shown to exercise a predictable and
potentially controlling effect on the balance of payments.
4.
Finally from the overall empirical analysis, we shall derive
some policy implications for the future with respect to the
Ivory-Coast's economy.
In this regard we shall present the
merits of the model as to its effectiveness to bring about
domestic economic stability.
These objectives are carried out in an empirical analysis of the
balance of payments of the country under study with a model that keeps
the main characteristics of the country.
Though the model to be set
up in Chapter IV of the project contains substantial modifications of
the traditional monetary approach, its main feature is a preoccupation
with a short-run equilibrium analysis of such approach.
C.
Pl an ~ the Study
The study consists of six major chapters:
Chapter 11 is a review of the existing literature on the monetary
approach to the balance of payments.
We first present the general
'. ,,"
:-~... ~".:'., - -~'.,. '~-"
:'
• • i'·
.
". ;--
concept of the monetary approach to the bal~nce of payments and then
we explore the available empirical works as they relate to our macro-
dynamic model.
Chapter ill presents the anatomy of the Ivory-Coast's economy.
This chapter is divided into two main sections.
The first section
consists of a general presentation of the country's economic resources
and its foreign trade policies.
The second section is an elaborate
discussion,of the banking system and nonbanking financial institu-
tions.
Chapter IV is mainly the presentation of the model suitable to the
particular characteristics of the country under study, as described in
Chapter Ill.
Therefore, a section is devoted to the model's analyti,.
cal framework and its assumptions.
Another section presents its spe~
cification and its comparative dynamic results.
Chapter' ~ is a follow-up of Chapter IV since our concern is to
present the statistical methodology relevant to the' empirical analysis
of·themoael de~igned. ~efore presenting the empirical results; we
briefly·.describe in a' subsection, the data used in the study.
Ina
finar.·sectio"n we 'present 'analytically-our empirical results.
Cha'pter ~ in this final chapter the summary of the whole study is
presented:
Then-we~derive some policy recommendati~n~ and concluding.
remarks, based up6n"'the theoretical framework as developed~n Chapter
IV and the empirical results obtained from the analysis-of Chapter V.
We close this chapter and hence the entire study by commenting on the·
limits of the model and its policy implications.
CHAPTER IT
REViEW OF LITERATURE ON THE MONETARY
APPROACH TO THE BALANCE OF PAYMENTS
Our objective in this chapter is two-fold.
First we present the
monetary approach to the balance of payments as it was developed by
its pioneers [Meade, Alexander, Polak, llohnson, Mundell, Guitian].
Second, we review the available empirical work which deals with the
relationship between domestic credit creation and the balance of
payments akin to the analytical framework developed in the present
study.
More importantly we investigate on the modifications brought
ahout as the monetary approach to the balance of payments is applied
to the developing economies.
The analysis of these modifications
should highlight the relevant assumptions to build up the model for
the developing economies in general and the Ivory-Coast's economy in
particular.
On a theoretical plane, the monetary app~a'c\\ta:Ar0j~f~e, balance of
fC~'-'t- '"
. /1/.
payments is an extention of domestic monetaf~~m~ifl}rnational
n;;o
economy in that it views the balance of
(~t'l1. J~'
pa~.:..me~ts as~aness~ntiallY
J~_. \\
.~' ~Q
r:
r'l;
monetary phenomenon.
In other words, money ,15:z,CO~.
key el ement in
e
Q
d7ev~
l1]e/'1t Sl.lpe~\\
the long-run both as a disturbance and adjustment~too~
in a nation's
balance of payments under a fixed exchange rate regime.
Therefore it
is ascertained that the economy can adjust the nominal supply of money
to what it demands by exporting or importing money through deficits or
surpluses in the balance of payments.
The overall monetary approach
can be stJmmarizerl as follows:
A deficit in a nation's balance of
11
payments results from an excess in the nation's money supply over its
demand.
The excess supply of money flows abroad and represents the
deficit in the nation's balance uf payments.
On the other hand, a
balance of payments surplus arises from an excess demand for money
which is satisfied by an inflow of money from abroad.
After the
excess supply of money has flowed out of the nation or theJnation"s
excess demand for money has been satisified by an inflow of money from
abroad, the deficit or surplus in the nation's balance of payments is
eliminated.
In other words, the concept of money as stock and of
monetary adjustments as adjustments of actual to desired stocks,
constitutes the rationale of the monetary approach.
From the above theoretical presentation of the monetary approach
it can be deduced that a nation's monetary base is composed of two
variables:
An endogenous variable (or foreign assets) which is
uncontrollable by the banking system of a small open economy and the
domestic credit which is the fiduciary issue of the same money base.
Since the latter variable is backed by domestic assets, it constitutes
the policy-controlled variable of the Central bank.
The macrodynamic monetary model used in the present study was
first constructed for economies with highly developed financial
markets [Polak, J. J. and Argy, Victor: 1971J 10 .
In applying such
models to the developing economies, substantial modifications have been
introduced to reflect the characteristics of these economies.
We
identified four basic differences between the two models:
10For a detailed exposition of both types of model see: Credit
Policy and Balance of payments: ,J.,l. Polak and V. Argy: IMF Staff
Papers (1971).
12
1.
In the model constructed for countries with developed finan-
cial markets, the income equation is determined as the sum of
consumption (C). investment (I). qovernment expenditures (G)
and exports (X) less imports (M) or Y = C+I+G+(X-M), whereas
in the other case, income is related to the monetary base, to
reflect the constant velocity of money in the special case of
the developing economies.
That is:
Y = vMB•
2.
In the first model, the capital balance (K) is assumed to be
composed of an autonomous component (V) and sensitive to the
interest rate (R):
K = V + gR.
In the second model, capital
flows are exogenously determined.
3.
The interest rate is explicitly treated as a function of
income (Y) and money base (MS) in the first model, yielding
R = eY - fM B• In the second model the interest rate is insti-
tutionally determined.
In other words, the dynamics of the
money market forces are assumed away.
From the specification
of the interest rate equation in the first model and the
income equation of the second model, the money base equation
in each model can be derived respectively as:
M
e
1
a:
B = f Y - f R
1
b:
MB =- Y
v
Equation (b) is the limiting case where money is totally ine-
lastic to interest rate, so that l/f = 0 (as f~oo) and elf = I/v.
:".
i3,'
4.
Another modification introduced in the two models lies in the
specification of the impo'rt function.
'In,theformer case~ the
~omestic demand ~o~import.is a function of national inco~~
level only:
IMP = mY.
In the latter model, in addition to
the national income level, the other explanatory variable is
the changes in the level of net foreign assets lagged one year
and an autonomous component:
IMP = ~O + ~lYt + ~2(F_I-F_2).
The differences between the two models bring about a set of
assumptions which justify the modifications when the original
model is applied to the developing economies.
The assumptions
to be spelled out below are also valid for the country under
study in the present dissertation.
1.
The country fulfi 11 s the "sma 11-count ry " assumpti on since it
is a price-taker in the world market for its export products
and import-goods as well.
Therefore the country faces in the
world market elastic curves for its export and import-products
and as a result it is'i~capabl:e to affect the world terms of
trade •
., '.
2.
There is little organized capital market because of a lack of
fUlly-de'velopedfinancial institutions ..
The only important
s6urceof financial intermediati~n is the b~nking system.
:~.'
,,;'
Thi~ banking system operates in association with many oth~r
banking institutions outside the national boundaries but
approximately of similar economic structure.
The paucity of
14
the financial sector results in imposing some constraints on
the interest rates and credit allocation.
Therefore the free
play of market forces is assumed away in this sector of the
.
1
11
natlona
economy
.
3.
The lack of a fully-developed market for near-monies, reduces
to an insignificant level, the demand for speculative balan-
ces.
And the holdings of money are predominantly for transac-
tions balances and insensitive to interest rates.
Furthermore, the interest rates are exogenous to the overall
economic structure of the country under consideration.
4.
The existing capital market is imperfect for one main reason.
The interest rate on credit in this market is relatively
sticky, even if the demand for credit exceeds the supply at
the prevailing rate of interest.
Moreover, the capital move-
ments between the Ivory-Coast and the rest-of-the-world are
llFor a discussion of financial institutions in LDCs see:
Stabilization Policies in Developing Countries:
Some Policy
Recommendations: Andrew Crockett: IMF Staff Papers 28(l9Rl) and "The
role of money in stabllizatlon policy in developing countries" Park,
Yung Chul: IMF Staff Papers 20(1973) pp. 337-413.
... , ',~;J
15.;
insensitive to interest rates.
Therefore capital flows are
12
treated exogenously in the study.
Our next objective is to investigate on the available empirical
works on the monetary approach to the balance of payments in the fra-
mework of a macrodynamic model.
Although the model used in the pre-
sent study has received substantial interpretations in the literature,
related empirical works are still scarce, especially in the context of
developing economies.
Therefore our empirical literature survey will
be very limited in scope.
From a theoretical perspective, Herring and Marston [1977] con-
ducted an analysis on countries national monetary policies.
As they
related their analysis to the balance of payments, they observed three
alternatives in the domain of stabilization policy.
First, they
concluded that if the coefficient associated with foreign exchange
reserve (or sterilization-coefficient) were perfect: that is R = -1,
then the supply curve of bank reserves becomes vertical and the
balanc~-o'f payment's e'quilibrium depends exclusively on domestic econo-
mi c factors.:
Second they established that if R = 0, then changes in
fo:rei g~ 'r~e~'e'r'ves' wl11 be refl ected one for one i n ch~ng'~s in bank
reserves.
Finally in the case of the imperfect st~rilization ~l<R<O,
the result is.a higher slope of t~e supply curve of bank reserves
relative to the case in which there is no sterilization,hence
12The exogeneity of the interest rate in the study does not
justify its exclusion from the model's set-up.
However, it has been
overlooked, because of its perfunctory role in the monetary approach
to the bal.ance of payments.
See Chapter 11 on the literature survey
for further evidence.
.
,-.
'0 • •--,
•
,,:'-:.:'
•
,;' ~:
" " -
....
1:7' ,C'
adjusted for new; rijectforl'sa:nd 'withdrawal s ~ 'whe"re Yt represents nomi-
~
--- ".
'nal domestic i'ncome and t:IJAistheamount of net domestic credit
creati.on, white' X , M and Kt are'exports, iniports and net" capital
t
t
{ri~lows res~ecti~ely.
' .
_".
"
0
Furthermore, Bolnick" stated that Polak's model
is nothing else but a reflection of a quantity theory transmission
mechanism:
for changes in the money supply, given the assumption of
constant money demand in relation to income, result in disequilibrium
of cash balances and induce individuals and businesses to look for a
restoration of the equilibrium through changing in spending, leading
to income changes in proportion of the original shift in the stock of
~....
money.
Bolnick's interpreation of Polak's original model has some
common characteristics with this project's model, especially the
reduced form income equation.
Indeed, in both equations, among other
predetermined, variables, domestic credit appears as an explanatory
variable.
Then the cons'i'ant" velocity assumption is a behavioral rela-
tionsh~p and t~e chan~e in the ~oney stock caused by changes in exoge-
no'usvariables'suCh as domestic credit"is the trigger element which in
,;:
·t , r , .
o"t;l' -!
.""
" ,
01-
<?
turn 'c'au'ses changesi n, income.
~:..,~ .:.," _.
, ik'~t~ 'Origirial setting, Pola'k's ~~del and many of its variants
were based on discrete analysis.
Yet, in i'nterpreting the model, ,
Prais, S. J. [1 961J15', '~~forinulated it into continuous an'alysis.
Such
contiriuous an~Jysis, he claimed, removes any ambiguity of
15prai~, S.J.: "S~me mathematical notes on the quantity theory of
money in an open economy" IMF Staff papers (1960-/)1).
'-. '.
"
'.:
'.'/.
...... '•..~ r
.
.
'~~~.~Qes.JL):ar,e .g,iyen,.by.t~~:,tr;ade:balance,.~~~h .~·hat: dL/dt ,:"7 ,.,X:"M".:' where
~,h,~.tlme path.of export.s,(X) j~ exogenously given and.the import func-
tjon•...(M) is determineA :by':)1;-;:::: mYoc-:"The dynami:cs, of the model i·nVolv,es
.
r,
'.'
,
.
•
' , ' F
. '
in~omest money and changes in the ,balance of payments as specified in
our;models and the previpus'ones.
Howev.er. contrary to our model
where domest.i~ 'credi t.creati on is the rel evant pol i cy-vari.abl e t the
c.~u,cial adju.stment variabl.e in Pra;is model is represented ~y .domest,ic
e~pen,djture C" whkh is equal ,tp incomet,.except:. that excess :deman.d.:,(cir
sur;pl,lJs)' of;l i·qui dity. ,will cause peqple to reduce (or i'nc.rease.)
,
',: .~'..,' .---.
.
': .
"
'.
.~,
'-
~
. . . .
,
c,"
"
,spendingaccordingly~ ",:Thts dorriest·i cexpenditure is d'etermi ned as: ','
, :.... ,'t~.~.:·~(;~...~.. ~ ~:,~~~,~~;,f
'.~ _·:; ..::0,~·'~-'·".,··:' "f~;~.' -:'; -...
~-
1
"
. ;
.;..
' .
, ....
•.
,
_
-
, , ' .
" ~{'tf,'~~.:~~~.~V':)fx'rlJ:~~!("..f{~,~.~,';cp.n~J.~~<~:d~ h·is mode1 ,by: defi ning'::D~a~i9pa';, ~:>'~,.' ... '
, :::·-i.ri~ijiti~~~~cf~~-t\\o:~{:,~~ s,:>~,-Y.~~~c.;~·~.;fi- ;r.t~.' :Then; the .f5u·rthen,;~·xP.J aj h.ed·.~'",:/:· .. '. j'
~.' ~~:r~~i~I~:~~~1?~~r ,t~~::f"{-o<" ~;... -:~:.. -:~#~J~·~~~~:i~~\\ ..;-"'~~:
~:. ~.. _" -,.~,~~:.:\\-r-;: .. -,;.". :-.~~~~i~~-.;". ~.~-_. .
'~,;:~:,
:>- .. -
' _ : - •
.·..·~·."~:~"~Zfi?;ia:~JM}i;tfr~ :@ZP·~~A·'"~)##lh,.tUO~e to' that, eYlv4agtcI:.,,-':. ':."
... ;':'!"':;:, f'~j":;~:J.
c,··:'·):~l~lt~!~~i~~::~x::;~:;L ".,,-:.
.~<,c:f~~lai;~~1~~,q.JLij· 4.,0~ftJ~.~(;·~Y,_ft1Hqr'lea~j~~ent o.tone..... '~.,
~;~:~~,.;P.H4l'
."
, c i ..
-_?~ •
-.
...~;:
"
In our (I1odel, the equilibrium income equation has been determined
under the assumption of the quantity theory whereas Prais,'· ~~9del has
e~olved into an explicit statement of the qtantity theory with the-
.... :
adjustment in income coming, about solely through disequi.li~ri~m of
; :.~'f. ..... "
19
cash balances.
There seems to emerge a general consensus as to the many similari-
ties that Polak's original model shares with the traditional quantity
theory of money.
Indeed, the common characteristics shared by the two
models have resulted in empirical studies of the link between money
and economic activity in many countries, especially the developing
ones.
For instance, in his application of the Polak model to the
Nigerian economy, Gray,Clive [1963J16 obtained many results which,
indeed, support the basic thesis of the model.
More fundamentally,
Gray was interested in verifying the degree of economic stability that
could be brought about through the use of monetary instruments to
design a viable stabilization program.
Using net credit creation as
the policy-variable, he defined it as the nation's money supply com-
ponent which increases the total money supply without adding an
equivalent amount to the country's official foreign reserves.
With
regard to the balance of payments, he found that three elements have a
correct estimate of the impacts of credit creation on the balance of
pay~ents.
They are:
- The amount of investment that will result from any given amount
of domestic credit creation,
- The increased output to result from any given amount of credit
creatlon, and
- The favorable balance of payments impact of the increased output.
16 Gray , Clive S. "Credit Creation for Nigeria's Economic
Development": The Nigerian Journal of Economic and Social Studies
p. 247.
Printea-in 12 Pt. Bem50 at--rhe 15adan unTverslty Press.
Nigeria March 1964.
20
Putting together these three elements. he concluded that the
resulting net increase either in exports and import-substitutes leads
to an increase in money income equal to the annual rate of the
increased production times the inverse of the marginal propensity to
import, i.e.: 1/m or the foreign trade multiplier.
He also showed
that the new credit creation generates extra income which is partly
invested and partly saved.
The extra amount invested and saved will
in turn be equal to the marginal propensity to save(s) times the extra
inc~me generated or ~ 6OC. And investment out of extra income thereby
generated amounts to ~ 6DC.
Hence, the total investment induced by
m
domestic credit creation in the model is determined by:
I = ~DC + ~ 6DC = (1 + s/m)6DC.
When all investment becomes fully
m
productive, the output is estimated to increase at an annual rate of
(1 + s/m)~ ; where k represents the ratio between the investment(s)
and the value-added resulting at full capacity utilization.
On the
whole, the empirical test conducted by Gray, supports the notion that
ch
°
anges 1n th e money suppl
y '
Vl a domes to1c
~t'~
.
Cge'I:!~1~~ crea ~1,0n, 1 d
n uce
1:C\\~·~~
spending changes which affect income, im~;.\\:t<~,0n4~~
~~i~~~fe1r~@t balance of
payments positions.
~s 1.'
,
I)
.~
The results of Gray I s study constitute a, subs-t:antj;a;l stabil i zat ion
t:'t-~/bll ~lJ"pe\\t\\~ ,J"
program in the framework envi saged in the prese'iit study.
In effect,
he showed the multiplier effects of domestic credits on the various
sectors of the national economy.
21
Another empirical work related to our area of study is that of
Charles Schotta [1966]17 applied to the Mexican economy.
In his
study, Schotta developed two models to carry out his analysis:
a
classicial monetary model and an income-model.
The income model is a
basic Keynesian income determination set of equations with no monetary
sector, while the classical model is mainly that of Prais as referred
to earlier in the present chapter.
Since he viewed the Mexican eco-
nomy as conditioned. by the export dependency, he argued that the two
models should operate internally to change income in response to
export sector fluctuations.
However, what is important for our pur-
poses is that Schotta claims his classical model is essentially that
of Polak.
Hence, though the system does not assume explicitly that
any autonomous domestic credit creation does take place, it is postu-
lated that changes in money income depend on changes in the stock of
money, that imports depend on income and that the balance of payments
is a major determinant of the quantity of money.
Assessing econo-
metrically the contribution of each model~ Schotta found that the
multiplier theory explains between 44% and 50% of the variance of
national income in Mexico, in contrast to the 70% of the variance
explained by the monetary model.
These statistical results do indeed
suggest that the monetary model is likely to be a better predictor of
changes in income and prices in Mexico than the simple income model.
