NOVEMBER 1994

RESEARCH PAPER TWENTY-SIX

EXCHANGE RATE
DEPRECIATION, BUDGET
DEFICIT AND INFLATION -
THE NIGERIAN EXPERIENCE

FESTUS 0. EGWAIKHIDE
LOUIS N. CHETE
and
GABRIEL 0. FALOKUN

ARC HIV OMIC RESEARCH CONSORTIUM

102039
OUR LA RECHERCHE ECONOMIQUE EN AF1UQUE
 DRC - Ub.

Exchange rate depreciation,

budget deficit and inflation
the Nigerian experience
Other publications in the AERC Research Paper Series:

Structural Adjustment Programmes and the Coffee Sector in Uganda by

Germina S semogerere, Research Paper 1.

Real Interest Rates and the Mobilization of Private Savings in Africa by F.M.
Mwega, S.M. Ngola and N. Mwangi, Research Paper 2.

Mobilizing Domestic Resources for Capital Formation in Ghana: The Role of

Informal Financial Markets by Ernest Aryeetey and Fritz Gockel, Research
Paper 3.

The Informal Financial Sector and Macroeconomic Adjustment in Malawi by

C. Chipeta and M.L.C. Mkandawire, Research Paper 4.

The Effects of Non-Bank Financial Intermediaries on Demand for Money in

Kenya by S.M. Ndele, Research Paper 5.

Exchange Rate policy and Macroeconomic Performance in Ghana by C.D.

Jebuni, N.K. Sowa and K.S. Tutu, Research Paper 6.

A Macroeconomic-Demographic Model for Ethiopia by Asmerom Kidane,

Research Paper 7.

Macroeconomic Approach to External Debt: the Case of Nigeria by S. Ibi

Ajayi, Research Paper 8.

The Real Exchange Rate and Ghana's Agricultural Exports, by K. Yerfi Fosu,
Research Paper 9.

The Relationship Between the Formal and Informal Sectors of the Financial
Market in Ghana by E. Aryeetey, Research Paper 10.

Financial System Regulation, Deregulation and Savings Mobilization in

Nigeria by A. Soyibo and F. Adekanye, Research Paper 11.

The Savings-Investment Process in Nigeria: an Empirical Study of the Supply

Side by A. Soylbo, Research Paper 12.

Growth and Foreign Debt: the Ethiopian Experience, 1964-1986 by B.

Degefe, Research Paper 13.
Links Between the Informal and Formal/Semi-formal Financial sectors in
Malawi by C. Chipeta and M.L.C. Mkandawire, Research Paper 14.

The Determinants of Fiscal Deficit and Fiscal Adjustment in Côte d'Ivoire by

0. Kouassy and B. Bohoun, Research Paper 15.

Small and Medium-Scale Enterprise Development in Nigeria by D.E.

Ekpenyong and M.O. Nyong, Research Paper 16.

The Nigerian Banking System in the Context of Policies of Financial

Regulation and Deregulation, by A. Soyibo and F. Adekanye,
Research Paper 17.

Scope, Structure and Policy Implications of Informal Financial Markets in

Tanzania by M. Hyuha, M.O. Ndanshau and J.P. Kipokola, Research
Paper 18.

European Economic Integration and the Franc Zone: the Future of the CFA
Franc after 1996. Part I: Historical Background and a New Evaluation of
Monetary Co-operation in the CFA Countries by Allechi M'bet and
Madeleine Niamkey, Research Paper 19.

Revenue Productivity Implications of Tw Reform in Tanzania by Nehemiah E.

Osoro, Research Paper 20.

The Informal and Semi-Formal Sectors in Ethiopia: a Study of the Iqqub, Iddir
and Savings and Credit Co-operatives, by Dejene Aredo, Research
Paper 21.

Inflationary Trends and Control in Ghana, by Nii K. Sowa and John K.

Kwakye, Research Paper 22.

Macroeconomic constraints and medium-term growth in Kenya: a three gap

analysis, by F.M. Mwega, Njuguna Mwangi and F. Olewe-Ochilo, Research
Paper 23.

The foreign exchange market and the Dutch auction system in Ghana, by
Cletus K. Dordunoo, Research Paper 24.