17 Schotta, Charles Jr.: "The Money Supply, Exports and Income in
an Open Economy: Mexico, 1939-63" Economic Development and Cultural
Change (14) 1965-66: Oct.-July.
22
The cruci al poi nt of .~chg:t.1a ~.~ _mod~Ji.~. ~l!.~J._irl_._a...~)~RQ!J:..:-_Q.cL~nt~d
.------ -._. ---- -_.-._-------
economy, moneta ry p.().li C;:.L1'IJJ.LY tetd an .efft~J~Dt s~~~iJjz31:tj.Qr:L.RJ:.9.9 ram.
_...-'~---
.
-
...
_.~.. ~-
--".-...
Our model is based on the same rationale.
The empirical studies reviewed above reveal certain common charac-
teristics.
In effect, most policy-implications derived from the
empirical tests conducted in the import-sector are based on the speci-
fication of the domestic import demand as a function of national
income level only.
That is IMP = mY.
This specification ignores
other macroeconomic variables likely to play substantial roles in
establishing the import level of a developing economy.
This short-
coming has been overcome in our model-specification by adding the net
foreign reserves-variables to the import function.
This variable is
to explain the extent to which foreign reserves are needed in the
acquisition of import-goods.
Moreover in most models reviewed,
domestic credit was either implicitly or explicitly the policy-
controlled variable for stabilization policy purposes.
The policy-
implications of this variable on the various economic sectors of
rel ated cbuntri es result in IIbuil t-i n-economi c stabi 1i zer".
In
effect, backed by domestic assets, the domestic source component of
the monetary base serves to stabilize the national economy whenever
disturbances originated either internally or externally.
Therefore
from earlier studies, we deduce that a general improvement of the
current condition of the country under study is possible through an
efficient management of local financial resources.
This will allow
the domestic banks to reduce their needs of foreign capital and to
ensure a financing of new. credits necessary for the country's economic
development.
CHAPTER III
THE ANATOMY OF THE IVORY-COAST'S ECONOMY
The analytical presentation of the Ivorian economy that we under-
take in this chapter is divided into two major parts.
The first part
consists of an overview of the Ivory-Coast's economy and th~ second
part is a detailed analysis of the country's monetary sector.
A.
General Perspective ~ the Economy:
1.
Economic Structure and Resources
The history of the country ~~ ~ political and economic unit
dates from 1893 when it became a colony of the French Empire.
With a
population of about eight million inhahitants, of which the foreiqn
community approximates two million, and an annual population growth
rate of 4 percent between 1965 and 1975 and 4.07 percent between 1975
and 1980, the Ivory-Coast is considered as one of the fastest growing
nations in West Africa.
The population census of 1975~ estimates that
by the year 2000 the country's population will be between 14 and 15
million inhabitants.
This population is spread over an area of
521,061.29 square miles.
Despite such a population rise, real income
per capita increased hy an estimated 3.6 percent a year.
In Table
111-1 below, the ratios of GDP to the corresponding population figure
for various years give in column (3) annual per capita GOP.
The sta-
tistics in column (3) indicate that nominal per capita GDP increased
,j
24
by an estimated annual average of 26.46 percent over the period of
18
1960 to 1980.
TABLE 111-1
DEVELOPMENT OF THE IVORY-COAST'S GDP: 1960-1980
Gnp
Population
Per Capita GDP
I
Year
(mill. CFAF)
(mill. inhab.)
(thousands CFAF) i
1
I
1960
142.615
3.735
38.183
I
1961
161.422
3.840
42.037
i,
1962
168.350
3.945
42.674
1963
197.810
4.050
48.842
I
1964
239.675
4.165
57.545
!
1965
239.586
4.300
55.718
~
1966
257.975
4.430
58.234
I!
1967
275.681
4.560
60.456
I
i
1968
I 326.468
I 4.765
68.514
1969
365.568
4.940
74.002
I 1970 I 415.326
I
5.115
81.138
1971
440.074
I
5.264
83.60n
1972
472.480
5.960
79.275
1973
565.267
6.200
91.172
1974
738.661
6.460
114.344
1975
834.545
6.720
124.200
1976
1.113.954
7.022
158.640
1977
1. 539. 265
7.267
211. 816
1978
1.740.562
7.563
230~142
1980
2.223.50
8.189
271.522
r-----l-------..r-----------""----------'
Source:
La Cote d' Ivoire en Chiffres: 1980-81 Ed.
The remarkable increases in nominal per capita GDP can be linked
to the performance of the domestic export sector.
The Ivorian
18The currency of the Ivory-Coast is the CFA Franc, which is
pegged to the French Franc, the intervention CU~iency. at the fixed
exchange rate of CFA Franc 1 = FF 0.02.
The official buying and
selling ratES are CFAF50 =- HI.
ihe eFA Fr-allc floats against the
dollar between 1960 and 1968: US$l = CFAF2GI}-247: between 197
and end
of 1976 US$l = CFAF20S-255 and between 197Q-1QP3: IIS~l
(FAF 55-415.
25
economy is based on a few agricultural commidities such as coffee~
cocoa and timber.
These crops consistute the largest export sector
with a small manufacturing sector which is still nascent.
The impor-
tance of agricultural activities in the Ivorian economy is shown in
the share of the total population engaged in farming activities.
Takin~ account of the farmers living in urban areas, the proportion of
the population classified as agricultural amounts to 80 percent of the
country's population; generating approximately 85 percent of the value
of total exports.
Due t~ the small size of the domestic market and a growing
indust~ial sector, the country depends on foreign markets for the sale
of its agricultural products.
Similarly the foreign markets provide
the home-country with raw materials, finished goods and the capital
equipment necessary for the promotion of the growing local industrial
se~tor.
This externally-oriented structure of the Ivory-Coast's eco-
nomy is the leading source of revenues necessary for the promotion of
i.
.~
th~ countryl's o'verall economic growth.
A critical eva'luation of the
export data repo,rted i.n Table 1r'1-2, g'ives reliable insig~t into the
:Ivorian economic development for selected years.
In effect, from the
observati6n of·thei~ ~tatistics, the following conclusion can be
drawn:
1.
Export volume growth rates are high in most years with an
exceptional acceleration after 1945.
This was mainly the
result of earlier research in cocoa and coffee production.
26
2.
With the high proportion of cocoa, coffee and timber in total
exports (85 percent), higher growth rates for these products
are equivalent to a higher growth rate of GDP.
3.
On the whole, the data indicates that gaining independence
did not disrupt production of agricultural export-commodities
as it did in many other countries.
TABLE 111-2
HISTORICAL DEVELOPMENT OF MAJOR EXPORT-PRODUCTS: SELECTED YEARS
Export -va 1ue I
Exports (tons)
Coffee, Cocoa I
& Timber
Year
Coffee
Cocoa
Logs
(bill. CFAF)
I
I
I
I
I
1900
24
n.a.
13,420
n.a.
1920
17
1,036
46,000
.02
1940
15,610
45,360
23,220
n.a.
1945
37,870
26,940
10,070
.82
1950
54,190
61,690
106,000
13.80
1960
147,500
62,870
640,000
33.30
1965
185,700
126,400
1,500,000
51.80
1970
173,904
179,156
n.a.
92.80
1974
261.486
208.522
3,033.647
177.70
1979
275.000
312,000
1.950.932
348.627
n.a. = not available
Source:
J. Dirck Stryker:
Exports &Growth in the Ivory-Coast:
Timber. Cocoa and Coffee. Yale University, Economic Growth
Center Discussion Paper No. 147 (New Haven: June 1972) and
La Cote d'Ivoire en Chiffres, 1980-81 Ed.
27
nespite the comparative advantage that the country enjoys in the
agricultural sector, the secondary sector of the economy has con-
siderably grown over the past twenty years.
At the time of indepen-
dence in 1960, the share of the industrial sector in Gnp was estimated
at 15 percent.
Thirteen years later this has jumped to 30 percent.
The primary sector whose contribution was estimated at 47 percent
declined to 29 percent over the same time period. This indicates that
other export products grew rapidly as well.
2.
Foreign Trade and Economic Policies
The foreign trade of the Ivory-Coast ~s predo~inently
influenced by its relations with one particular developed country in
the early years of independence:
FRANCE.
Yet, a diversification of
markets and sources of capital emerged sooner.
Thus, though France is
still the main trade partner, the proportion of exports and imports
originating in France between 1960 and 1974 was cut in half.
Over the
same time period, the proportion of trade to the USA and the EC
countries more than doubled.
Also, it should be noticed that the Ivory-Coast's "economy has
grown under the so-called policy of "Liberalism".
It consisted of
granting a host of incentives and exemptions to foreign capital, labor
and expertise in order to develop and promote the local industrial
sector via import substitution.
Furthermore the investment code
granted duty-free import privileges to some priority firms in order to
buy raw materials and other intermediate inputs on the best cV~11able
terms in the world market.
28
The relationship between foreign trade performance and economic
growth can be linked to the development of the terms of trade.
In
effect, in the financing of investments, the development of the terms
of trade is an important factor.
~~_~~!_i_~_~~!~21__~_erms of trade i ndi-
cC!.-t_~_ ~~at...~~~ average price of a countr~t'.~._ba_~.~_e~ o_f._.~~'po_rts_d.i dn_ot
.. -._----~ .
k~_Race with the average price of its imports.
A way to compensate
..
-
.. -._ ..... - ..
.
~. . .
~'"
- ' . "
.....
-
.
.
is to increase exports in volume, but this may be difficult in the
case of the Ivory-Coast since it is one of the world's largest produ-
cers of its main export-products and exporting much more could affect
prices negatively.
In Table 111-3, the deterioration in the terms of
trade is implicitly expressed in the fluctuations of the ratios of
exports to imports.
Such fluctuations in the terms of trade are
likely to give the economy a stop-ana-go character.
The Ivory-Coast's
authorities have been able to counter these fluctuations at times by
establishing'a price stabilization fund for agricultural products, by
drawing on reserves and by borrowing from ab~oad •
.,
TABLE III-3
OEVELOPMENT OF THE IVORY-COAST'S TRADE BALANCE' SELECTED YEARS
(mill. CFAF)
Exp. /Imp.
Year
Exports
Imports
(%)
1960
40.2
36.9
108.9
1965
70.5
62.9
112.0A
1970
130.190
107.704
120.80
1971
126.558
I
110.834
114.10
1972
139.541
114.317
122.6
1973
190.857
157.521
121. 3
I
1974
291.800
232.300
125.6
I
I
1975
254.572
241.392
1OS. 5
I
1976
392.501
311.607
125.9
!
I
1977
:
529.212
I
429.56fi
'
1""\\ '"
1""\\
lLJ.~
1978
524.382
1979
534.847
I 522.502
100.3
I
528.85n
i
101.1
I
1
Source:
La Cote d'lvoire en Chiffres and Calculations
·--~..7=-/,·- ..
29
.....
3.
The Balance of Payments of the Ivory-Coast:
A Critical
Evaluation
Table 111-4 provides a summary of the relevant statistics
on the different components of the Ivory-Coast's balance of payments
from 1963 to 1980.
The main characteristics displayed by these data
are regular surpluses in the Trade Balance.
The balance on current
account registered persistent deficits throughout the years, though
substantial capital inflows did take place.
The persistent deficits of the current account are substantially
due to higher factor payments in the form of debt payments, trans-
ferred profits and foreign workers' remittances.
Since the deficits
in the services and unrequited transfers accounts outweigh the trade
balance account aurpluses, they result in deficits in the overall
current account.
Yet, the deficits on current account, have been offset in most
years by the inflows of capi~al as indicated by the s'tatisti'cs ih
column (3).
Indeed, during the' first decade after independence,
inflows of direct private investment in the Ivory-Coast's economy were
quite significant.'
Moreover, a study by the World Bank in 1977, indi-
cated that long-term borrowings both by the private and government
sectors have been increasing ever since the 1960's and 1970's.19
Given the size of the current account deficits and the level of
net capital inflows, the balance of payments of the Ivory-Coast
19Bastiaan A. den Tui'nder, ibid.
30
registered surpluses in most years except in 1967, 1971-73 and 1975
and 1979-RO.
The determination of domestic credit policy is the responsibility
of local monetary authorities.
Also, the credit guidelines which are
highly influenced by the interest rate policy, have resulted over time
in higher allocation of credit to the private sector.
In other words,
increasing amounts of domestic credit are oriented toward those
priority-activities which are mostly undertaken by the private sector.
Indeed, th~_~ata in column (7) show that the proportion of total
domestic credit allocated to the private sector increased over the
years except_ln 19_C52..
Tile cumulative annual average rate of increase
in domestic credit to the private sector amounted to 20.1R% over the
1963- 80 pe r i od •
I n add it ion, cha nges in pr i v~!~ __ <jQ~?Ji~__~J:_ed-iL_~~ !:e
positive in all years except in 1905 when to~al domestic credit
outstanding decreased by CFAF 2.57 billion.
all years (column 6 of Table 111~4). _This is an indi~atiqn that the
public sector is not keeping pace with the domestic credit policy
, : 1 ,
resulting in a disproportionate increase between credit" available to
i
the public sector and the amount of investment undertaken in that sec-
tor.
The overall effect of domestic credit policy on the balance of
payments has been mixed over the time period covered in the study.
Approximately over one-third of this time period, the balance of
31
Table 11 1-4
Statistics ~ BOP an~ nome~~ ~ed~~
The Ivory-Coast's Economy; 1953-19RO
(annual data in billion of CFA Francs)
I· .... -'- -.---------t--_· --._-----, ..•- ~_..--r----------- -,.._---.._-----.----..._------ .._---_.. .-.
~.~.-
~---:'--1-
-1
( 1)
( 2)
(3)
(4)
(5 )
(6 )
(7)
(R)
(9)
(10)
(11 )
Trade
Balance
Balance
Total
Publ i c
Private
.:.DC
;.nc
l_Y~ilrS
Balance
on CA
on KA
BOP
DC
DC
nc
DC pub./TnC
:. T•DC
Publ i c
Private
.._---"
--- ------
_ .. .....-_.---
~
•.~----
- - - - - - -
1963
9.0n
-4.50
9.30
4.80
27. 44
··8.82
36.27
-.32
1. 43
-5.6B
7.13
6
1%4
10.30
-4.40
8.80
4.40
37.86
-9.09
47.24
-.24
10.42
- .27
10.97
1%5
7.60
.10.40
14.90
4.50
35.29
··9.31
44.72
-.26
-2.57
- .22
-2.52
196fi
11.60
.13.40
16.50
3.10
37.92
-9.99
48.17
-.26
2.63
- .68
3.45
I
1% 7
9.80
.15.60
9.30
-6.30
47.69
··6.75
54.77
-.14
9.77
3.24
6.60
I
1968
25.70
- 1.30
10.00
8.70
55.18
- H1.33
65.51
-.13
7.49
-3.58
10.74
I
1959
2fL 70
- .80
10.30
9.50
70.26
. 8.80
79.06
-.13
15.08
1. 53
13.35
lQ70
17.RO
.16.00
25.60
9.60
76.45
-15.23
92.68
-.21
6.19
-7.43
13.62
I
Iq 11
15.30
.29.10
25.90
-3.20
95.10
-15.25
111. 35
-.17
18.65
-.02
18.67
IQ7?
Z;>.20
.30.20
11.30
-lR.90
122.59
-10.21
132.91
-.OB3
27.59
6.04
21. 56
19n
20.30
.47.50
45.20
-2.30
150.15
-25.22
175.37
-.170
27.47
-15.01
42.46
I
1974
60.00
.29.30
50.30
21.00
205.55
-3R.09
243.74
-.19
55.49
-12.87
68.37
:.
19 7~
Ifi. 70
-85.70
50.30
-35.40
269.20
-25.30
292.50
-.095
61. 55
12.79
48.76
j971i
4/.40
.49.11
56.91
7.8D
370.40
-21.50
3R9.00
-.058
104.20
3.70
96.50
1977
9Q.64
-43.92
BR.52
44.70
529.50
159.10
I -76.00
601. 20
-.14
-54.40
212.20
10 7n
I
I.RR
-117.2R
207.R6
36.57
588.RO
-107.50
690.70
-.18
59.30
- 31. 50
89.50
IQ7'!
!
4.00
-2R4.19
231. 95
·52.24
677.RO
I -114.00
774.10
- . 17
89.00 \\
-6.50
83.40
1911(J
2R.55
-342.R3
311.45
-31.11R
R35.00
~ -54.70
RA5.50
-.065
157.20
59.30
92.40
i
_____.______
L __
I,
._ i
_ . _ _ _ _
_ M _ _. _ . · . _
• • •
"
_ _ 0,, • • • • • • •
SOlJr'c:e'
Item 1-7:
International Financial Statistics/lMF; La Cote d' Ivoire en chiffres/Ministere de I 'economic, des Finances
and du Plan, Ed. (1980-81)
32
-'
payments registered deficits.
The surpluses and deficits registered
by the balance of payments from 1963 to 1980 can be restored either by
changes on the capital account or transfers on current account or by
adjusting current account payments or receipts.
Furthermore, since
foreign reserves are often available in insufficient amoun~s and
suitable external financings are generally lacking, desirable balance
of payments positions can be brought about by adjusting current
account transactions.
B.
The Monetary Sector of the Ivory-Coast
1.
An Overview ~ the Banking System
The Ivory-Coast is a member of the Monetary Union of West
Africa (UMOA).
The treaty establishing the Union was signed in 1962
by Benin (then Dahomey), the Ivor"y-Coast, l"1auritania, Niger, Senegal
and Upper-Volta.
Togo joined in 1963 while Mauritania withdrew in
1977.
An agreement on cooperation between France and-the Unlon mem-
bers was also signed in 1963.
Therefore ever since the early years of
independence, the monetary system of the Ivory-Coast has been closely
related to the treaty of the Monetary Union and the domestic monetary
policy was left to the'responsibility of the Central Bank of the West
African States (The BCEAO).
The monetary system of the Ivory-Coast defined via the Monetary
Union has the following characteristics:
33
1.
A central bank supervised by the Monetary Union, responsible
for defining and implementing monetary policy in the domestic
economy.
2.
A comm<?n curr:~ncy:
The CFA Franc issued by the BCE-AO for all
member-countries.
Even though the. Monetary Union is con-
sidered as a common curren'cy area., the notes issued are
distinguished for each country of issuanc~ by a letter
following the serial number:
Letter A for the Ivory-Coast.
Moreover, separate monetary accounts for foreign reserves and
credits are kept for each member-country.
The existence of
such independent accounts originated in the creation of a
National Committee responsible for implementation of monetary
policy in the domestic economy.
However, the credit-ceilings
are determined by the board of directors of the BCEAO and the
National Committee is responsible for their allocation bet-
ween commercial banks and local firms.
In general the indi-
vidual ceilings on commercial banks rediscount with the BCEAO
are subject to a limit of 50 percent of their ex~ected credit
operation.
The medium-term rediscount ceilings for each
country are established on the basis of the estimated number
of investment projects likely to require medium-term credits.
3.
The BCEAO assures the convertibility of the CFA Franc into
French Franc at a fixed exchange rate.
~if;:"ifimB.~~~_i\\t.~~~
~I'?$2'AY¥i~~~~
_.
__
' . '
. .
,
-
.- . . . _... -
..··1.--
~~•.. -~-...._-_ .. -...-_.... - ~
, - .
.
_ . .
.
. , . -.
,
34
4.
The me~ber-cc~~tries als~ agree to centralize their reserves
in an operations-account with the French treasury and to
maintain the free convertibility within the Union.
2. Further Development ~ the Banking System
The failure of the Monetary Union to control effectively the
money supply in the region and in each member-country via the Central
bank rediscount and liquidity ratio, has led to the reorganization of
the Union several times.
The modifications brought into the treaty
were mainly to provide member-countries with central banking opera-
tions more responsive to their increasing needs for economic develop-
ment and integration.
This is a gradual move towards monetary
autonomy in each meillber"-CoLint ry.
In 1973, when the Monetary Union began to be restructured, a new
level of authority, the Conference of the heads of state was added
above the Council of Ministers.
The former decides on ~atter~ related
to membership in the Union whereas the latter formulates monetary
policy.
Moreover, at the national level, the National Monetary
Committee has been replaced by the National Credit Committee.
The
tasks of the new committee consist, among others, of proposing the
maximum amount of credit to be allocated to each country over a period
of twelve months.
More importantly, it decides on the use and a11oca-
tion of credit between commercial banks and financial institutions and
the treasury as well.
Also. under the new treaty, specific ceilings
on commercial banks and indi~idual
firms were remove~ and the
....
35'
-....; ..
allocation of credit w'as left to the tliscretion of the National Credit
Committee.
In July 1975, more important changes were introduced in the
Monetary Union of West Africa.
They are designed to:
(1)
Ensure a better utilization within the Union of the member-
countries' financial resources
(2)
facilitate financing for the economic sectors that are given
priority in national development objectives, and
(3)
increase the role of national citizens 'in the management of
their economies.
With regard to the first objective, the Union's liquidity level
will be based on the needs of economic development and on the monetary
and foreign reserves position of each member-country and of the Union
as a whole.
In addition, a member government is now entitled to a
BCEAO permanent credit line equivalent to 20 percent of its tax
receipts during the preceding budgetary year.
In the past, the credit
line was set at 10 perceht for ~limited period, but was changed to 15
percent in 1966.
The credit line can now be granted automatically
with no time limit.
With regard to the two other objectives. the BCEAO is provided
with new policy instruments to direct credit allocation.
In January
1976, the control on volume of credit by the BCEAO was changed from a
system based on discount ceilings for banks and individual enterprises
to a system based on allocatinq to each country a ceiling for each
country's recourse to BCEAO resources.
And within this ceiiing,
specific credit-ceilings should be allocated to the treasury, other
,
public institutions, and the private financial institutions.
The reformed BCEAO was more flexible than its predecessor.
While
the new system was supranational in organization, there was flexibi-
lity which allowed a separate application of BCEAO-approved,monetary
policy instruments at state level.
The structure of the banking system in the Monetary Union was
designed to reflect strong political and economic ties with
Metropolitan France.
Therefore in matters of international transac-
tions, all foreign exchange dealings of each member-country were
centralized in the French tr'easury and all foreign transactions of a
non-franc nature were to be conducted at the Paris foreign-exchange.
The balances of these transactions are recorded in the operations-
account (Common Reserves Account).
The monetary agreements that ori-
ginated under this account have the following characteristics:
1.
The BCEAO is committed to deposit in this account all its
foreign exchanges irrespective of the origin of the funds.
2.
The BCEAO has the right to request against th~ CFA Franc, all
foreign assets existing in the Union whether held by pUblic
or private institutions.
3.
In the case of insufficient amount of foreign assets existing
in the Union, the BCEAO can overdraw from its operations-
account any amount of French Franc necessary for its economic
transactions.
:,
4.
The operations-account stipulates a free-convertibility
clause.
In other words, if inflationary pressures are being
experienced by one of the member-countries, this latter
country will be allowed to increase the level of its imports
and thereby meet the excess demand resulting from lflflation.
Another advantage of the free-convertibility of the regional
currency is to protect each country against the adverse
effects of its own economic crises.
The advantages derived from the working mechanism of the
operations-account have been substantially affected however by the
different reorganizations of the Monetary Union of West Africa that
took place in 1973 and 1975 respectively.
Under the resulting new
treaties, the BCEAO is allowed to keep up to 35 percent of its foreign
reserves in monetary assets other than French currency in other
central banks.
This modification ultimately brings in some flexibi-
lity in the financial negotiations of the BCEAO with other countries
in general and in particular with France, since it reduces the power
of the operations-account cont~ary to the earlier stipulations.
Another variable that plays a substantial role in the monetary
sector of the Ivory-Coast is the interest rate.
In the section that
follows, we review how it is d~termined and its impacts on some" econo-
mic sectors.
3.
Interest Rate Policy
The interest rate policy followed by the Ivory-Coast1s
authorities was in line with the policy-objective of the Monetary
Union of West Africa.
In other words, a policy of low interest rates
20
has been designed and implemented by the BCEAO.
Such policy was
believed to achieve three major economic objectives:
1.
The desire to increase the le~el of investment among dif-
ferent sectors of the economy,
2.
The desire to improve the allocation of investments among the
country·s economic sectors,
3.
And finally the need to lower the financial costs so as to
avoid the possible inflationary effects of interest rates
liberalization.
As a result of these policy-objectives, the interest rates range
changed between 1972 and 1975 from 3.5-8.5 percent to 5.5-11 percent
to the present 8.0-13.0 percent.
Moreover, small industrial and agri-
cultural Ivori an borrowers now pay 5.5 pl us a spread of one to three
percentage points for a total of 6.5 to 8.5 percent a year.
Foreign
borrowers were ,charged 5.5 percent plus a spread of five percentage
points fo~a total of up to thirteen percent a year.
In 1974, when
this policy was implemented, the discount rate which is used by the
20S
.
P
.
L· t
erglo
erelra
el e:
Interest Rate Policy in West Africa.
IMF
Staff Papers, March, 1982.
'.
:
: -.-
.. ~..... '.
. ',c.'
' .. ~~ ,
BCEAO as an instrument of credit operation remained at 5.5 percent
during the same period.
This interest rate differential with France
has resulted in allocating about seventy percent of the available
domestic credit to foreign-owned enterprises, thereby making recourse
to foreign borrowing unnecessary for the foreign private sector which
is already allowed a large transfer of funds under the existing policy
of foreign-investment.
Table 111-5 presents a detailed structure of the interest rates in
the Ivory-Coast and their levels in the various-sectors of the
national economy.
TABLE 111-5
INTEREST RATES STRUCTURE IN THE IVORY-COAST (%)
-I
Jan. 1973 to
From
I
Item
Jan. 1973
Jan. 1975
July 1975 .....
!
0
Basic disc. rate
3.50
5.50
Priority: 5.50
Normal:
8.00
Interest Rate charged
by :Comm. banks
Short Term
1.01scountable
4.5-6.00
6.5-9.5
6.50-7.50 (a)
2. Nondiscountable
6.0-9.00
8.0-11.0
13.00 Max. (b)
Medi urn Term
1. 01 scountabl e
5.5-7075
7.25-10.5
6.5-8.50
(c)
2. Nondiscountable
8.0-8.50
11.00
13.00 Max. (b)
...
. I~t. Pd. rin Deposits
17l)em:Depos its
.00-2.50
•00-3.75
.00-3.00 Min •
2. Jime ()eposits
.00-4.50
2.50-6.50
3.25-6.50 Min.
3.Savl";'qs Accts.
3.25
j
4.75
5.-50
Source:
BCEAO
a:
For crop financing and advances on stocks
b:
All other credits
c:
For small and medium scale enterprises run by nationals and
mortgage loans Of less-than CFAFIO million to nationals.
NOTE:
In the new system the distinction between rediscountable and
nondiscountable has been eliminated.
"'--"~",·'r~......-····.~"···· '~To~'"'t~)~~?:~;;~
• • • • •
f '
. , :
The traditional argument held as to the multiplier-effects of low-
interest rates on investments and economic growth in developing econo-
mies has been recently attacked in the economic literature.
Indeed,
McKinnon [1973] has shown that an interest rate-ceiling will increase
the aggregate demand for investment but realized investment level
would decrease as a result of insufficient savings generated by low
interest rates.
The second general implication of interest rate
ceiling is that it creates rationing of available funds among all com-
petitive investors who are willing to borrow at the institutionally
determined interest rates.
Therefore, as the market mechanism is prevented from functioning
by extensive regulations of interest rates, the necessary real credit
expansion will not be forthcoming.
Moreover because of the poor variability of interest rates in the
domestic economy, the bond market mechanism as postulated by Keynes
does not hold good.
In the Keynesian bond model, the price of bonds
rises as the interest rates fall and vice versa.
Since in the economy
under study the interest rate is nearly static over a long period,
market variations in the prices of bonds are not to be ~xpected.
The distortions brought about by the policy of lowinterst rates
have resulted in substantial reorganlzation of the domestic monetary
system 21 .For instahce in 1975, the interest rates in the Ivory-Coast
were aligned with those prevailing in France.
This meant an increase
21 Such improvement has been made possible under the reform of the
Monetary Union that took place in 1975.
'1~'f~~~'''}~3;0~''~~~'7~f1lli1~
-", .'~J '~
, .....
; '
.'
~.:
,.
::-_:
"-,.
of·two oercentaoe ooints in nonpreferrential rates (that is interest
n.,.:,
•.r .;..
0-,;
" .
••••
~~
')~."
~;(~
~.-'_:'
, o J
:
•..•
- ) _ .
. ,
..; . •
' , : - .
' ; : . '
' , . ,
rates ap~lied_to the f{nancing of exports otherthan~~ricultural
.'~':-'
:-~-';'i:' .~:'::.:
'. .~:.:
'~:;:.:~<:., .~~.::~
,/~
'.-.'
exports and ,priority-investments).
Other major refor:1l1s have been ini-
. :
.
"
":
(:.~-._.
.~ .': '::;' ;..~
' .
.
--:
: "
,'-
:
:l
~ .. ,:,: " .~ -. '-;'
. . . .
tiated to stimulate .local savings through higher interest rates on
:.0."
deposits.
In 1973, small deposits began to earn 2.5 ,percent a year
and the highest rate was brought to 6.5 percent.
By mid-1975, the
minimum rate inc~eased to 3.25 percent and the maximum rate on larger
amounts were subj~ct to negotiations.
4.
Th~ Development of Specialized Financial Institutions
In the Ivory-Coast, banking and credit institutions
include four commercial banks, five development banks, two financial
institutions operating in a specific area and two leasing
corporations 22 •
The capital of most of these banks is joint1y owned
by the government of the Ivory-Coast and thei r for.ei gn owners.
In
1979, ,three American banksop~ned offices in the Ivory-Coast:
.
~:.-
. '-- ,
. ; . '" '1:
: j
~,...
• ~
Bar~lays Bank, Citibank and ~hase Manhattan Bank.
Their capital is
.,:
.~ '~" '
~;-=:: ;:' ~
'-';:' ..
2~corlimerdia'l ~;Banks:"c Bariq~~);jnternat; onal ePourTe Connnerce'and
,L I i ndustri e· ,~nC6te'd 'Ivoi re ){:~JGlCI) ;.,Soc i ete general e 'de, Banque en
.Cote "d~'71voire-(SGBC IF,; $Qci,~t'e"liJvojri enn'e de:Barique:( S.rB) and Banque
. Il1t~rlJati9nare Pour.:l'Afd:qu~f:! ,OEC':j.dentale (BIAO)."'"
'
:·:r·"; :"De'velopmenCBanks:- :Ba'nque "Ivo;-rienne'de Developpement
~Industri.al (BIOI); Credit'._de~la Cote d'Ivoire'(CCI);Barique Nationale
pour ~de Developpnient Agricol e (BNDA);' La Compagni e Ffnanci ere de 1a
Cote ~I Ivoire ,(COFINCI); and La Banque Nationale pour l'Epargne and Le
Credit'(BNEC). '
"
Special Fields & Leasing:
La Societe Africaine de Credit
Automobl le (SAFCA); La $oclete Ivoirienne de Financement (SIF); La
Societe Africain de Credit Bail (SAFBAIL); and La Taw Industrial
Leasing Corporation.
r:?lii,i't~~if~~;rJf.r;f~"1{;:T:tr;!'Jw?i)'~I~~~~G:~~;~0~r,;::;~'!2~',;n;;';J7~~~;'i:r~1'~itii{~~~1~~~~,
~:~
-
. , .
_ . .
• ,
< .":
:
•
?
."<.'::-
"" ~
,-..-. ,
...
100 percent American~owned.
Their primary objective is to act as
".-,;
.
.. " ~ .' ,.
financing agents for the local goverr.ment in such big projects as the
f· -: .
'. : .;
j
.;~.,
..~.
iron ore and the paper pulp project.
': ~- ':".' ;<.~
~'"
In addition. there is a public institution. The "Caisse Autonome
dlAmortissement" (CAA) which manages the public debt. holds and mana-
ges public deposits and mobilizes local resources through i~suance of
tax-free bonds.
The government also established in 1968 a fund to provide guaran-
tees for credits to small Ivorian enterprises.
Finally a stock-
exchange was established in April 1976.
Its activities consist of
"ivorianizing" capital invested earlier by foreigners.
Sales of
shares of foreign-owned enterprises. to Ivoirians. have been success-
ful. rising from CFAF 1.3 billion 1975 before the stock-exchange
opened. to CFAF 2.5 billion in 1976.
Ivorianization of bhe banking
and business activities is progressing steadily.
At the end of 1975.
local participation in the equity of commercial and development banks
was 54.5 percent, about one-fifth of it in private Ivorian hands.
"
.,
For commercia.l banks to devise eff'ic1ent tools for the mobiliza-
ffb-ri of 1oc~l;" sa<~1 n;g~~:ih'~~ estab1i shed that savers have hi gh pre-
ference for liquidity and that interests they earn are of s~condary
. '
importance.
This finding is typicil of the attjtude_~isplac~d by
7;-'
people in rural areas.' Alt!houghthey have accepted modern money, a
good ~~m~er o~ ihem still view the.banks with suspicion as a safe
place to deposit their money, because of the impersonal nature of the
banking institutions.
However. a promising initiative in savings
mobilization was the creation in 1975 by the government of the "Banque
Nati onal e~d I Epargne etdeeCredit I; (BNEC) .-: 'Its' capital 1s 100:: percent -
owned -by the· Ivorian government.~, The main' thrust of tM s:natfonal
savings and ·loan 1s to' mobi11:le.1 o'cal savings' in urban: and rural areas'
for re1ending on long-term mortgages.
This'type of-bank is similar to
a savings and loan association and provides customers with_a highly
liquid sa~ings instruments~ An;indi~ation of the increased potential
for mobilizing savings in the int~rior is that the number'of-windows
has increased from 35 in 13 locations in 1964 to 95 in 37 loc~tions irr
1974.
In order to ensure a continuous supply of agricultural output and
raw materials re~uired by industries, efforts were made to establish
agricultural financing institutions.
These instit~tions grant credits
to farmers and raw material producers to expand and modernize thei r
operations.
In 1959. the "Caisse Nationa1e de Credit Agrico1e" (CNCA)
was set up in the Ivory-Coast in a bi d to decentra1i ze and to-reach
areas remote from-the capital-,' Abidjan.
Regional' cf;!ntres'were accor-
;
..
-:
d~~~lY~~st~b1ished~nd known as "Centres Regio~aux de Coordination and
Cooper,ati'on-:'AgI"Jcolen(CCCA)~ : The -:regiona1 ,off.ices' iof the 'CNCA main-~
tafne-df{los~F~i~on'tacts with farmers-:andcooperativ~s who:hadbene-
f1~t'~~:~f;~om ~CCA c~editS, to ensure that credi'ts were not mi sused.
' ..
This fi~!pcing of agricu1turil actiVities has been relinquished by the
-CeCA and it is now carried out by another specialized institution:
The National Agricultural Credit Bank of the Ivory-Coast (BNDA).
The
Ivorian government owns 66.7 percent of its capital.
The desire to promote rapid economic development and to be self-
f
,'.,:=s:ys· _:~_;;,: ~~.::\\-.:~'_:;;::-::;_:'" ?--"';;:~;~::':::.:?-:- -::.~-~ .. _~-.:':''';T.·:':?'''''~~<:7':-~';'*~~~:'~''-'~''':'';''!':'"",7'7~~'''''c?,,,~:,.:~~ ""
_<..:. :<~:::":;':'"~:': '". :0;-;':
..J
>~.~'"'-:';".,
44
reliant in industrial output led to the establishment of sp.v~r~1 deve-
lopment banks in the Ivory-Coast.
For instance in 1964 the "Banque
Ivoirienne de Developpement Industriel" (BIDI) was established.
Its
.
capital is 100 percent owned by the Ivorian government.
Among its
many activities. it provides long-terms loans for the acquisition of
medium and large scale local industries.
These non-deposit financial institutions can cause monetary
disturbances with the funds at their disposal through their lending or
investment patterns.
Therefore the branch of the BCEAO situated in
the Ivory-Coast exercises some control over their lending practices.
It indirectly influences the allocation of long-term loans through its
directorship in the development bank's board and through ceilings on
medium and long-term commitments for each bank.
For development
banks. these ceilings are defined by the 8CEAO as 150 percent of the
sum of the bank's equity and its long and medium term resources.
Medium~term rediscountable loans guaranteed by the government are not
included in the ceiling.
On the whole. we can conclude that the financial infrastructure of
the Ivory-Coast offers some basis for asset portfolio diversification.
":..;
.-; ",:--,~:.
"~" -;"·:"1
;~ " "
CHAPTER IV
THE MODEL
In the context of a monetary approach to the balance of payments,
we develop in this chapter a macrodynamic model for stabilization
policy purposes.
There are five main assumptions that underlie the
model· s set up.
They are expl i citly spell ed out in Chapter II where
we surveyed the available empirical literature on the monetary
approach to the balance of payments in the context of the developing
economies.
The rationale for these assumptions are twofold:
First,
the model is constructued for a small open developing economy
operating in a regime of fixed exchange rate; therefore its power to
influence the terms of trade is very limited if not insignificant.
Second. in such an economy, financial institutions are still rudimen-
tary and inadequately organized to deal with the international econo-
mic structure in which it is engaged.
Thus, there is a lack of
sophisticated monetary tools.
Under these circumstances. the nominal
supply of money is beyond the control of the domestic monetary
authorities.