Exchange rate depreciation and the structure of sectoral prices in Nigeria

under an alternative pricing regime, 1986-89, by Olu Ajakaiye and Ode
Ajowu, Research Paper 25.
 Exchange rate depreciation,
budget deficit and inflation
the Nigerian experience

Festus 0. Egwaikhide
Louis N. Chete
Gabriel 0. Falokun
Nigerian Institute of Social and Economic Research (NISER)

AERC Research Paper 26

African Economic Research Consortium, Nairobi
November 1994
© African Economic Research Consortium, 1994

Typeset by Centre for the Study of African Economies,

University of Oxford,
St Cross Building, Manor Road,
Oxford, OX1 3UL,
England.

for the African Economic Research Consortium,

P.O. Box 62882, Nairobi, Kenya.

Printed by The Regal Press Kenya Ltd.,

P.O. Box 46166,
Nairobi, Kenya.

ISBN 9966-900-13-6
Contents

List of tables
List of figures
Acknowledgements

I Introduction 1

II Proximate causes of inflation in Nigeria - some

preliminary observations from trends and relationships 3
III Review of related studies 11
IV Conceptual links and the model 15
V Model estimation and solution 22
VI Policy simulation results 29
VII Conclusion 32

Appendix A: Definition of the estimated model variables 34

Appendix B: Unit root tests 35
Appendix C: Inflation rate and discount rate 36

Notes 37
References 40
List of tables

1. Nigeria: Monetary survey, 1970-89 5

2. Long-run solution of the cointegrating relationship 24
3. Estimate of behavioural equations 25
4. Theil's inequality coefficients and
their decomposition for some key variables 28
5. Effects of exchange rate depreciation on
budget and inflation: a simulation experiment
(mean annual growth rate, 1984-89) 30

List of figures

1. Inflation rate and budget deficit 4

2. Inflation rate and growth in Ml 6
3. Inflation rate and growth in real GDP 7
4. Inflation rate and official exchange rate 8
5. Inflation rate and parallel market exchange rate 8
Acknowledgements

The execution of this empirical enquiry has benefited from the valuable
comments and suggestions of the participants and resource persons at the
various workshops organised by the AERC, especially between December 1990
and May 1992. In particular, Benno J. Ndulu's initial reactions to the proposal
for this research, together with the technical assistance of Mohsin S. Khan and
Christopher S. Adam helped to improve every section of this study and we are
happy to acknowledge it here.

We are also thankful to Olu D. Ajakaiye, Rasaq A. Olopoenia, Mufutau I.

Raheem and Fidelis 0. Ogwumike, for helpful comments and suggestions.
However, the views expressed are those of the authors and do not necessarily
represent those of the AERC and NISER.
I Introduction

One of the thorniest issues in Nigeria's policy arena today is how to put
inflation under effective control. The control of inflation has been central to
both monetary and fiscal policy in the last few years, as demonstrated in the
various budgets and policy statements. Historically, the origin of the current
inflation dates back to the 1970s, when the revenue accruing to the government
from the non-renewable oil resource rose steeply. With the increase in public
expenditure, enhanced by oil revenues, there was vast expansion in aggregate
demand. With the inelastic supply of domestic output, inflation inevitably
resulted. The rapid growth in money supply, as a result of the monetisation of
the earnings from oil, also exerted upward pressure on the general price level.
The price of crude oil slumped in the world market during the first half of
the 1980s. Thus, Nigeria's crude oil, which sold at slightly above US$41 a
barrel in early 1981, fell precipitously to less than US$9 by August 1986. This
triggered a series of developments in the economy. One example is the state's
fiscal crisis, as reflected in the persistent and substantial budget deficit which
cumulated to approximately N17.4 billion in the five years between 1980 and
1984. Monetary policy became highly expansionary as a large part of the
deficits incurred during this period were financed through the creation of
credit. Indeed, the total domestic credit to the economy recorded an average
annual growth rate of 29.9% in 1980—84 and most of the increase was
attributable to net claims by the government. Simultaneously, two-digit
inflation at a mean yearly rate of 20.2% was registered, clear evidence,
perhaps, in support of the monetarist proposition. But the inflation in 1984,
which stood at almost 40%, is often explained in terms of acute shortages of
imported goods and services imposed by inadequate foreign exchange earnings,
a derivative of the steep fall in crude oil prices.
With the deepening internal and external disequilibria, it became imperative
to adopt the Structural Adjustment Programme (SAP) from July 1986. The
SAP, which is predicated mainly on the principle of 'getting prices right', has
exchange rate reform as its central focus1. In pursuit of this, the Second-Tier
Foreign Exchange Market (SFEM) was introduced in late September 1986 and
since then the naira has depreciated sharply against the US dollar and other
2 RESEARCH PAPER 26