All they can do is to determine the ex-ante quantity of
money by changing the domestic component of the BASE.
Then the whole
model can be defined in the following terms:
the determination of a
relationship between domestic credit (exogenous variable) and the
balance of payments (endogenous variable) in the Ivory-Coast between
1963 and 1980.
The objectives to be achieved through the model have
been explicitly defined in Chapter I.
46
The structural form of the model yields two estimable equations
for gross domestic income and the national import level.
These two
equations are crucial in determining the reduced form equation of the
balance of payments.
Indeed, in carrying-out the comparative dynamic
analysis of the model, we obtained a reduced form equation of the
balance of payments which is a non-homogenous second-order difference
equation.
Upon total differentiation of this reduced form equation,
we derived the relationship between the balance of payments and the
predetermined variables of the model.
Defining 5uch a relationship as
an equation of the balance of payments and setting it equal to zero,
assures that the balance of payments is in equilibrium.
From this
balance of payments equilibrium, we could solve, for instance, for the
relevant variation in domestic credit which is consistent with such
equilibrium.
In other words, choosing domestic credit (DC) as the
policy-controlled variable and then solving the balance of payments
equilibrium, we derive our fundamental equation which specifies the
relationship between domestic credit, exports, capital flows and net
foreign assets.
This equation is the rationale of the model because
it is at the origin of the policy-implications to be derived from the
whole study.
In the subsequent sections of this chapter, a detailed analysis of
all the model's steps as outlined above will be presented.
~~~~.~!%7Sg:~~~~,1~~'~"""'''if~~~t~=?~~;:~;.c~~s.:::,o:~,.;::;1:S".:;::::~,:.-=:;j:,:~";
:,..
'.'
,
47
A.
Analytical Framework
In seeking viable alternatives to bringing internal and
external stabilization 'to the economies of developing countries facing
a balance of payments constraint, different economic theories have
been suggested.
The openness of these economies to international
trade introduces a potentially additional source of money,
Hence~
money can be supplied by the banking system through credit expansion
and/or purchases of international reserves.
Such alternatives add a
constraint to the policy options available to the domestic policyma-
kers concerned with a stable balance of payments objective.
In the context of this study the analytical framework to develop
the relevant economic policy, to achieve the internal objective of
~ice stability, and the external objective of a balance of payments
equilibrium, will be based on the Domestic Credit policy approach to
the balance of payments.
The principal objectives of such a stabili-
zation program are to reduce the domestic infl~tion rate and improve
the home-country's external position.
The original ver~10n of the model being considered was developed
by Polak. J. J. [l957-5RJand :Polak and Boissonneault [1960J.
Further
developments of the model have been suggested by other economists such
as Guitian, M. [1973J.
The Credit-Approach to the balance of payments stresses the inter-
dependence between the markets for goods and services and for addition
to the stock of money.
In an open economy maintaining a fixed exchange
rate we shall be concerned with nchieving equilibrium in the money
........ ,'. '.',
:-1....'
. ,.~" ..
-
market and in the balance of payments.
These two conditions imply
"
equl 'h i
.1_r.um ' t
,n h
~ e mar ket f or goo d:
s and ser",'ces
•
but not in the market
"
for stocks.
Given a rate of growth in the quantity of money demanded,
an equivalent rate of growth in the quantity of money supply can, in
principle be derived from any combination of changes in domestic cre-
dit and in net foreign assets of the banking system.
In other terms,
an increase in the quantity of money demanded i~ compatible with dif-
ferent balance~ of payments outcomes.
Hence it follows that
equilibrium in money flows does not need to coincide with equilibrium
in the balance of payments.
Specification of the latter as a desired
change in international reserves (i.e. a reserve target) determines
the appropriate change in domestic assets needed to satisfy the
desired change in money holdings.
Furthermore, it has been well established in the literature that
increases in domestic credit to either the private or the public sec-
" tor iffect the balance of paymenis, since these increa~es raise aggre-
gate demand and" especially import dema"nd [Polak and Boissonneault;
.' ~
i"
1966, F]~mming and Bo;~so'nneault; 1961 and Argy; 1969-1971].
In.d~veloping countries where a ,sizeable proportion of government
~-;:
:-
~':... , .
.~.
" -
~xpenditures is used to finance development budgets which are often
characterized~by a high'"foreign~input component, an increase in credit
by. the monetary authorities to the public sector can adversely affect
the Trade Balance and thereby impose pressures on the balance of
payments,
More importantly in many of these developing nations,
~~?·-:~'!:,.~~~;:.~;;:::~.~~z.\\m~\\::~,""~";·ff"'· "i~4:;,:4;'~?'~;"_'~''''''''''-~.>-;·'::;"0j;;7~'''~~~'''4;:~~S;;;;~;;.~~'~S'=:"""$~';';~i~~..:f4:i~ij':~'~k':'::-':;
..
49
increases in domestic credit by commercial banks to the private sector
are generally used to finance imports, often of consumer goods.
Therefore the relationship between changes in domestic credit and the
balance of payments is fairly explicit in many developing countries.
Consequently policymakers concerned with achieving economic growth
while minimizing pressures on the balance of payments would be
interested in the magnitude of the multiplier of domestic credit on
the balance of payments and consequently the effectiveness of the use
of domestic credit as a stabilization policy tool.
In general, assuming no change in the stock of reserves,
equilibrium in the bal~nce of payments requires a rate of domestic
credit expansion expressed as:
n~ d~~ = n~ (~~ + t~) 23 With
given growth rate of output (dY/dt = Y) and rate of inflation
(dP/dt = P), the balance of payments as a proportion of the stock of
reserves, (I/F)(dF/dt), is entirely determined by the percentage rate
of domestic credit expansions, i.e.,:
~ = -dc if ~ = ~ = o.
23 The expression is obtained by taking the time-derivative of the
domestic money market equilibrium, i.e.:
Ms = Md or nc + F = kPY.
The time-derivative of the equilibrium equations yields:
dDC
dF
dP
dY
.. '
----at + df = k( df Y + df P) or:
DC + F = k(YP + YP).
Setting F = 0
and since the elasticity of the demand for money with respect to each
of its arguments is equal to I, we obtain the relevant percentaae rate
of domestic credit expansion, necessary to generate the balance-0f
p~yments equilibrium as:
50
(Foct~cte 23:
continued)
DC = k(YP + PY) = k[YPnp + PYnyJ
Dividing throughout by DC, we obtain:
DC
k
DC
= DC (YP + py)
0.£.0.
NOTE:
By relating the money stock to total expenditures through the
demand for money, we obtain the Cambridge School specification of
equatinn of exchange which stresses the reasons for holding money:
M = kPY where
M
amount of money in circulation at a moment of time
Y
real national output of goods and services per unit of time
p
= price of Y
k
= proportion of PY that the community desires to keep enough to
purchase.
r t\\= kPY
, kP = M/Y
aMI aP = k Y
aM Y
Y
aY M = kP M = ny = 1 •
B.
Implementation of Domestic Credit-Approach ~ Stabil-ization -
Program ~~ Regime of Fixed Exchange Rate:
In a regime of fixed exchange rate, the channels through which
the credit approach operates can be determined by examining the beha-
vior of the different sectors in the economy.
As the economy grows,
an increase in the quantity of money demanded reflects the decision of
the community to use part o! its increasing income and wealth to add
to its money holdings.
This is a saving-decision by which surplus
."~·-·'·;'~·~{:~:t~.:~.~~,·':. ;~~1:.~I~lc;~i'l:;···· ::-. ,"'.,~\\:.tl'-'-" ....-.;.. ..~.
." -
"
'.: ~ :~-. "~'- .; -..0::'-
. ";t ,"
~:." .~.
".
. w• •
seCto'rs~releasereal 'resources to··.ilccomm(ld~te claims on the banking'
. ~ . i _
" . " '
..,,(~
-
' ..-.- -i" ::- ;.~ ._"_"'~ .·~·':·r-.···" ."',.:' .. .;;:~., .:·"~~T:-,
:.
sYste'rri~"".[;'Th-ese 'resources are"'channe1ed to '"t hese'e'conomi c un its whose
expendhu're exceeds 'their income~that 1'5 ·t()'dehCit':~pe~ding units.
Deficit':spendi rig uni ts'horrow~ the surpl ~s"of th~ ~urpl us-s pendi n'g'
units through the banki ng' system~ whi ('h accepts 'thei r pr~mfs~ to pay
in the future.
The credi tex'paris f~n'of the ba'n~i'ng system is there-
fore tied to the availability of resources that the public is willing
to save 'in the form of assets created by ihe'~anking system.
Therefore, the existence of a well-functioning banking system contri-
butes to the separation of the saving and investment functions as
reflected in the decision of the surplus sector to hold money and for
the deficit sector to supply domestic assets.
However, the banking
system can extend credit in amounts different fiom those made
,available to the p~blic.
In this case, prices and the balance of
payment~ will~perform the necessary ~djustments. T~~s, in an open-
. . ~~
economy, credit pol icy becomes an efficient method .of managing' aggre-
.. gate "de'mana and ,the' bal ance ofpayinent~:
.
..
'<':<from"t,hi'S: an~iftical"frame~brk ," i1?ern~rg~es'~t~~{' an/att~mpt by the
d'omestlc'('authorities to reduce or eliminate a deficit or to create or
increase"iasurplus'will require control of domestic credit.
On the
o'rie'h~lI,d, 'resourc'es can be released to expand. ,On the other hand,
imports can be'_.reduced~
~y thi s process, the bal anceof payments out-
come would be consistent with the objectives of the local authorities
and at the same time will satisfy the demand for money.
Given a balance of payments target as an initial condition, the
..... -
52,
:
.' .. '.: ....
• .">~ '. ~~.
• ~- "0 '.'
design of credit policy will be based on theestimationcof the.econn-
my's demand Jor: money ,'Tn'terms ofthe'r:'el evant;,set :oJ·.i ndependent
variables.
Assumptions on, or projections of, thetime,path of the
1atter determi ne - ceteri s - paribus .- the quantity of ;money demanded.
This time path may consist of two different elements: . normal growth
in money holdings in response to normal growth' in the economy, and
adjustments of desired to actual money balances if the two a~e not
initially equal:
That is, if there is.a stock disequilibrium·io the
money market.
This estimated time path combined with the inter-
national reserve target, determines the appropriate domestic credit
expansion.
Therefore, a relationship is estab1ished between dOffiestic
credit expansion and the balance of payments.
This relationship
indicates that excessive (or insufficient) domestic credit expansion
relative to the growth of the demand for money wiJl induce a balance
of payments deficit (surplus).
This transmission mechanism provides a
complex-link between domestic. credit, an~ the balance of payments
through major, variables such ias:exports,:.capital flows~ net foreign
assets and domesticincoine~~:iEhls:""l;"ne,of
reason:ingc:impl.ies a;large
.'
: .
::\\ ~
degree of stability in "the ri1(fn.ey,:ma'~kets and the major'impact of the
adjus~ment process fall s on }heAl ows of expendi ture channel ed through
the ~oods and services "market.'
C.
Balance of Payments and {apital Mobility
So far, the analysis has been conducted with the current
account of the balance of payments.
However, the inflow of foreign
53
capital has contributed and continues to contribute substantially to
the economic growth of the Ivory-Coast.
Indeed, in most years, the
current account deficit has been offset by the inflows of foreign
capita 1.
Because of the stickiness of interest rates in the country
- - -
------.
-_.
----.- -
.
under study, it is assumed that capital movement is imperfe.ct between
the home country and the rest of the world.
This assumption is com-
patible with the economic characteristics of the country under study
since it fulfills the assumption of a small country.
Therefore, the
introduction of capital movements in the model will not modify adver-
sely the overall analytical framework.
Rather, it will strengthen the
relationship between the domestic credit and the balance of payments.
Indeed the composition of the balance of payments would additionally
depend on the conditions prevailing in the gouds and services and
capital markets.
The effects of the capital market on the balance of
payments would be manifested through the supply of foreign capital to
the domestic economy.
Since the balance of payments continues to be
one of the channels through which the supply of money is adjusted to
its demand, the inclusion of capital mobility increases the effec-
tiveness of domestic credit policy to influence the international
position of the domestic economy.
The overall analytical framework can be summarized as follows:
In
an open economy operating under a fixed-exchange rate regime, it beco-
mes possible for the pUblic to change the nominal quantity of money
through international transactions.
As a consequence, the nominal
supply of money adjusts to Uw derna~d for money through the ba;c::ce of
payments.
Thus, the banking system and in particl:lar the monetary
..'.' '-e.""','. "~-'" ""'}}~~~'~'ri-::;.i;:;i~.:
.......
"
author-iti es do not di rectly contre1 the total money supply but, only
, the part that is made' available thr'ough domestic credit.
Furthermore,
the attainment of 'a quantity of money desired by the community is com-
patible with different movements in international reserves that may
not be consistent with the balance of payments objective of the
authorities.
These considerations did indeed lead to an ,emphasis on
the role of domestic credit as the control variable for stabilization
purposes.
The above theoretical analysis can be portrayed in a diagram-
m~tical setting that will depict the balance of payments curve and
capital ·mobility in the framework of a policy-mix model.
Beforehand,
we. should stipulate that the slope of the balance of payments curve
will depend·on the degree ef capital mobility.
If capital is perfectly immobile, then K = 0 and the BP' line
r
becomes(vertical:
this position is independent of the interest rate.
On the other hand" i.f capitali s perfectly mobil e, then K = -~
r
and
the. d<5mest'ic interest rate cannot diverge froin the world interest
. .
.
.
" rate.,~ ::~Gi:ven,;'that·"thf~' i s-:n1'~st·.;li kely-to.apply ··i n the long-run ~ there
. . . .::~~..~:~
.
is:stiJl'scope for an upward-sloping BP curve in the short-run.
.
.
. .
. -
:lndeed, since we ~assume imperfect capital mobility, implying:
_w·< ~r < 0, then the BP curve will be up~ard-sloping.
Secondly, internal balance irLtl}e.modells.referre.d.t.o_.aLtYll
--_
__
..
----- -~.
... -.-_.-
employment level ~nd pric~ s~ab~lity and is identified on the IS-LM
schedules as the vertical line Y = Y :
the IS-LM assume their usual
f
slopes.
',"
.
55
r
BP
~
/
I
LM'
I
~
I ---
I
---
---
I
I
--- --- -- __ IS'
.--- ~--t .---
--
~ .-/
I
~.-
I
LM
\\
I
;1
~v
I
I
I
1
Yf
Y
FIGURE I:
POLICY-MIX MODEL
r:
domestic interest rate
56
r
BP
ISj-
--... --... --... --...
--
--...
---- ---
--
---
---
--- --
---
--
IS=:---
2
--... --...
--- -- --
--
--
---
.-
--- --
--- --
""'"'-
--- ---
---
IS:----.....
---
-
-- d--'" --...
1
--...
-- --
Yf
Y
FIGURE 11:
SIMULTANEOUS ADJUSTMENT PROCESS I~ A
POLICY-MIX MODEL
Suppose the economy is at an initial payments imbalance. that is a
deficit, at a below full employment level of output represented by
point (a).
Under our one-policy variable model, the use of financial
policy would move the intersection of the IS-LM curves from point (a)
to point (b) where the BP curve and the Yr line intersect.
This
policy-mix model presumes that there are two traditional policy
instruments available to hrin(] ithollt it sirnlllti1npnll<; intprnal (1nrl
57
external balance such as at point (0) in Figure 124 •
However, if
fiscal policy is too inflexible to permit effective use or the finan-
cial market in government securities is insufficiently developed to
permit active monetary policy, the system reverts to a one policy-
instrument
such as domestic credit in our case and two objectives to
bring the economy into stability simultaneously.
In the present con-
text, fiscal policy can be used to supplement the automatic monetary
response to payments imbalances.
If the fiscal authorities pursue
internal balance without regard to external balance, the system will
converge to simultaneous internal and external balances at point (b)
in Figure 11.
The resulting BP deficit however induces a bakcward
shift in the LM schedule until external balance is restored at point
(c).
Fiscal policy is_ applied again to restore internal balance and
the system ultimately converges to point (f) where both internal and
external equilibria are achieved.
n.
Specification of The Model
- - - - - -
Numerous variants of the model used i~ the present study have
been suggested to fit the specific characteristics of certain
countries such as the developing ones.
Under a version suggested by
Bolnick (1975) and Franco (1979), the relationship between domestic
credit and changes in net foreign assets and balance of payments has
24 We could view each instrument {monetarv and fiscal Dolicvl as
\\ . , ;
'
... ,
being assigned to a particular objective.
For example expansionary
fiscal policy might be assumed to restore internal equiloibriul1l wio,ile
restrictive monetary policy is assigned to ensure external balance.
58
been made explicit in the case of a small open economy.
However,
given the other variables, the model developed by Bolnick and Franco
shows that there exists a particular rate of change in domestic credit
which will bring about and maintain equilibrium in the bal~nce of
payments.
To achieve such an objective, the model's analysis rests on
the principle of income determinations through the constant velocity
approach of the quantity theory of money.
Therefore it is assumed
that in order to be money income, there must be a quantity of money
circulating in the economy which is related to income by velocity.
Secondly when money is used to buy imports, it disappears from the
money supply.
This is because it is paid over to a commercial bank
(or Central Bank) to finance the purchases of foreign exchange with
which to pay import bills.
It can only be replaced by new money
generated in one of three ways:
exports, capital flows and net
domestic credit creation.
Consequently, the apparent differences of substance between the
model developed by Bolnick and Franco and the original version lie in
the modifications of the standard income determination, reflecting the
special conditions of a developing economy.
Hence, the' model implies
that savings, investments, taxes and government expenditures are not
important factors in determining the behavior of the economy.
By relating income determination to the supply of money, the basic
income mechanism of the model can be specified as:
( 4. 1 )
vIt
"M
{
"'B
. <:t
59
Equation (4.1) embodies the fundamental assumption of "constant velo-
city" since it assumed that all income is spent.
In other words, the
circulation of money stock can be related to the amount of money
expended in the economy during a given time period.
Therefore if we
d
o
th
1
1
f th
oth
dOt -
d
0
eSlgnate Pi as
e average prlce
eve
0
e 1
commo 1 y an
qi as
total quantity sold, then Piqi represents the total money value spent
on that commodity.
For all n-commodities in the aggregate economy we
n
obtain:
+ P q
=
P.q. = PT; where P is the index
n n
L
1
1
i = 1
of the prices in the economy, T as the index of the quantity of goods
traded.
The product PT then represents the total monetary expen-
dituies on all goods sold in the economy.
Since v = PT/M; where M is
the domestic money stock, we obtain the original equation of exchange
as My = PT which is the relationship described in equation (4.1) where
Y is equivalent to PT.
t
Under these specifications we obtain the same results as those of
the earlier specification of the quan~ity theory of money, i.e.:
- The general price level (P) would vary inversely ~ith respect to
the volume of trade done by it,
- it will vary directly, however, with respect to the velocity of
money circulation,
- and finally it will vary directly with respect to the quantity
of money in circulation (M).