major currencies. Quantitatively, the naira, which traded at N4.62:$ 1.00 at the
inception of SFEM in late September 1986, had, by the end of 1989, exceeded
N7.65:$1.00, a change of almost 65.6%. During the same period, inflation leapt
from barely 5.0% to almost 41.0%.
This development shows that the depreciation of the naira has a role to play
in Nigeria's recent inflationary process. Concomitant with this is the substantial
budget deficit operated annually by the Federal Government in the last decade
or so. Part of the budget deficit is financed through bank credit which directly
affects the money base. This also exerts upward pressure on the general price
level. All this suggests that there are many sources of the current inflation.
While the channels through which exchange rate depreciation affects prices are
well known, the extent to which this phenomenon engenders price inflation in
Nigeria is still not well researched.
As part of the attempt to fill this lacuna, this study examines the
quantitative effects of exchange rate depreciation on budget deficit and
inflation in Nigeria. This is achieved in two stages. First, a structural model of
the interaction between exchange rate, budget deficit, inflation, and government
revenue and expenditure is constructed. In doing this, we are influenced by
recent developments on cointegration and the error correction model. Second,
a simulation experiment on the impact of exchange rate movements on the
general price level and budget deficit, in particular, is conducted. Policy
implications of the simulation exercise are then derived.
This report presents trends in the relationship between exchange rate, budget
deficit and inflation. It is complemented by recent empirical literature on
inflation. Thereafter, a conceptual link between the above macroeconomic
variables is sketched. The macro model developed is rooted in some of the key
linkages and reflects some of the indications from preliminary observations
from trend analysis.
Estimates of the individual equation are presented and results discussed.
This is followed by the solution of the entire model, using Theil's inequality
coefficient and its decomposition to evaluate the model's performance. Finally,
we present the policy simulation results and our concluding remarks.
II Proximate causes of inflation in Nigeria —
some preliminary observations
from trends and relationships

A proper understanding of the determinants of inflation in Nigeria requires

adequate discussion of the movements of the relevant variables over time. In
response to this, we shall focus mainly on trends in inflation, budget deficit,
monetary growth, growth in real GDP and exchange rate movements in the
period between 1970 and 1989.
After the civil war in January 1970, there was a need to restructure the war-
ravaged economy under the economic Second National Development Plan
(1970—74). A major constraint confronting the government was the inadequacy
of funds to execute the Plan's projects and programmes. Fortunately, the price
of crude oil shot up in the world market due to the Middle East Crisis
(1973—74). The average export price per barrel of Nigerian crude oil rose from
a tiny level of US$4.13 in 1973 to US$14.74 in 1977 and to US$35.18 in
1980. Combined with this were various petroleum tax reforms at the start of
the oil boom2. The aggregate effect of these was increased government
revenue. Quantitatively, the federally collected revenue, which stood at N633.3
million in 1970, rose to N4.5 billion in 1974 and by 1980 had exceeded N15.0
billion.
Parallel to this was vast expansion in aggregate demand, led by growth of
public expenditure. Indeed, in the decade 1973—83 the Keynesian public-
expenditure-led growth (enhanced by oil revenues) was pursued by
government. So it is not surprising that the Federal Government's outlay,
which was approximately N839 million in 1970, leapt to N4.9 billion in 1975
and stood at over N14.0 billion by 1980. During this period, central
government expenditure grew at a compound annual growth rate of 6.9%.
While it is generally appreciated that the rise in public spending was to ensure
that a large proportion of the population benefited from oil exploitation, this
expenditure generated persistent budget deficits, as it exceeded revenues, in the
second-half of the 1970s (except in 1979). Between 1975 and 1978, the
cumulative budget deficit was about N4.8 billion. As a percentage of GDP,
budget deficit fluctuated between 2 and 6.7% in the period. However, as
4 RESEARCH PAPER 26