60
More fundamentally, Polak statp.s:
"The. aMumption 06 a c.OMtant ~o 06 ..tl1.c.ome. to mone.lf
obv..toU.6ly J.JnYJUeA that the. ..tnc.ome. -6bte.am c.aVll1.ot c.han~e.
e.xc.e.pt a-6 the. qua~y 06 monvJ c.hal'l.ge6.
If, theA.e.~o/te.
we. c.an e.xpla..tn c.hange6 ..tn the. qua~l( 06 mone.lf, we.
would have. a .6a..;t{.,o 6ac.to/tlf e.xolanation 06 ..tnc.ome bl1
me.an.-6 0 6 mo nUaJuj analH.6..t-6."
[Polak, 1957, p. 11].
The assumption of a constant relationship between money and income
can be rationalized by the fact that mQst developing economies lack
significant active markets for near-monies.
This implies that the
speculative demand for money is likely to be minimal if not
nonexistent.
Given that the demand for money balances is largely for
transactions and precautionary purposes, equating these needs to be a
constant fraction of the flow of income seems to describe the struc-
tural characteristics of a typical developing economy very well.
In addition, the model stipulates that injections of primary
importance consist of domestic credit creation, exports and net
foreign assets while the primary leakages are composed of imports.
These injections will affect gross domestic product through the com-
position of the nation's money base.
From assumption (5) stated in
Chapter 11, the monetary oase is specified cs:
61
(4.2)
MS ::: DC + F
where:
M ::: Nation's money base
B
DC = Domestic component of the nation's monetary base
F = International reserves or foreign component of the ftation's
monetary base
The domestic component of the nation's monetary base is the
domestic credit created by the nation's monetary authorities or the
domestic assets backing the nation's money supply.
The international
component of the nation's money base (F) refers to the international
reserves which can be increased or decreased through the balance of
payments surpluses or deficits respectively.
This latter operation is
often expressed as:
oFR = B where B represents the balance of
payments.
The relationship between domestic credit and the balance of
payments in the present model is examined by tracing out the path
along which credits spill over into imports.
In the economy being
considered, it is assumed that there are two sectors which generate
import-demand:
the government and the private sector.
Indeed, in the
Ivory-Coast, like in many developing countries, the government genera-
tes import demand mainly through the development budget. which tradi-
tionally has a high component of foreign inputs.
Similarly the
private sector generates substantial import demand largely composed of
consumer goods which is fuellerl mainly by the extension of domestic
62
credit.
In its structure, the import demand function can be specified
as a function of gross domestic pr~duct and changes in the level of
net foreign assets lagged one year.
The relationship between the
import function and the net foreign assets is based on the assumption
that at the beginning of every year t, when the Central Bank allocates
a given quantity of foreign exchange to the import program, the size
of the program will be determined in large measure by the change in
net foreign assets over the periods t_ 1 and t_ • The other variable
2
of the import function, the gross domestic product expressed in nomi-
nal terms, is determined by the money base via the quantity theory
equation (4.1).
The influence of domestic credit on imports is
generated via the impact of changes in credit on the money supply and
consequently on gross domestic product and hence on imports.
Furthermore, in the model, it is assumed that the magnitude of the
foreign exchange allocation is determined mainly by changes in net
foreign assets during a preceding period.
Therefore the following
import function obtains:
( 4. 3)
In the model, it is assumed that domestic credit has no impact on the
level of exports.
Secondly, though the Ivory-Coast has comparative
advantage in its main export products (coffee, cocoa and timber). the
factors determining the quantity exported are exogneously fixed.
The
consumption of these primary products being concentrated in countries
63
with high per capita incomes, tile world primary commodities prices
( Pw) , and the GNP of the OEe!) - countries (yw) are the explanatory
25
variables of the export supply function
. In other words since we
assume that the economy under analysis is a small price-taking eco-
nomy, the price of its traded goods are determined independ€ntly
of
its domestic monetary aggregates.
Therefore, in the model it is sti-
pulated that the total domestic export represented by (X) is given.
Another relation specified in the model is the net captial inflows
which in turn determine the capital account.
The net inflows of capi-
tal are obtained by summing up the changes in net foreign assets and
imports minus total domestic exports.
This relation gives an identity
which reads:
(4.4)
K
Thus the balance of payments in the model is defined to be equal to
changes in net foreign assets.
Its identity reads as:
(4.5)
or
(4.5')
F
25
~.
~ ~
T ... e
~ t - ~ " s~ " c - 1 -,. , n1· F~ C - r - ~ - c
, , 1
.
.
•
II1
_
a L l
LI
01
:>
'J
" O l l \\ . . < :
u,,-iur,u prif',ory COIn;;ivui"ies prices
and OECn-GNP a's explanatory variables of ttle ,:"xport supply function in
devel opi ng count.ri es has [)-epn delllOnstratpd hv l'lanv ecnnomi sts such as
Adarns and ,]. Rehrlllan (1 r.J7()).
~
~
64
The whole model presented in the analysis can be summarized into the
four linear equations (4.1) -
(4.2 1 )
-
(4.3) and (4.5) with four
endogenous variables and five predetermined variables, two of which
are lagged endogenous variables.
E.
Comparative Dynamic Results:
The notations of the model are:
f)C
Domestic credit
Y
Gross domestic product
t
MB
Nation's money base
F
Net foreign assets
F
and F_
F lagged one year and two years respectively
-1
2
Imp
Total domestic imports as the sum of private and
public imports expressed in domestic currency unit
K
Net inflows of capital
v'~O'~1'~2
Parametric constants, where v represents the velocity
of money, ~l is the marginal propensity to import and
~2 is the impact coefficient of changes in net foreign
assets lagged on imports~
~O is autonomous imports.
The endogenous variables are:
Yt , MS' F and Imp.
The predetermined variables are composed of three exogenous
variables:
X, DC and K and two lagged endogenous variables F_i and F_ '
Z
Equation (4.1) and (4.3) are stochastic behavioral, while equation
(4.2') and (4.5') are identities.
· :--.-' -~"-~~-; .
65
In order to answer the policy questions raised in the objectives
of the dissertation, we ought to transform the structural form of
identity (4.5') into a reduced form and on the basis of this, derive
the relevant dynamic results so implied.
Thus, upon substitution of equation (4.1) and (4.3) into equation
(4.5'), we obtain the reduced form equation of the balance of payments
as:
( 1- ]lZ ) F_I + ]ll _2- ]l1 v0C+X+K- Ilo - ~
(4.6)
F
1+]l1 v
where 11: = -fl1 E:t - nt ·
By totally differentiating equation (4.6) we obtain the rela-
tionship betwee~ the balance of payments and the predetermined
variables.
It is equal to:
or
where:
1- iJ2
lJ2
~lV
1
1
-~-~-
0. 1
1+ lJl v
~
1+ !J v
':t:3
1+ lJ, v
0.11
~
1
1+]l, v
". ','
. '-';' . ":'.
':,;-
.: ",
." ...'.'
,"
f . ~
Setting equation (4.7) equal to zero assures that the balance of
payments is in equilibrium; and solving for dDe which is consistent
with that equilibrium gives:
(4.8)
where:
l-~
112
1
p
::: - -
O ::: -
13
:::
:: -
1
IJ v
2
3
°4
l
l-lV
j:l v
The domestic credit multiplier on the balance of payments can be
expressed as:
a(X-IMP+K)
(4.9)
dDe
Express ion (4.9) can: be rewritten as:
aF
(4.9 1 )
[since anc l ::: 01 . .
where:
aF/aDC is the partial derivative of the balance of payments
with respect to domestic credit.
Taking such derivative from equation
(4.6), we obtain the relevant domestic credit multiplier as:
'.
(4.10)
Empirically, we expect the offset-coefficient (a) to be between minus
\\
one and zero.
That is: -1 < a < O.
And over time, if a can be deter-
mined to approach -1, then we can conclude that capital beeomes per-
fectly mobile in the country under study.
Equation (4.8) exhibits a predictive behavior for the attainment of
a stable balance of payments.
Indeed, it states that the domestic
po l.i<:.'y'm~k~ rs..~~ '1. q~!~rm.i'le. t.h~~b~.ng~._ i.D..d.Q!!l~~tJ.~_cr~~jt '~hi c~. is
necessary tQ bring a..b.oyt .a. bala~ce of payments equiJibI.iull1.' ..g_~v~~.p.~r
ticular values of ne~ foreign assets, exports and neti_n.!.1.o.w~_~f ~api-
tal with the knowledge of the parametric constants.
In other words,
at the beginning of every period t, the policymakers can solve for dOC
by substituting values of dX, dK, dF_ 1 and dF_ 2 into Equation (4.8).
__ If the actual increase in g()m.~~Ji~ cr~dit is equal to dDC, then the
-~;;.~
.~-
---"-
-._- .
resulting balance of payments will be in equilibrium; whereas an
increase in domestic credit less than dDC will generate a surplus.
These specific val ues of the domestic credit resulti n9 in three
different positions of the balance of payments can equally be obtained
from a general interpretation of the domestic credit multiplier
derived in equation (4.9 1 ).
In other words, if the multipler-value is
positive, then an expansion in domestic credit will generate a surplus
in the balance of payments: if, on the other hand, the value is nega-
tive, a deficit will emerge as a result of an expansion in domestic
credit.
Finally, changes ill domestic credit will not affect the
"F"-J':7f~;~~~~:~~~~~;.,~:r~_'~,":~~~~::~~~~'~~r'~X;: ,~'
,
,
68
balance of payments if the muitiplier value is zero.
This latter
position of the ,balance of payments is that of'equilibrium existing in
the economy; thus indicating that the overall economy is in such an
equilibrium position where no further policy-action would result in
any substantial changes.
For empirical analyses, equatiorr (4.1) and (4.3) will be estimated
and the results would be used to evaluate equation (4.7) and (4.8) and
(4.10).
This task is undertaken in Chapter V, where. in addition. we
present the relevant statistical methodology.
The regression analysis will cover the period 1963-1980.
Since
equation (4.1) and (4.3) are overidentified. a two-stage least squares
26
method will be used
•
26Since the model exhibits a dynamic structure, the stability con-
ditions of the equation of the balance of payments will be determined
and presented in Appendix A of the dissertation.
,
'.;;'':==.0."
~ <> ~-~., :.---:--:::,-~.-- .Ji.~-.~ ~- -- -;- - - ~-tl --:- - . - - _ -
( t t
. \\"':~
;or... ..
CHAPTER V
STATISTICAL METHODOLOGY AND EMPIRICAL RESULTS
The objectives of this chapter are threefold:
first, we are going
to seek the statistical methodology appropriate for the model deve-
~
l~ped in Chapter IV.
Second, the empirical results obtained from the
application of the relevant statistical methodology will be presented
and interpreted accordingly, and finally the stability of the system
will be verified, using the estimates of the system of equations.
A.
Statistical Methodology
In order to carry out theoretically and empirically the objec-
tiVes of our study, the theoretical framework developed in Chapter
Ill, has led to the formulation of a system of simultaneous equations
with dynamic structure.
In addition, we adopted a stochastic specifi-
cation of the behavioral equations.
That is, we added a disturbance
term: to the income determination equation (4.1 )an'(j the import demand
function as defined in equation'(4.3).
The pres'ence of the stochastic
erro,rs ;·is justified on ·the grounds that they represent the influence
on there9ressand'of many omitted variables, each with an tndividually
small effect.
With respect to the error-terms~ we assume that the
di_~!':1!~_(tr1~e._,Y.J~~;1:9r:S__oj,_the,..~Y~J~IlL~L~, __no.rl1l.aJ lY._A~~Yi b.':l_~~~ __~~ ~_~,9_1!l_. v.er.. -
tors
..
.-~
with zero mean and. covariance-matrix
-
- - -
"
.-... - -
-.
._.
which is homoskedastic and
non-autoregressive 27 •
We adopted the assumption of homoskedasticity
27 The classical reference for these assumptions is:
Kmenta, Jan:
Elements of Econometrics, MacMillan Series in Economics, Lawrence C.
Klein, Consulting Editor (1971).
70
mainly because we have developed an aggregate model in which macroeco-
nomic data will be used in the empirical estimation.
Furthermore,
homoskedasticity assumQtion is justified on the grounds that, for
--~_.~- - _._-_..------ ---_.- -- -----
instance, the aggregate income function as specified in equation (4.i)
will exhibit variations in incomes of similar magnitudes throughout
the observation period.
The second assumption is justified on the
-----'>-'-- - ~ _.- -- _._
-.- ,..- ~----
basis that we are dealing with annual observations rather t~an monthly
or quarterly data which are often likely to generate heteroskedasti-
city.
It has been observed that a system of simultaneous equations often
exhibits a special feature which is of particular interest in studying
such a system.
That is:
the appearance of endogenous variables amoD9
the explanatory variables of at least one of the structural equations.
Indeed, such is the case for our behavioral equations (4.1) and (d.3).
Similar observations apply to identities (4.2) and (4.5).
Each of
these equations contains at least one endogenous variable as an inde-
pendent variable on the right-hand-side.
Since this endogenous
variable cannot be considered as fixed in repeated samples, we would
conclude that it violates the classical linear regression mold.
Especially, in such a situation, the classical assumption of unbiased-
ness of OLS estimator of the coefficient-matrix (r,~) ;s violated.
Therefore an alternative method of estimation is often suggested.
Usually, three alternative approaches are suggested to estimate the
simultaneous system of equations.
They are:
the naive approach, the
limited information approach and the full-information approach 28 •
In
28F
d
. 1 ' ,.
.
f
or a
etal eel G1SCuSSlnn 0
each appr'cach, see:
Econo:netric
Models, Techniques ann Applications, r~.n. Intriligator, Chapter 11,
pp.
368- 4(0.
Prent i ce -~-Tnc~FnfJl eWf)oo Cl iUs, New ,le rsey (1 q7P,).
71
general, the naive approach gives biased and inconsistent estimators
because of the inclusion of endogenous variables among the set of
explanatory variables.
In other words, it does not distinguish bet-
ween explanatory endogenous variables and included exogenous
variables.
The limited information method includes many specific
~
estimatori such as:
Indirect Least Squares (ILS), two-stage least
squares (2SLS), k-class estimators, and limited information maximum
likelihood (LIML).
These estimators are often referred to as instru-
mental variable estimators (IV).
One of the common characteristics of
these estimators is that they use information as to which variables,
both endogenous and exogenous, are included in the other equations of
the system but excluded from the equation being estimated.
Therefore
they utilize all identifying restrictions which pertain to the
equation being estimated.
As to the full information approach, it
basically consists of estimating the entire system of simultaneous
equations simultaneously.
This approach includes three-stage least
squares (3SLS) and full information maximum likelihood (FIML).
As a result of inconsistency and biasedness of estimators
generated by the naive-approach and the simultaneous e~timation method
of full information, our empirical testing will be based in general,
on the limited information approach: and especially on two-stage least
squares (2SLS).
We chose this estimation technique not only because
it yields consistent and unbiased estimates but also it is cost-saving.
As mentioned earlier, our system of equations contains behavioral
equations and identities as weil.
For the identities, the coef-
ficients are automatically known and hence they do not contain any
stochastic disturbance terms.
72
The model developed in our study consists of a simultaneous system
of four linear equations (4.1). (4.2'). (4.3) and (4.5').
Since we
have four equations and four endogenous variables (Y t • Ms' F and IMP)'
we conclude that the system is complete.
The structural form of our simultaneous system of equations can be
transformed into the following linear form:
Yt - vM S
Et
Ms - DC - F
n
I MP - ilO - ill Yt - il2(F _1- F- 2)
T1t;
F - F
- (X-I u ) - K = n
-1
I-r?
In matrix notation we have:
,.
\\
,"
I
I
1
-v 0
01\\ 1 ' 0 o 0 0 0 n
I r~ I
0
1 -1 o
Ms
0
o -1 0 0 0
X
0
o
r
0
-~ 0 0
0 - il2
DC
[:"1
1 F
il2
o 1 1 JlIMP
o -1 0 -1 -1 0
K
j
I
F-1 I
F
I
" -2;
B
y
+
r
t
X
E
t
(4x 4)
(4xn)
(4xn)
(6xn)
(4x n)
Hence, we can express the above system of simultaneous equations
into the following compact format:
8
y
~
r
v
!I
E
( 5. 1 )
( Gx r, )
(G 'n \\
,l'", ;
(Gxr:)
, v
)
\\"xn
( Gx n)
( 11 xii)
(4xn),
(lIxf))
( f)x n )
(4xn)
'.:;
.
,.,'
';-
.
;;.:
' "
.-t<
.....'
". ~
;'~
-,
..
~~:-.
w~ere:".
rr;":=~:~b~b~f ofob'servati on{ .-
.
". , . '
.·F·,·
-,'
,t,,:~·:;; :.:~~. :: :>:_~C: .;!~:~ .~.1"~'~: ~1~:';'l::~~ :··1-.f..::~ .,,' . ~;. ~
G~:numbir of 'endog~nous variables
"K '=':rid~be~of predetermined variables
X'~ matrix of K predete~ined variables
-
Y = matrix of G endogenous v~riables
B,r = matrices of the structural coefficients
: , "
E = matrix of the stochastic disturbances
With resp~ct to the error-terms we assumed that disturbance vec-
tors of the system are normally distributed random vectors with zero
mean and the variance-covariance matrix is homoskedastic and non-
autoregressive.
In other words:
wh~re~:. is the variance-covariance matrix and it is equal to~
.....
J.~~dimension of ~ reduces ,to (2x2) because, in our system of
~imultaneous equations we have two behavioral equations. These two
behavioral equati~ns are the only ones incorporated in the variance-
covariance matrix.
74
"
~.:.,
......:
In estimating the system of simultaneous equations as specified
above, the first question which is often asked is whether the struc-
tural equations are identif..ie.d...-In general, an equation is said to be
identified if it is either exactly identified or overidentified.
The
identification status of a system of equations is often carried out
under two conditions:
the order condition and the rank condition.
In the case of identification by linear exclusion restrictions,
using the A matrix of all coefficients of the structural form:
the
order and rank conditions are respectively expressed in the following
manner: 29
1.
The Order-Condition
(that is the necessary condition) for iden-
tification stipulates that the number of linear restrictions
imposed on each equation be at least the number of included endo-
genous variables less one:
K ) G-1.
2.
The Rank-Condition
A necessary and sufficient condition for the
identifiability of a given equat_ionof the structural system is
":..!
.. '
that it be possible to form,:at l~ast one nonvanishing determinant
of order G-1 from the columns of the matrix A, corresponding to
the variables excluded~ priori from that equation.