indicated in Figure 1, there seems to be no correlation between fiscal deficit

and inflation.
In part, the budget deficits were financed from bank credit and the balance
came from external loans3. Thus, monetary policy, like fiscal policy, was
expansionary during the oil boom. Table 1 shows Nigeria's financial data for
the period 1970 through 1989. The table demonstrates that there were increases
in domestic credit in the economy, especially from 1976. It is almost incredible
that the average annual growth rate of domestic credit stood at about 65% in
1976—80. In absolute terms, total domestic credit accelerated from N3. 1 billion
in 1976 to about N10.7 billion in 1980. Out of this, the share of net credit to
the central government fluctuated between 20.2 and 40.8%.

Figure 1 Inflation rate and budget deficit

40 — Inflation rate
Budget deficit

30

>. 20
0
C
a,

" 10
U-

\'

—10
\ \/ \ .1
'I
,\ -...

—20 I I I I I I I I I I I I I I I I I

70 72 74 76 78 80 82 84 86 88

The growth of credit to government had two important direct effects. First, it
vastly expanded aggregate demand and second, it accelerated the growth of
domestic money supply. There was a rise in total imports as a result of the
upsurge in aggregate demand that could not be met by the available supply in
the economy. Available data shows that merchandise import registered an
average annual growth rate of almost 23% in the five years from 1976—80.
This was actually enhanced by the expansion in bank credit arising largely
from fiscal deficit. Thus, the direct correspondence between budget deficit and
the current account deficit is easily appreciated.
Table 1 Nigeria: Monetary survey 1970-89 (N million)

C redit to government Credit to p rivate sector

Net foreign Domestic Value Percentage Value Percentage
assets credit N million share (%) N million share (%)
Year A B C (C/B)

1970 154 1,141 666 58.4 475 41.6

1971 280 1,127 532 47.2 593 52.6
1972 244 1,274 518 40.7 749 58.8
1973 414 1,399 545 39.0 847 60.5
1974 3,500 -176 -1,310 744.3 1,121 -636.9
1975 3,668 1,161 -631 -54.3 1,750 150.7
1976 3,396 3,073 622 20.2 2,382 77.5
1977 2,962 6,010 2,452 40.8 3,459 57.6
1978 1,420 7,706 3,143 40.8 4,485 58.2
1979 3,228 8,662 3,313 38.2 5,126 59.2
1980 5,607 10,689 3,539 33.1 6,744 63.1
1981 2,556 15,746 6,299 40.0 8,917 56.6
1982 1,057 21,410 10,328 482 10,567 49.4
1983 808 27,590 15,465 56.1 11,071 40.1
1984 1,423 30,423 17,823 58.6 11,823 38.9
1985 1,816 31,900 18,297 57.4 12,822 40.2
1986 4,463 36,409 18,827 51.7 16,690 45.8
1987 6,865 40,209 21,157 52.6 17,918 44.6
1988 7,974 50,680 27,488 54.2 20,857 41.1
1989 20,044 36,337 6,879 18.9 25,832 71.1

Source: International Monetary Fund, International Financial Statistics Year Book 1990.
6 RESEARCH PAPER 26

More important to the current discussion is the pressure which monetary

expansion exerts on price inflation. Following the ebullient growth of domestic
credit, it was not surprising that the mean annual growth rate of money supply
(MI) stood at 33.7% during 1975—80. This would probably have been higher,
but for the negative growth rate of 8.2% recorded in 1978. On average, the
period of high monetary growth coincides with a high inflationary trend, as
demonstrated by the growth of the consumer price index, even though price
inflation generally declined from 33.3% to almost 10% between 1975 and
1980. This is summarized in Figure 2.

Figure 2 Inflation rate and growth in Ml

90r — Inflation rate

I
——— Growth in Ml
801-

70

/_\ ,,

—10 I I I I I I I I I

70 72 74 76 78 80 82 84 86 88

In the 1970s, there does not seem to have been a strong contracyclical
relationship, in line with theoretical expectation, between the growth of output,
proxied by the annual growth of real GDP, and inflation (see Figure 3).
Although real GDP recorded an impressive annual growth in 1970—72, double-
digit inflation was registered in 1970—71 and inflation fell sharply to
approximately 3.3% by 1972. Between 1974 and 1980, the expected
relationship seems validated, though very weak. This tends to support the
argument by Schatz (1984) that a sizeable proportion of the increase in
aggregate demand was dissipated through inflation. The decline in Nigeria's
other-than-oil GDP in the 1970s to the early 1980s possibly reinforces this
argument4.
 EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 7

Figure 3 Inflation rate and growth in real GOP

Inflation rate
Growth in real GDP

>.
C,

4,
a.
4)
U.