That is:
rank (~ A) = G-1
29 See Fisher, Franklin M:
The Identification Problem in
Econometrics: (1976). Ch. 2, Robert E. Krieger PUblishing Company,
Huntington, New York,
". :'.<"::~:~:,~:';'(W?'~:f~~,':;"~fO'·"'.~~ ..."·":~~~:?:r?\\: ..~~"·::"::~'~;~:~~i~:;';Jf~~f:?:B·'r?;;·;~?~·~",."· -
· ,;,.. "'l:"~";"~~If~l~!t~
..
' . :
, : ' .
_0
'...
,
.. ":~\\~
'
.
0'
>" ~/,
where
- ~ is an NxK matrix whoie elements are known constants such that
each row of • correspond~ to a~linear homogenous restrictions on
the hth equati on of thesystem~ In other words:
~ A = 0
- A is the matrix of the coefficient-matrices (B,r) of the struc-
tural form.
In the case"of ~~clusion restrictions being considered, each
column of • has one and only one nonzero element, that!element being
equal to unity.
And any set of such columns will be independent if
and only if the unit elements come in different rows.
In order to determine the identifiability of the two behavioral
equations of our structural model, we first determine the A-matrix:
BI
A = (rr)
or:
I 1 0 -~1 0
. 1
0
0 \\
I '-v
I
\\
I,
0
-1
0
1
i
I
/,
\\
0
0
1
1
I
I
\\
Ii 0 0 -~O 0
A
\\
=
I
I
;
,
0
0
0
-1
I
!,
!
0
-1
0
0
i,
I
0
0
0
-1
II
Cl
a
- ~2
-1
0
0
~2
i
0
i
,
· The~i'pr1orirestr1ct1ons~~n~t~e'h~~~quation of the -system can be
summar1 zed as: -
~
a
=
o
h
[rh x (g+k) J (g+k)x1
Each row of ~ implies one linear restriction on a •
h
Considering the first behav10ral equation (4.1), involving coef-
,
ficients summarized by the first A column vector aI' all a priori
I
restirctions for this equation can be written in the form:
~1 a1 = 0
For this equation, the necessary and sufficient identifiability
condition is:
rank (~1 A) = G-1.
For equation (4.1) we determine ~lA as follows:
1
0 -\\.11 0
...
001 0 n 00000
,-vII
0
0
0
-1
0
1
..'.
000 1 0 0 0 0 0 0
0
-1 0
1
0
0
1
1
1I
000 0 100 0 n 0
0
0
1 1
0
0
-lJO
0
o 0 0 001 000 0
0
0 -lJO 0
0
0
0
-1
A =
o 0 0 0 0 0 1 00 0
0
0 0
-1
--
0
<P l
-1
0
0
I
o 0 0 0 000 1 n 0
o -1 0
0 I
I 0
0
0
-1
I
I
I
010
0
I
000 0 0 0 0
0 0
-1
i
I
0
0
- Li"
0
!
I
C.
I
o 000 0 0 '"'v " f"l ,
v V. 1
i"
"
I
-1
n
0
IV
\\j
- iJ2
"
I
iJ?
'J
....
I10 0 )J2 0
~~::~~~~r','i~:F?r'~C~'!~~fJX;78~?~~: ·~r~'··'·(~~~·/_/."'7::~(/:.-~:",~~~;::- "';0"'-,; "":.'" ~
.'
.-.."
~ - "'};
l:. ~
•. _~.
::_.-.-
•
. ,.... _r:
Since ranlc.'(~iAf~;T.3 and:'rankC<pi:: 8)'G;,;r, we 'conclude that equation
(4.1) 1s over1de~t1h~d.
Therefore a '2~SLS t~chnique will be used to
estimate it.
Similarly for eqiJation (4".3)"we shall determine 4l A
3 :
0 - u1
0
f-: 1 0 1
-'
0 1 0 0 0 0 0 000
0
-1
0
1
-v
1
0
n
o 0 1 0 0 0 0 0 0 0
0
0
1
1
.0 -1
0
1
A
<P3
=
00000 1 0 000
0
0
-uo
0
0
0
0
-1
000 0 001 000
0
0
0
-1
o -1
0
0
l o o o o o o o o o o . J
0
-1
0
0
0
0
0
-1 j
"
0
0
0
-1
0
0
-u2
-1
0
0
u2
0
rank(~~A) = 3 = G-1.
, Since ranlc.(4l3A) = 3 and ranlc.(~3 = 5 > G-1. we deduce that the
im~ort dem~nd ~quation (4.1) i~ overidentified.
Thus ~ 2 sLs estima-
tion technique will be used for its regression as in thi previous
case.
From the reduced form solution~f equations (4.1) and (4.3) of the
.....
simultaneous system and their identification conditions, we deduced
that these two equations are overidentified.
And henceforth, we chose
two stage least squares method of estimation.
The application of 2SLS technique, for instance. to equation (4.1)
will proceed as follows:
the first stage consists of finding a
A
regressor Ms that resembles Ms' but independent of the error-term £t'
/\\
To find MS' we use OLSto regress the reduced form equation of Ms on
all the predetermined variables of the system that is:
X, DC. K,
F_
and F_ •
From this first stage regression, we obtain the fitted
1
2
A
value of the dependent variable MS to replace MS in the structural
equation (4.1).
Since the predetermined variables are respectively
/ '
,
independent of the error-terms £t and '-, so is M .,
Hence the fi rst
B
A
~
stage allows us to construct a variable (M ) which is linearly related
B
to the predetermined variables of the model and which is purged of any
correlation with the error-term in the income determination equation
/'.
(4.1).
In the second-stage, after substituting M for MS in equation
B
1\\
f\\
(4.1), we obtain y = vM
£*:
where M
B +
B is uncorrelated with the
adjusted error-term (£*).
We can now estimate the parameter (v) by
applying once again OLS regression technique to the structural form
equation y = "
vM
+ £*.
Then we obtain a consistent estimator of the
S
income equation parameter (v).
If additional predetermined variables
appear in the structural equation like in equati.or'\\ (4:3), 2SlS would
also estimate those parameters consistently.
These two steps prove
that indeed, the 2SLS technique is instrumental variable (IV) estimation.
Data Description and Specification ~ Structural Equations ~~ Estimated
So far, we referred to equations (4.1) and (4.3) as the relevant
structural equations of the __ yst.em to hp estimatAd e:npi,icCllly.
They
a re:
79
( 4. I)
( 4. 3)
The explanatory variables and dependent variables of the two equations
above, have been defined in Chapter IV in an explicit manner.
However, it should be noticed that the variable used for foreign
reserves (FR) is the net foreign assets - components of the IMF money
supply defined as M (or money base) for the country under study.
And
1
the variable used for DC is the domestic component of the same HI.
Hence, we obtained the money supply identity (M ) as specified in
1
identity (4.2) in Chapter IV.
These two equations are fundamental to
our study because the estimates of their parameters will help evaluate
the other identities of the system and thereby describe in a systema-
tic manner the overall economic structure of the Ivory-Coast from 1963
to 1980.
The data used in the study consist of yearly observations on the
variables.
The data on Gross Domestic Product (GDP), the balance of
payments, imports and exports (X) are obtained from a publication of
The Ministry ~ Economy, Finance and Planning ~ the
Ivory-Coast:
"La Cote-d'Ivoire en Chiffres" 1980-81 Edition.
However, the data on
domestic credit (DC), net foreign reserves (FR) have been drawn from
the International Financial Statistics published by the International
Monetary Fu~d for various years.
Using the data collected on net
80
foreign assets, (F), imports and exports, we computed the net inflows
of capital as defined in equation (4.4).
All data are reported in
Table 5-1 below and expressed in national currency units:
the CFA
Franc.
The statistics on the GDP show that from 1963 to 1969, the overall
increase in the Ivorian GDP amounts to 84.50%.
Similarly from 1970 to
1979 the cumulative variations in GDP amount to 368.19%.
These two
percentages are indeed an indication of the forward trend of the
Ivorian economy in the first and second decades of its independence.
These positive variations in GOP have resulted in different economic
cycles as portrayed by the positions of the balance of payments (see
Table 111-4:
Chapter Ill) in various years.
Equally appealing are the
figures on the net foreign assets:
from 1963 to 1969, the cumulative
variations in net foreign assets in the Ivorian economy can be eva-
luated at 212.22%.
This is an indication of the high level of foreign
inputs in the development process of the country in its early years of
cooperation with the outside world.
However, from 1970 to 1980, we
observed a sharp decrease in the inflows of foreign as?ets in the
domestic economy.
From a level of 38.60 in 1970, it declines to 2.30
in 1974 and becomes negative in 1976 and 1979.
Among the various
reasons that can be found to explain the lower degree of foreign capi-
tal involvement in the Ivorian economy over the second decade of its
independence, we shall retain the one dealing with the domestic policy
of foreign investment.
Over the first decade of independence, the
development strategy adopted by the local policymakers, allocated a
'~-:-~--::-~--:--:-----:--~'"""""======.-,.="-===~~.,......
..
I
81
host of incentives to foreign investors who invest in infant-
industries or so-called priority-industries.
These incentives ranged
from tariff-protection to higher taxes on raw materials.
In addition.
the investment-code granted duty-free privileges on the imports of raw
-
materials utilized in the production process of those goods and ser-
vices considered to be of basic necessities to the nation.
However.
as the economy reached a certain development stage. especially in the
second decade of independence when most investment contracts signed in
the decade of the 1960's expired. most of the incentfves granted to
foreign investors were either not renewed at all or were reduced to a
level where the total costs of production sharply increased.
Hence.
many foreign investors gradually reduced their domestic production
activity.
And the inflow of new foreign investments has become more
and more insignificant over years.
The data on imports and exports (both include goods and services)
have been both increasing over the sample period.
These statistics
display a peculiar characteristic found in the rate of increase of
of both imports and exports.
Increases in both domestic exports and
imports of foreign goods have been remarkable.
Imports and exports
have been increasing at the same pace.
Although the import-substitu-
tion strategy adopted by the Ivory-Coast in the early years of inde-
pendence has resulted in equipping the country with some
light-industries and thus reduced the import of basic necessities, the
overall import level is on the upward swing.
This characteristic of
i .•;",?:>;~~;~cl~;;e~~L~i":::<::~ ··''''·:;';·:''f~~i~f'~~~"l~~
~: ~'.,
.
. .•~ ...•-;..
imports. in dev~,lop~ng nations can be rationalized in the following
manner,
In the early stages of economic ~evelopment, the growing
domest ie' i ndustri al sector necessitates forei gn raw materi al s for its
op'eration and "fore'ign capital goods are equally required to build up
the country's infrastructure.
The import of these two fact9rs of pro-
duction often equalizes or sometimes outweighs the reduction in
imports brought about as a result of a domestic policy of import-
substitution.
Therefore as the country grows and produces domesti-
cally its basic needs, the needs for its industrial sector increase
equally.
Because both imports and exports increase at about the same
rates, the surpluses that are derived from the Trade Balance are unli-
kely to be large enough to cover deficits generated in the balance of
payments as a result of the negative transactions carried out in other
posts of the balance of payments.
This is especially so wit~ respect
to services and unrequited transfers.
Therefore, it seems that even if a country used successfully all
.,;
the familiar commercial policy tools to reduce its import~ of basic
needs, it isunl i kely'that the same pol icy will be appl ied to the
'imporbi- "ofinateri al s needed in the i ndustri al 'sect'or:: of the" economy
(at 'least in the shclrt-run framework).
However, in an,effort to
cbntrol the couritry's economic activity, the poli~ymakers may imple-
ment financial policy to stabilize the balance of payments via the
domestic monetary sector.
..
--~'T--
-;-
-
~
83
B.
Empirical Results
In the previous sections of this chapter, we presented the sta-
tistical methodology deemed necessary to estimate empirically the
structural equations of the system.
In a second step, we ungertook
the analysis of data to be used in the regression and at the same time
we commented on the cyclical trends of the data and derived some
implications for economic development of the country under study.
Similarly, the balance of payments has been evaluated with regard to
the data trends and an ag9regate stabilization policy has been thereby
suggested, based on controlling some macroeconomic variables of the
domestic monetary sector.
ihe empirical results to be reported in this section have been
computed with 2SLS techni~ue as mentioned earlier.
These results are
useful on many counts:
the statistical results obtained for equation
(4.1) will provide empirical evidence to the hypothesis that in deve-
loping countries, the primary motive for holding money balances is for
transactions and precautionary purposes.
Furthermore, its statistical
,
significance will provide additional support to the conclusion already
30
derived from other studies
that the money base (M ) is a signifi-
1
cant variable for income determination except in countries with
hyperinflation.
Equally important is the estimate of the income-velocity
30 For reference, see Demand for Money in Developing Countries,
Some Theoretical and Empirical Results, Chorn, G-Huey \\.long: .Journal of
~tary Economlcs-1(1977): North-Holland Publishing Company.
84
of money base.
This gives uS an indication as to the degree of bank
expansion in a particular developing economy, that is an indication of
the extent to which the banking system is being spread to the rural
areas of the country and of rural involvement in transactions.
To begin the analysis of our empirical results for equation (4.1)
we report below, the estim~tes of the regression of that equation,
2
including the coefficient of determination (R ) and the t-statistics.
Yt = 3.11
MS
( 4• 1)
( .0844 )
the t-statistics
36.810
2
R = .9595
S. E. = 115.36
O.W.
.7832
The empirical findings of equation (4.1) do support most of the
hypotheses stated above.
Indeed the very high coefficient of deter-
mination R2 = .9595 and the t-statistics = 36.816, do respectively
indicate a strong explanatory power of the independent variable over
the dependent one and secondly the explanatory variable is highly
significant.
Therefore, on the basis of these statistical results the
following conclusions can be drawn with respect to the economy under
study.
First, the highly significant power of the explanatory
variable does confirm the argument that in a growing dualistic economy
where the agricultural sector is the backbone of the economy, aggre-
gate transactions are determined by the nation's money base (M ).
In
8
other words, since the primary motive for money holdings is for tran-
'85
sactions purposes. transactions will depend on the aggregate stock of
.
money.
This shows the constant relationship established between the
two variables in an economy at its early development stages.
Second.
the high velocity of circulation has some significant implications:
it indicates how efficiently monetary impulses act upon th~ real sec-
tor of the economy by facilitating the rate at which real assets are
transacted in the different sectors of the economy.
Therefore there
is a strong indication that the economy's degree of monetization
increased and that financial intermediations are improving.
Before consolidating all the statistical results in the framework
of domestic credit approach to the balance of payments. we ought to
report and analyze the empirical findings of the domestic import
equation (4.3).
IMP = 30.83
+
• 26Yt
+
.406(F_ 1-F_ 2)
(4.3)
(6.789)
(.00859)
(.294)
the t-statistics = 30.323;
= 1.383
S.E. = 15.550
2
R
= .9934
D.W. = 1.330
Many interesting conclusions can be derived from the empirical
results of the domestic import equation as reported above.
The high
2
value of the coefficient of determination (R
= .9934) indicates that
both independent variables have a strong explanatory power.
For indi-
vidual variables, we can see that the valup of the t-statistics for
86
Gross Domestic Product is highly significant whereas the t-statistics
for the lagged net foreign assets indicates a relative significance
level at 10 per cent.
The relatively higher explanatory power of the
gross domestic product over that of net foreign assets does support
the proposition that as an economy grows, the accompanying higher
domestic income level generates higher demand for foreign goods in the
private and public sectors.
This process disequilibrates balance of
payments position.
The balance of payments position is further
frustrated as inflows of foreign assets contribute to increase the
import level in the public sector.
However, when interpreting the
estimates of the regression coefficients, the proposition described
above, still applie? but in a reverse order.
Indeed, the marginal
propensity to import (~l = .26) is less than the impact coefficient of
changes in lagged net foreign assets (~2 = .406).
In this case, a one
unit change in the domestic currency will result in .26 units being
spent on imports whereas if net foreign assets are projected to
increase by one unit, .406 units will be spent on imports.
In any
case, whichever interpretation is adopted, the empirical results sup-
port the contention made earlier.
In other words, whatever factor of
production (i.e. domestic or foreign input) is being considered in a
growing economy, the increase in either one will bring about higher
demand for foreign goods, which in the final analysis has negative
impact on the balance of payments.
From the empirical estimations of equations (4.1) and (4.3) we can
identify the estimates of the reqression coptfi(ipnt~ n~(essary to
___ ~_
~
__ ."
_
_ ,-...)t_
:Ir~'
•
j.
i
::m~. -C"
i
R7
i
'I ~
i,!
f
TABLE 5-1
!
i
11il!
The Ivory-Coast:
Regression ~ata:
19~n-1980 [billl}ns of CFA Francs in current pricesl
-t
--.-----:~-~-
~(=:::~----,-
.f
,
- - - - ' - - - - - - · · - - - - · · - - 1
i
lIi
~ !
Endogenous Variables
! Predetermined Variables
I
~11J:
Years
I
GDP( Yt )
MB_
'_--r:-_~~~=- ,_~. \\---y
~-. DC
F'~_l~F-2_._._",,~..(_~)
,
1963
197.80
36.44
9.GJ
66.80
~2.20
27.44
6.19
9.30
~
~::,c
1964
239.70 .
50.39
12.53
78.80
~]1805.0! 37.26
9.00
6.19
8.er
1965
239.60 7
50.94
15.QS
78.50
Sf.70
35.29
-12.53
9.00
14.90
1966
258.'00
56.98
1~.06
R5.90
90.40
37.92
15.65
12.53
16.50
ili..
1967
275.70
62.14
14.4593.30
96.50
47.69
19.00
15.65
9.30
1968
326.50
77.58
22.\\~O I
106.90
125.~2G.
55.12
14.45
19.06
10.00
1969 1
365.60
98.36
28.:10
124.20
143.9\\0 I I
70.26
22.40
14.45
10.30
I
1970
415.36
115.05
38j~:
156.20
161.00 \\\\1
76.45
28.10
22.40
25.60
1971 I
440.10
129.80
34.7~
167.80
164.90
I 95.10
38.10
28.10
25.90
1972 1
472.50
136.59
13..sq
179.30
181.10 \\
122.69
34.70 I
38.60
11.30
1973 I
565.30
160.26
10.lq
171.60
191.90
150.16
13.90 I
34.70
"45.20
:1-~
1974
738.76
227.35
,21.711
241.40
301.40
205.65
10.10 I
13.90
50.30
~
1975
834.50
269.50
: )';.3,0
249.60
266.30
'~67.2
21.70
10.10
50.30
1976
1114.00
367.80
-\\2.~0
311.608
392.50
~Or4
2.30
21.70
56.91
1977
1582.50
565.30
.1,5\\80
429.566
529.212
5}.~~?