Between 1974 and 1980, the mean yearly growth rate of nominal exchange
rate (official) was -2.5%. This is an indication that the naira appreciated
against the US dollar during the period. Juxtaposed with this was a high rate
of inflation which stood at 18.2%. A statement on the type of exchange rate
regime adopted during the period under review may provide a useful insight
into the observed relationship5. The local currency was fixed against the US
dollar up to 1970, and in December 1971 the Nigerian pound was tied to the
dollar. But a system of independent exchange rates (i.e., the naira exchange
rate was independently fixed against the US dollar and the UK pound sterling)
was pursued from 1974, although in practice the dollar/sterling cross rates
actually determined the naira exchange rate against the US and UK currencies.
From February 1978, the naira exchange rate was based on a basket of
currencies of Nigeria's major trading partners (UK, France, Japan, the
Netherlands, Switzerland, the US and the former West Germany). One point
is probably obvious from this sketchy presentation of Nigeria's exchange
rate
regime: the exchange rate policy was not employed as a principal instrument
of economic management in Nigeria until the mid-1970s. The refusal to adopt
an exchange rate policy to correct Nigeria's balance of payments
difficulties
of the 1960s is a graphic manifestation of this. Thus, exchange rate movements
bear little relationship to inflation, as shown in Figure 4.
With respect to the parallel market exchange rate, the story seems to
be
quite different. While the mean growth rate of official exchange revealed that
it appreciated in 1976—80, the parallel rate depreciated annually by
1.9%, on
average, over the same period. Despite this, the yearly growth of consumer
8 RESEARCH PAPER 26

price index prior to 1981 seems not to mirror the shadow price of foreign
exchange, as shown in Figure 5.

Figure 4 Inflation rate and official exchange rate

250 — Inflation
(official
EO
rate
exchange rate)

200

L100

50
/
l/t

I I I I I I I I I I I I

70 72 74 76 78 80 82 84 86 88

Figure 5 Inflation rate and parallel market exchange rate

— Inflation rate
EP (parallel market exchange rate)

>.
0
C

0•
LL.
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 9

The appreciation of the official exchange rate artificially cheapened

competitive imports in relation to locally produced goods and services. With
the massive import of consumer goods, raw materials and spare parts, the issue
of imported inflation is easily appreciated when it is realised that in 1974—80
the mean yearly growth of the import price index registered 10.7% and
inflation stood at 18.2%.
This was basically the situation between 1970 and 1980. From 1981, the
structural weaknesses of the oil-propelled economy were exposed with the
collapse of crude oil prices in the world market. It is almost incredible that the
price per barrel of Nigerian crude oil, marginally above US$41 in January
1981, declined steeply to less than US$9 by August 1986. This had a direct
impact on government revenue, as federally collected revenue declined from
N15.2 billion in 1980 to N10.5 billion in 1983. The unchecked expansion of
public expenditure under the civilian administration, which could not be
serviced adequately by the shrinking revenue, generated a persistent budget
deficit in the period. Not surprisingly, between 1981 and 1984, the cumulative
budget deficit of the Federal Government was approximately N 15.4 biflion.
Yet, the relationship between budget deficit and inflation (see Figure 1) in the
reference period is practically non-existent. Perhaps this is due to the
monetization of the deficits that exerted upward pressure on the general price
level.
A striking revelation in the first half of the 1980s was the increase in
domestic credit, with a sizeable proportion of this attributable to net claims on
the government6. From N15.7 billion in 1981, the total domestic credit to the
economy leapt to more than N30.0 billion in 1984. Correspondingly, net credit
to the government accelerated from N6.3 to N17.8 billion. Although the annual
growth of Ml was generally low in the first half of the 1980s relative to any
period in the 1970s, the inflation rate climbed phenomenally from 21% in 1981
to almost 40% in 1984, although there was a big dip in 1982. Figure 2 shows
that inflation in the early 1980s may not have been explained by monetary
factors alone. Rather, structural factors could have accounted for inflationary
trends during this period.
It will be recalled that there were supply shortages, particularly in 1984, due
to import restrictions to correct the chronic balance of payment difficulties.
Indeed, Figure 3 demonstrates that real GDP registered negative growth rates
in 1981 and 1983—84, and the negative rates correspond to high rates of price
inflation. The depreciation of the official exchange rate could have exerted
some pressure on prices, as shown in Figure 4. But the picture that emerges
in Figure 5 demonstrates a strong positive correspondence between inflation
and the parallel market exchange rate relative to the official exchange rate
identified in Figure 4.
10 RESEARCH PAPER 26