- 2.60
2.30
88.62
1978
1740.60
633.604!~.~0
522.502
524.382
58o\\S.O
35.80
- 2.60
191.38
1979
1944.7
640.60
ft~!7.20
582.850
532.850
677.\\~C I
44.80
35.80
231.95
1980
2223.50
626.3
(.. 20./
636.96
665.52
835.{)~, {"" -37.20
44.80
311.452
ltr.:
J '
I
~'ltI.~
I
+
,~_._._
I"~ __.....
I
't" \\\\
....---,------.-
SOUrCE!s:
IM~/Intern~tional Fii,n;'q Sta~istics (yariou~:, Ye~rs); Month~y Bll~ftin of S~atistics
(~~'~~~nNat lOns ); Lr""i\\~'!vOlre en Ch, f f res (Mi n1Stere de 1Ecor\\\\l desF, nances et
~,
(a)
The capital account sta~:, ))'," , nc 1ude nonmonetary "ap' tal and errors \\~m, ss, ons.
88
determine empirically the domestic credit ~ultiplier of the ba1anc€
of
payments:
v = 3.11
=
11
0.260
"'1
= 0.406
~2
\\
Then the domestic credit multiplier (aF/aOC) defined in equation
(4.10) is:
aF
~1 v
.8086
aoc - -
= - O.447n
[I' '0'
1
I' .-"
+ ill v
1.8086
!
I
J
Therefore, the offset-coefficient for the economy being studied is
-0.4770.
The rationale for determining the domestic credit multiplier
is evidenced in its usefulness to generate balance of payments
equilibrium whenever defined with. respect to foreign assets.
This
value has many implications:
First of all, the model yields the
expected sign and size for~.
In other words, finding -.4470 ~ ~ ~ 0,
con fi rms our earl i er hypothes is in Chapter I I wh.icb__ ~jsumed. i':'lpe!,~~ct
capital mobility.
However, we cannot draw a definite conclusion as to
the size of a in the long-run where it is supposed to be equal to
minus one.
This is so, because in the country under study if we can
ascertain that the local authorities are establishing financial
markets and other related institutions to obtain control of the
country's monetary fllnctioninr:J and hp'f1':'-o. ::"'~':~-;ce:'ln "independent"
89
monetary policy, it remains certain that capital movement between
developed and developing nations is highly influenced by institutional
rather than real economic variables.
This latter determinant of capi-
tal movement, naturally, compromises the attainment of perfect capital
mobility within a reasonable time period.
The other inllJlication of
the size of 0 is that it indicates that the inflows of foreign capital
has considerably contributed to finance the fluctuations on the
current account balance.
In addition, it provides a strong indication
that the behavior of the overall balance of payments is dominated by
the trends exhibited over time, by foreign capital.
Therefore aF/anC
C3f.
be interpreted as the effects of one percentage change in domestic
credit on both the current account and capital account.
Here, the
empirical value of the offset-coefficient indicates that as domestic
credit increases by one million of CFA Francs, reserves will be
lowered by ,'CFA Francs 447,nOO.
The second impl ication of the domestic
~. "---. ------.~---.~---._--~_._---_._._----_._------
credit_multipJ.ler._wjJl_J:>J~J.elatedto the sterilization policy.
That
... ,
--------_._----_._-_._~---
.._~.- ,-
is, if we ~~~_'!!~_.a_!l__? ut~no~ous_i nflow--9_Lf9.l'~igD-capi taJ_!tLl
m!Ll_j()~_~, with nodomestic_sJed__Ltzation policy. this inf~ll
ge~~_~_~!~ ~~f~~ttiD_9__Ol,l1nQwof~X mi 11 ions of CFA Francs (0 bei ng the
3
offset-coefficient i.e. aF/aDC).
And the net effect on the balance of
payments is thus (l-~)X with complete sterilization, the net effect
will strictly be limited to X.
The above general proposition can be illustrated in a diagram
which represents the combination of interest rate (r) and income level
(y) for which the domestic o<;t.put market (is), !lIoney :llarket (LM) and
the balance of payments (SP) are r,~spectively in e~u-Il ibrium.
The
B9A
r
BP
y
FIGURE Ill:
Sterilization Policy and Domestic
Credit Multiplier
full equilibrium is represented by the intersection of the 3 curves at
point (E).
Earlier assumptions about the slopes of the IS-LM-BP cur-
ves still hold.
The autonomous inflow of forei~n capital which is
offset by OJX amount of the domestic currency of the banking system
brings the system from equilibrium point (f) to the balance of
payments disequilibrium point (n). by shifting the LM curve outward.
With no domestic sterilization policy. the LM curve moves back to its
original position (E) as a result of the lower foreign component of
the money base after the whole adjustment mechanism has heen
completed.
However, if the ·l~()netary authorities decides to sterilize
8gB
the balance of payments
effect~ the system would stay put at point
(0).
In other words. the monetary authorities must decrease
(increase) the domestic component of the money base by the same amount
as the foreign component increases (decreases). by an amount equal to
the balance of payments surplus (deficit).
In order to determine the change in domestic credit which is
necessary to result in balance of payments equilibrium while at the
same time considering changes in net foreign assets. in exports and in
net capital inflows. we ought to evaluate empirically equation (4.8).
tiW
-
?i!D~ii'il r~",~•
•
~
' • • • •
,
- , • • '
_ . • • • •
'.-.~' •• -
_-::::-:~. >~
'i ".
,j
90
That is:
1-~2
~2
1
1
dOC = - - dF
+ -
dF
+ -
dX + -
dK
(4.8)
~1v
-1
~1 v
- 2
~1 v
~1 v
or
dDC = .594 d F
.406 dF
1
1
.8086
-1 + .8086
-2 + .8086 dX + .8086 dK
Hence:
-- ------------_._----_.---------
r---------
+ .502 dF_
+ 1.236 dX + 1.236 dK
I
dDC = .734 dF_ 1
2
Il----------------
In equation (4.8) we have been able to determine the values of the
parametric constants associated with the predetermined variables which
give the change in domestic credit necessary to bring about the
balance of payments equilibrium.
The second set of values needed by
the policymaker to conduct an effective stabilization policy via
balance of payments equilibrium using domestic credit is the numerical
values of dF_ , dF_ , dX and dK.
With the knowledge of the numerical
1
2
values of the coefficients and predetermined variables at the
beginning of every period t, the policymaker can solve for dDC by
substituting values of dF_ , dF_
1
2, dX and dK into equation (4.8).
Then the following monetary policy implications can be deduced:
If
the actual increase in domestic credit is equal to dOC, then the
balance of payments will be in equilibrium: hence bringing about
simultaneous equilibrium in the home-country in tprms of interr,al
stability and external stability is equally achieved.
However jf the
91
increase in credit is larger than dOC, the end-result will be a defi-
cit in the balance of payments, while an increase in credit less than
dOC will generate a surplus.
A conceptualization of these outcomes
will lead us to record balance of payments deficits and sur~luses for
the Ivory-Coast's economy over the period of 1963 to 1980 and then the
changes in domestic credit associated with those different balance of
payments outcomes.
Table 5-2 below, provides the relevant data on
balance of payments surpluses and deficits.
In addition, we shall
report in that table the changes in domestic credit that would have
been necesary for the balance of payments to have been in equilibrium.
In other words, using equation (4.8) we shall derive the comparative
dynamic results of the model with regards to the predictive values of
dOC.
92
TABLE 5-2
Required vs. Actual Changes in Domestic Credit for
Balance of Payments Equilibrium
BOP Surplus
Actua 1
Required
Deviations
or deficit(-)
flOC (a)
flDC(b)
(c)
,
Years
(1)
(2 )
(3)
(4)
J
--
1965
4.50
-2.57
3.136
5.706
1966
3.10
2.63
16.792
14.162
1967
-6.30
9.77
2.709
-7.01)1
1968
8.70
7.49
34.066
27.176
1969
9.50
15.08
27.005
11. 925
1970
9.60
6.19
48.220
42.030
1971
-3.20
18.65
15.392
-3.258
!
1972
-18.90
27.59
4.502
-23.088
1973
-2.30
27.47
38.275
10.805
1974
21. 00
55.49
128.414
72.924
1975
-35.40
61. 55
-36.776
-98.326
1976 I
7.80
I 103.20
155.736
52.536
1977
44.70
159.10
194.831
35.731
1978
36.57
59.30
146.767
87.467
1979
-52.24
89.00
86.496
-2.504
1980
-31.37
157.20
201).572
49.372
ti
-
--
Source:
Tables 5-1 and 5-2 and Calculations.
(a)The actual flOC'S are obtained as the difference between DC in
two different periods, say 1962 and 1963.
(b)We obtained Req~ired ~DC's by substituting values of changes in
Net Foreign Assets, in Net Inflows of Capital and in Exports into
Equation (2.8).
(c)The deviations are just the difference between Actual and
Required DC's.
93
In column (3) of Table 5-2, ~e have the statist~cs which represent
the required change in domestic credit creation for the balance of
payments to be in equilibrium.
In years where the balance of payments
exhibited surpluses, the required changes in domestic credit as pre-
dicted by the model are higher than the actual changes in domestic
credit.
For instance, in 1966 and 1970 and 1976, the balance of
payments registered surpluses of 3.1, 9.6 and 7.8 billion of eFA
Francs respectively and domestic credit rose by 2.63, 6.19 and 103.20
for each corresponding year.
However, the model predicts that
--------
.----
.__._._---
increases of 16.792, 48.22 and 155.736 billion of eFA Francs respec-
tively would have brought the balance of payments into equilibrium.
But sincp- the actual increases were 14.162, 42.03 and 52.536 respec-
tively short of the predicted increases in domestic credit, surpiuses
were recorded.
And if the actual increases in domestic credit were
greater than the predicted ones, then the balance of payments would
have been in deficits.
Indeed for the years 1967, 1971-72, 1975 and
1979, we observed that actual domestic credits were higher than they
should have been.
Hence the balance of payments registered deficits.
Yet, the model did not yield consistent results throughout the
observation period.
Some peculiar results are obtained in 1973 and
'1980.
In these years, the balance of payments recorded deficits of
,.----
2.3 and 31.37 respectively and actual domestic credit was increased by
27.47 and 157.20 for 1973 and 1980 respectively.
According to the
model'l s dynamics described so far, for the balance of payments t.o be in
equilibrium, the predicted values of domestic credit sholild hCi'vc: ',p.en
94 .
been less than the actual figures for these two particular years.
However, the model predictions are respectively 38.275 and 206.572.
A
heuristic explanation to these abnormal results may be found in
looking at the events which took place a year earlier, i.e •., in 1972
and 1979 respectively.
With respect to actual and required changes in
domestic credit in 1972 and 1979, the model exhibited consistent
results.
However, when looking at the statistics related to net
foreign assets in Table 4-1, some interesting facts appear then.
We
can see that the inflows of net foreign assets have been drastically
reduced.
From 13.90 to 10.10 hetween 1972-73 and from -37.2 to -208.7
between 1979-80.
These drastic reductions in the inflows of foreign assets can be
related to some specific events which occurred at the end of the
decade of 1960's and 1970's respectively.
At the end of the first
decade of the country's independence, government policy toward foreign
investors was considerably alte~ed.
Indeed, as we alluded to earlier,
a decision was made to remove most of the incentives granted to
foreign investors.
As a consequence, many foreign investors who no
longer found the investment opportunities as rewarding as they would
desire, either curtailed their production activities or undertook a
policy of invisible income repatriation in the form of higher prices
for their products and thereby created lower local demand for their
goods and services.
In addition new foreign investments were not
forthcoming as a result of the new policy of invpstment-corle.
Domestic policy regarding f~reiqn investments is slill lpss
95
favorable than it was in the 1960's.
Tile result is ther'efore that
foreign investments are still coming into the country but not at the
rate witnessed in the 1960's.
The policy decisions at the end of the
1960's continue to have some ramifications over the country's economic
pattern in the 1~80's.
However, other events have taken place which
have aggravated the overall economy in the late 1970's and early
1980's to the extent that the country was declared to be in the brink
of economic crisis.
As a result, a "Stand-by" loan from the IMF has
been deemed necessary to sustain the economy.
The huge decrease in
the net foreign reserves between 1979 and 1980 can be explicitly
explained by the policy-actions taken by the local authorities in
regard to the sale of the main agricultural products in the world
markets.
During the period of 1979-80, the wor'd markets for agri-
cultural products (coffee, cocoa) were drastically distabilized.
This
was mainly because the economies of the main buyers were experiencing
economic recession which reduced their purchasing of primary com-
modities for which many substitutes exist in their markets.
The lower
purchasing power of the developed countries as a result of high infla-
tion rate, has been manifested in their offering low purchase price
for the products of the developing nations.
For most agricultural
products, the prices offered were far out of line with preceding fore-
casts.
Since the Ivory-Coast is the w0rld's third lnrgest supplier of
coffee and cocoa, the local authorities decided not to supply the
world markets immediAtely with those years' produc~ions.
96
The ultimate objective of this policy ~as to hold the rl0~~stic
production from the world market in the expectation that price will
immediately go upward at the announcement of the decision.
Yet, pri-
ces never went up, and the products could not be sold at the going
pr ice at 1ate r date s, for 0 the r pro duce r sal rea dy sup P1i ed ·t hew 0 r 1d
markets.
Consequently the country lost the revenues expected from the
sale of coffee and cocoa and high deficits were recorded in the
balance of payments.
The overall economy was then taking a severe
downturn.
To compensate for the loss of foreign reserves during this
period, many economic policy alternatives have been implemented.
Domestic cos~s of living are skyrocketing.
The reslllting domestic
inflation, is a consequence of the massive changes in domestic credit
as predicted by QUi model though the b=lance of payments was in defi-
cits in those years.
On the whole, the model has yielded some consistent results as
expected from the domestic credit approach to the balance of payments.
On the other hand, some irregularities have been brought out by the
model's predicted results.
The rationale for these latter results
have been related to some specific conditions prevail'ing in the
domestic economy during those particular years of the observation
peri od.
The overall analytical framework has been based upon a macrodyna-
mic model whose empirical results are consistent and significant in
many resp~cts.
Hence all along the esti~ation process, it has been
impl icitiy assllrneo that the system 1S stilhle.
With
the empirical
reslllt.s proviriec1 hy the regression .of equations (4.1) and (4.3) we
97
will undertake in the next section an explicit investigation of the
stability condition of the model.
C.
Dynamic Analysis! Stability Conditions
In Appendix A, we presented the general conditions for stabi-
lity of a second-order difference equation, which is the equation of
the balance of payments after solving for its reduced form.
In this
section, our main task will consist of finding out whether or not the
whole system is stable.
This investigation will be conducted by
looking at the path that the balance of payments will trace out over
time.
Following Kmenta [1971, p. 593], a system is said to be stable
if in a situation where the values of the exogenous variab1es are held
constant through time, the mean values of the endogenous variables
settle down to some constant levels.
And an unstable system will
obtain whenever the process described above, follows a reverse pat-
tern.
That is, when the values of the endogenous variables either
explode or display a regular oscillatory movements.
One way of deter-
mining whether a system is stable or not is to refer ·to the final form
equation (or to examine the fundamental dynamic equation).
In our
case, our stability study will center around the trend displayed over
time by the fundamental dynamic equation of the balance of payments.
Considering only the homogenous part of the equation of the
balance of payments, equation (4.7) can be written as a homogenous
second-order difference equation with co~stant coefficients i~ the
variable F.
That is:
98
(4.7 1 )
or in a characteristic equation form, we obtain:
(4.7")
Following the stability conditions as specified in Appendix A, we can
compute the characteristic roots °1 and °2:
= -0.328
= - 0.224
We could also state the conditions above, in terms of the largest
roots.
That is:
p = Max(lo 1,
1
1°21) < 1, or p = .328 < 1, which
satisfies the necessary and sufficient condition for stability.
In terms of the n-determinants defined as ~1' ~2' ••• ' we found the
fall owi ng:
1
.892 ) 0
-.328
~:~tmry':r~t9~;~~~1~r9!;9'~.'.':
";'.
1
o
-.224
~2 =
o
1
-.224
= • 9498 > 0
-.224
o
1
Therefore, ~1 and ~ being respectively positive, they satisfy the
Schur1s theorem of the stability conditions of the characteristic
equation.
In other words, the characteristic equation is guaranteed
. /
to have roots less than unity insofar as 6 and ~2 are both positive.
1
Finally combining the characteristic roots and the Schur's conditions,
the overall stability of the system can be expressed in a triangular
form as did Routh and Hurwitz.
Then upon substitution of values of
° and °
1
2 into the Routh-Hurwitz stability conditions, we found:
1 - .328 - .224 = .448 > 0
(a)
1 - .224 = '-776 > 0
(b)
1 + .328 - .224 = 1.104 '> 0
(c)
Hence the Routh-Hurwitz conditions are all positive.
In addition to
the previous stability conditions which are all satisfied, we conclude
that the system is stable since the principal fundamental dynamic
equation is stable in all aspects of the conditions so far studied.
Therefore, over the short-run, our dynamic model is more likely to be
accurately predictive of the structure of the economy under study via
lOO •.
its balance of payments.
Henceforth, the long-run expected results
will be obtained.
In effect, the prediction that over the long-run,
the balance of payments surplus or deficit is temporary and self-
correcting under the monetary mechanism is verified.
CHAPTER VI
SUMMARY, POLICY RECOMMENDATIONS ANn CONCLUDING REMARKS
In this dissertation we sought to analyze the international econo-
mic structure of a small open developing country:
the Ivory-Coast, in
~
order to prescribe an efficient stabilization policy.
Such stabiliza-
tion policy was carried out in the context of a macrodynamic model
which analyzes the relationship between domestic credit and the
balance of payments of the country under study.
The model analytical framework which is basically an adoption of
the traditional monetary approach to the balance of payments has been
innovative in the sense that it favors domestic credit over money, as
the policy controlled variable.
More fundamentally from the theoreti-
ca1 framework to the empirical results, the model's objective was to
stabilize the economy as a whole internally by reducing the domestic
inflation rate and externally by bringing equilibrium in the balance
of payments.
The domestic component of the money base has been used
to achieve these two objectives and also to generate the dynamics of
the model.
The internal objective of lower inflation is achieved by
determining the appropriate change in domestic credit (60C).
Since
the latt~r is a domestic macroeconomic variable, it is expected that
the impacts of changes in domestic credit will be transmitted to the
domestic monetary sector on micro and macro-levels by restricting the
growth of liquidity.
In the process of developing our model, we used the principle of
income· determination.
This domestic income level has been specified
with respect to -the constant velocity approach of the quantity theory
of money.
This specification was adopted-particularly to depict the
special characteristics of the country under study.
This feature was
expressed in equation (4.1) and the related empirical results are
conclusive on many counts.
Indeed, the estimation results showed a
strong explanatory power of the independent variable over the depen-
dent one as expressed by the very high coefficient of determination
(R2) and a highly significant t-statistics.
These two statistics lend
support to the contention that the money base as we defined it in the
study, explains pretty well the national income or the natiQn's demand
for money and thereby establishes a constant relationship between the
two variables.
Similarly the empirical results of the domestic import equation
showed that independent variables have a strong explanatory power over
the domestic import demand.
This explanatory power is also evidenced
\\
by their very high
2
R •
Afte~som~-substitutions andmanipulations,w~;obtained the
reduced form-equation:ofthe balance of payments (4.6).
The total
differenti~tion of that equation whic~ is given in equation (4.7),
served as the beginning of our theo~etical set-up of the credit
approach to the balance of payments.
From equation (4.7) we deter-
mined the change in domestic credit as stipulated in the model's ana-
lytical framework.
Also, after deriving theoretically the domestic
credit multiplier from equation (4.6). we combined the empirical
results of equations (4.1) and (4.3) to obtain empirically the ~alue
103
of the domestic credit in equation (4.]0).
This latter value enabled
us to test the validity of the domestic credit approach to the balance
of payments and also to show the influence of the inflow of foreign
capital on the current account.
All the empirical results obtained above, sustained in ~ll aspects
the dynamics of our macroeconomic model, though some inconsistencies
appeared in the results for certain years.
Indeed, for 1973 and 1980,
the model predictions are inconsistent with its basic structure.
However, we found that these inconsistencies were the results of some
specific occurrences in the Ivorian economy during these particular
years.
Therefore, the policy implications of the model derived for
1973 and 1980 do not parallel as such, those of the other years of the
study period, but they are in conjunction with the economic events
that took piace in 1973 and 1980 respectively.
Consequently policyma-
kers interested in using such a macrodynamic model, ought to keep in
mind what to expect when some unexpected variables interfer in the
economy and deviate its functioning from its normal trend.
Despite these inconsistencies, the general monetary policy impli-
cations of the model have been in line with its theoretical foun-
dations.
It has yielded accurate predictions of the required level of
domestic credit to eliminate the balance of payments deficits, reduce
its surpluses and hence bring it to equilibrium.
Furthermore, in order to determine the equilibrium credit,
starting either from the ceiling level or from the floor level, poli-
cyrnakers are required to estimate in .'j(i'Jancp dOIGestic e,:ports and net
inflows of foreign capital.
The uncertainty sur'rounding the estimated
104
:-: ~
... ,"
values of these variables may impose some limitation on the model IS
predictive capacity and hence its effectiveness as such.
However,
these shortcomings can be overcome by assuming that the main export
products are sold in the forward markets.
And thus, the authorities
have in advance a fairly accurate amounts to be delivered t~ foreign
markets.
Secondly, in the Ivory-Coast like in many developing
nations, the inflows of foreign capital are determined in advance by
donor countries according to their own criteria.
Such is the case of
tied aids.
By the time the development project comes into effect,
local policymakers can determine with some certainty the range within
which the actual values of foreign capital to be received will fail,
gauging f~om the total aMount promised by the donors.
With the proba-
bility that there will be fairly accurate estimates of exports and net
inflows of foreign capital, the model developed in our study does pro-
vide a solid foundation for the implementation of domestic credit
policy as a viable stabilization program.
In debating regional spe-
cialization on a national basis, it should no longer be controversial
to start regional integration of many similar economies such as those
of the Western African nations, from one country's mac~oeconomic
achievements.
In other words, if a policy of domestic credit can best
serve the economic objective of the Ivory-Coast as demonstrated in the
study, there are then grounds to consider it on regional level.
And
the adoption of such a stabilization policy tool should come as a
complement to the already existing tools of stabilizing the different
econOlnies either indi\\/idually o~- cCJliec.Liv~l'y.
Finally, the validity Of the model's predictive capacity is
105
further supported by the stability study conducted.
Indeed, the
model's statistical results did not violate any of the stability con-
ditions that we determined in Appendix A.
Therefore, the overall
system was found to be stable.
The credit approach to the balance
of payments is more than an efficient method of generating balance of
of payments equilibrium in the short-run.
More importantly the con-
tention that over the long-run under the monetary approach, the
balance of payments becomes self-equilibrating, is verified.
The last remark to be made concerning the overall project consists
of the type of commodity considered all along the analysis.
The
model's analytical framework was based on the assumption of one
integrated market for traded and nontraded goods, so that the law of
one price holds.
All policy implications regarding the non-traded
goods have been implicitly incorporated into the general policy impli-
cations of the model as a whole.
Lumping together the markets for
traded and non-traded commodities has one particular adverse effect.
It hides the divergent inflation rates that certain countries
experience in spite of their fixed exchange rate system.
In effect,
it has been shown by Dornbusch (1973), Krueger (1974) and Parkin
(1974) that whenever non-traded goods are explicitly incorporated into
a model's analytical framework, the policy-implications are likely to
be more comprehensive than those obtained from a model whose theoreti-
cal setting is based on traded goods only as in our study.
For
instance such market dichotomy explains pretty well, why different
countries engaged in international transactions and opErating in a
system of fixed exchange rates, have di fferent inflation rates.
It
l05A
will be interesting in the future to incorporate non-traded goods
whenever a study along the same lines as ours, is undertaken.
In this
way the policy options made available under this project could be
strengthened.
·n.;
APpnmI x A
STABILITY CONDITIONS OF EOUATION (4.7)
The dynamic structure of the model leads to the determination of
the stability conditions of the equation of the balance of payments or
equation (4.7).
(1- 112 F-1 + 112 F_2- 111 VDC+ X+K - l1a - ~
(4.7)
F
1+ 111 v
Equat.ion (4.7) can be rewritten 'lS;;1. nClnhomngf>nolls second-order dif-
ference equation with constant coefficients in the variable F:
.
" " r , v , v .
,-
- ill v U'-'TI\\T",-pa- Lot
( 4.'7 I )
1+ U, V
1
or
k
We can express equation (4.7) in a characteristic equation form:
(4.7")
k.
where:
(1-)J2 )
° = -
1
1+)Jl v
)J2
° = -
2
1+)Jl v
- )Jl vnC+X+K- )JO-Lt
k
=
constant 1= f) .
1+)Jl v
Note:
k is the nonhomogenous part of equation (4.7').
Since the characteristic roots are ° and °
1
2 and on
1, the charac-
teristic equation can be expressed as:
(4.7"')
For the case of distinct and real roots and thus for stability both
1011 < 1 and \\° \\ < 1.
This rule is sometimes stated in terms of the
2
largest roots in the following manner:
Let p =: max(!ol\\' \\(21): then
a necessary and sufficient condition for stability is p < 1.
The above results can be obtained by applying the Schur's theorem
to the auxiliary equation (4.7").
It will mainly consists of exa-
mining the relationships among the coefficients on' 01 and ° .
2
Basically, the Schur's conditions tell us that equation (4.7"')
will have roots ]ess than unity if and only if the following n-
determinants are all positive:
--~~:~::?~,;;~~~T~];)'~
108
.":~
o
o
6n -1
o
o
.5n
o
o
°n- l' °n- Z
°0
> 0
D,
:::
-----------------------------------------------~-----
n
on
0
0
00
01
°n_1
°
o
o
6n-Z
n-1
o
o
where
> 0
0
°0
on
0
0
°0
°n -1
°n
:::
--------------------
> 0
D. Z
On
0
0
°0
"
On -1
on
0
00
Equation (4.7") can be transformed into a second-degree polynomial
equation that reads:
R~ ... ..••... .... ...""'.. ..~,,~-r-C=-·C=~----~'"--·~~-"~"r;:s~:c~·~J~~~'ll
.', ·:;rj(~
2
(- 6 E - 6 E + 6
2
1
0) Ft::; k.
where:
E is the transformation operator
Then combining the Schur's conditions, Routh and Hurwitz defined the
stabi-l ity regi on in a (6 ,6 ) pl ane as a tri angl e expressed by:
1
2
1 + 6
+ 6
> 0
1
2
(1)
1 - 6
> 0
2
1 -
6
+ 6
> 0
1
2
An empirical verification of the Routh-Hurwitz stability conditions
can be conducted by substituting the coefficient-values of ~1' ~2 and
v into the expressions for 6 and °
1
2 and then check whether or not the
above three conditions hold.
(l)The Routh-Hurwitz conditions as derived from the Schur's
theorem can be found in:
Foundations of Economic Analysis: pp.
429-1139.
p.
A. Samuelson, Cambndge: l1arvard Universlty Press, 1947.
-------,...."......,--.,.:-"_,.,...-=..
-_.._- - _-__ -
.
~.
-~
~.,
' 1
. '~:'
APPENDIX B
Computation of the Changes in the
Predetermined variables of Equations (4.8)
dDC = .734 dF_
+ .502 dF_
+ 1.236 dX + 1.236 dK
(4.8)
1
2
,
dX
dK
dF
dF
Years
-1
I
-2
!
-
I
1965
- 6.80
6.10
3.53
I
2.81
i
1966
I
8.70
1.60
3.12
I
3.53
1967
I
6.10
II
- 7.20
I
3.41
,
3. ] 2
I
;
i
1968
I
28.70
I
.70
I
!
I - 4.61
i
3.41
I
I
1969
18.70
.30
I
I
7.95
- 4.11]
I
1970
17.10
15.30
5.70
7.95
·1
I
1971
3.90
.30
10.00
I
5.70
I
1972
16.20
-14.60
- 3.40
I
10.00
I
I
I
1973
10.80
33.90
-20.80
i - 3.40
I
i
I
1974
109.50
5.10
- 3.80
I
-20.80
I
1975
I
-35.10
0.00
11.qO
I
I
3.80
I
-
I
i
I
.
II
I
I
I
1976
,.
126.20
6.61
I
I
-19.40
11. 60
I
i
i
I
I
I
I
I
1977
136.71
31. 71
- 4.90
I
-19.40
I
I
1978
I
- 4.83
102.76
I
38.40
I
4.90
I
-
I
j
i
,
I
i
1979
8.47
40.57
9.00
I
38.40
I
.
i
t
I
,
I
1980
132.67
79.50
I -82.00
I
9.00
f
I
I
i
I
I
I
!
I
_._J
..-
. ",...
:"' .. ,
APPENDIX C
Reduced Form Equation of Income Determination
(4.1)
Upon substitution of identities (4.2') and (4.5') into equation (4.1)
we obtain:
Yt = v[DC + F_ 1 + X - M + K] + ~ •
Replacing Mwith its functional form equation (3.3), Y becomes:
t
or
or
112
v.
or
Note:
v
6.,.)
By totally differentiating the reduced form equation of Yt , we obtain
the relationship between domestic income level (Y ) and the predeter-
t
mined variables.
That is:
Setting dY
equal to zero, assures that the domestic income level is
t
in equilibrium and we could then solve for dOe which is consistent
with such equilibrium.
~----_.~~
.,".' :~.~~:~~r~~~.'··:·-;
BIBLIOGRAPHY AND REFERENCES
Adams, Klein:
Stabilizing World Commodity Markets:
Lexington Books:
D.C. Heath &Company, Lexington, Massachusetts (1978).
Ando-Herring, Martson:
International Aspects of Stabilization
Policies:
pp. 147, 393 and 421:
Proceedings of a Conference held
at Williamstown, Massachusetts, June 1974.
Aghvli, Bijan B.:
Money, Prices and the Balance of Payments:
Indonesia 1968-73: Journal ~ Development Studies:
Vol. 13 No. 2,
Janua ry 1977.
Brown, c.P.:
Pr·imary Commodity Control:
Oxford University Press,
(1975).
Raldwin &Al.:
Trade, Growth and the Balance of Payments, 1965:
Rand
McNaiiy! Co.:
Chicago, North Holland Publishing Company: Amsterdam.
BolnicK, Bruce R.:
Interpreting Polak; Monetary Analysis in
"Dependent Economies": Journal of Development Studies, Vol. 11,
July 1975, No. 4.
Brunner, Karl and Meltzer, Allan H.:
Predicting Velocity:
Implications for Theory and Policy:
The Journal of Finance:
Vol. XVIII
No. 2, 1963.
Coppock, Joseph D.:
International Economic Instability:
The
Experience After World War 11:
McGraw-Hill Book Company, Inc.,
(1962) pp. 1-139.
Charles Schotta, Jr.:
The Money Supply, Exports and Income in an Open
Economy: Mexico, 1939-63:
Economic Development & Cultural Change:
114
Chorng-Huey Wong:
Demand for Money in Developing Countries:
Some
Theoretical and Empirical Results:
Journal of Monetary Economics
3 (1977) North-Holland Publishing Company.
Chenery. Hollis R. and Allan M. Strout:
Foreign Assistance and
Economic Development:
The American Economic Revi~w Vol. LVI.
Sept. 1966. No. 4, Pa rt l.
de Silva. K.E.A.:
Money Supply. Inflation and the Balance of Payments
in Sri-Lanks (1959-74):
Journal of Development Studies: Vol. 13.
No. 2, January, 1977.
Dornbusch, Rudiger:
Open Economy Macroeconomics:
Ch. 10:
Basic
Books, Inc. Publishers:
New York, 1980.
Exchange Rates and Fiscal Policy in a Popular
Model of International Trade:
The AmeriCan Economic Review:
December, 1975.
Floy. John E.:
Government Expenditure Policies in the Small Open
Economy: Canadi an Journa 1 0 f Economi cs:
1979.
Gurley. John G. and Shaw, Edward S.:
Money in a Theory of Finance:
The Brookings Institution:
Washington, D.C.
Helliwell. John F.:
Monetary and Fiscal Policies for an Open Economy:
Oxford Economic Papers. 1969.
Halliwood. Paul:
Stabilization of International Commodity Markets:
Jai Press. Inc., Greenwich. Connecticut (1979). Vol. 18.
Horwich, George and Samuelson, P.A.:
Trade, Stability and
Macroeconomics:
Academic Press, Inc. 1974, New York.
\\
Havrilesky, Thomas M.:
Monetary Macroeconomics 1978:
AHM Publishing
Corporat ion, Arl i ngton He; ghts, 11 -I i noi s.
Herring, Martson:
National Monetary Policies and International
Financial Markets: 1977 North Holland Publishing Company:
Amsterdam.
Heck, Harold J.:
International Trade:
A Management Guide:
American
Management Association, Inc., 1972.
Harrod, Ray:
International Trade Theory in a Developing World:
pp.
297-331 and Chpts. 13 and 14:
MacMillan ~ ~ Ltd, New York:
St. Martin's Press (1963).
Hahn, Frank H.:
The Monetary Approach to the Ralance of Payments:
Journal of International Economics 7 (1977), pp. 231-249:
North-Holland Publishing Company.
Intriligator, M.D.:
Econometric Models, Techniques and Applications:
Prentice Hall , ~ , Englewood Cliffs, New Jersey, 1978.
IMF Staff Papers:
Vol. 28, No. 1, March 1981:
Steinherr, Alfred:
Effectiveness of Exchange Rate Policy for Trade Account Adjustment.
26: 1979:
Hwa, E.C.:
Price Qetermination in
Several International Primary Commodity Markets:
A Structural
Analysis.
23 (1976):
Belanger, Gerard:
An Indicator of
Effective Exchange Rate for Primary Producing Countries.
20 (1973):
Yung-Chul Park:
The Role of Money in
Stahilization Policy in neveloping Countries.
26 (1979):
pp. 450-512:
Mathieson, Donald J.:
- - - - - - - - - - -
Financial Reform and Capnal Flows in a Developing Economy.
116
26 (1979):
Day, William~:
Domestic Credit
and Money Ceilings under Alternative Exchange Rate Regimes.
8 (1960-61):
Fleming, Marcus J. and
Boissonneault, Lorette: Money Supply & Imports.
8 (1960-61):
Prais, S.J.:
Some Mathematical
Notes on the Quantity Theory of Money in an Open Economy.
9 (1962):
Fleming, Marcus~:
Domestic Financial
Policies under Fixed and under Floating Exchange Rates.
Vol. VI, No. 1957 No. 1:
Polak, J.J.:
Monetary
Analysis of Income Formation and Payments Problems.
March 1982:
Sergio Pereira Leite:
interest Rate
Policies in West Africa.
Vol. 28: No. 1: 1981:
Crockett, Andrew D.:
Stabilization Policies in Developing Countries:
Some Policy
Considerations.
Vol. 28: No. 1: 1981:
Khan, Moshin ~ and Knight,
Malcom D.:
Stabilization Programs in Developing Countries:
A
Formal Framework.
20 (1973):
Guitian, Manuel:
Credit vs.' Money as an
Instrument of Control.
23 (1976):
Omotunde E.G., Johnson:
The Exchange Rate
as an Instrument of Policy in a Oeveloping Country.
Johnson, Harry G.:
The Monetary Approach to the Balance of Payments:
a nontechnical guide:
Journal of International Economics 7(1977),
pp. 251-268: North-Holla~d Publishing Company.
117
Kouri, Petti J.K.:
The Hypothesis of Offsetting Capital Flows:
A
CaseStudy of Germany:
Journal ~ Monetary Economics:
1 (1975)
pp. 21-39:
North-Holland Publishing Company.
Kravis, Irving B. and Lipsey, Robert E.:
Price Behavior in the Light
of Balance of Payments Theories:
Journal of International
Economics 8 (1978) p. 193-246: North-Holland Publishing Co.
Krueger, Larry, Monson, Akrasanee:
Trade & Employment in Developing
Countries, p. 239:
The National Bureau of Economic Research:
1981.
Kmenta, Jan:
Elements of Econometrics:
MacMillan Publishing Company,
Inc., 1971.
MacBean, Alasdair I.:
Export Instability & Economic Development:
Harvard University Press:
Cambridge, Massachusetts (196).
Morton, Kathyr~ and Tulloch, Peter:
Trade and Developing Countries:
Chs. 7 and 8:
Halsted Press Book:
John Wiley ~ Sons
(1977).
Miller, Ronald E.:
Dynamic Optimization and Economic Applications:
McGraw-Hill International Book Company (1979).
McKinnon, Ronald I.:
Foreign Trade Regimes &Economic Development:
.Journal of Internati anal Economics
9 (1979), pp. 429-452.
Onoh, J. K.:
Money and Bankinq in Africa:
Longman Group Ltd. (1982):
New York.
Swoboda, A.:
Monetary Policy under Fixed Exchange Rates:
The
Effectiveness, Speed of Adiustment anrl Proper Use:
Economic~ 4n
(1973).
Short, Brock K.:
The Velocity of Money per Capita Income in
~:"AfR/~
Developing Economies:
Malaysia and Sin~:::v.~~(~c~.:~~~~.~"~:_:~,O;f
Development Studies, January 1973.
I~, ~~ 1 ~
Taro, Yamane:
Mathemat i cs for Economi cs:
ft...~~El ementa ry- /SuJ!;vey : 2nd
/.r~~\\'.'
ed. Prentice Hall Inc., Englewood Cliffs, <N.e,~ Jersey..Q,'(1962).
- - - -
"--.~~.