The persistence of both internal and external disequilibria led to the

implementation of the SAP in July 1986. The adoption of a realistic exchange
rate via the SFEM, begun in late September 1986, is central to this. Available
statistics reveal that the naira has depreciated steeply against the US dollar
since then. On average, the naira, which before 1985 traded for almost
N1.00:$1.00, stood at approximately N3.32:$l.00, a change of over 130%. The
yearly percentage change of the official exchange rate has been very high
indeed. Simultaneously, the rate of inflation accelerated from 5.4% in 1986 to
about 41% by 1989. Figure 4 shows that there is some correspondence in the
direction of movements between inflation and the nominal official exchange
rate since 1987. Overall, the parallel market rate appears to correlate with
inflation more than the official rate.
Demand management aspects of SAP emphasize reduced public
expenditures and, therefore, a fall in budget deficit. Although, budget
deficit/GDP ratio peaked at 11.9% in 1986, it declined remarkably to 5.5% in
1987 and rose through 8.5% in 1988 to 9.0% in 1989. However, the magnitude
of the fiscal deficit increased from almost N5.9 billion to N15.3 billion
between 1987 and 1989. This was partly financed from bank credit, with the
rapid growth of money supply as the inevitable concomitant. In general, for
this period, Figures 1 and 2 demonstrate a strong association between budget
deficit/GDP ratio, money supply, and inflation.
In summary, two tentative conclusions emerge from trend analysis. First, the
Nigerian inflation since 1970 can be explained by a combination of monetary
and structural factors. This seems to confirm earlier empirical studies of
inflation in Nigeria (see Section III). Second, exchange rate movement
engenders price inflation. The parallel market exchange rate appears more
significant than the official exchange rate in this respect. This suggests that
prices may have adjusted to the shadow price of the exchange rate7, (an
empirical verification of this for Ghana can be found in Chhibber and Shafik
(1990a)). Evidence from graphical representation suggests that the devaluation
of exchange rates tends to slow the rate at which the parallel market exchange
rate depreciates.
Ill Review of related studies

There is a vast body of empirical literature on inflation and this is usually

dichotomized into two parts: the structuralist and monetarist perspectives8. Our
brief review here is not bifurcated into the structuralist-monetarist controversy.
We begin with studies on inflation in Nigeria.
In 1974, a national conference on inflation in Nigeria was organised by the
Nigerian Institute of Social and Economic Research (NISER), Ibadan. Several
aspects were addressed, but the papers prepared for Section Two of the
conference focused on the proximate causes of inflation9. In general, the
findings of some of the key articles reveal that neither monetary nor structural
factors alone explain the Nigerian inflation. Striking evidence from this
conference was that a combination of both factors precipitate the inflationary
process.
Prior to the NISER conference on inflation, few studies had addressed the
issue of inflation. The work of Oyejide (1972) is particularly appealing, as he
takes the impact of deficit financing in the course of inflation as the focal
point of his empirical enquiry. Having established the theoretical link between
domestic money supply and inflation from the Fisherine equation, Oyejide
determined statistically the impact of alternative definitions of deficit financing
on inflation. Evidence from this research demonstrates that there is a direct
correspondence, though not on a one-to-one basis, between the general price
level and measures of deficit financing over the 14 years from 1957_70b0. One
point of importance from this is that less emphasis on deficit financing may
limit the growth of price inflation in Nigeria.
The results of Ajayi and Awosika (1980) can be juxtaposed against this. An
important conclusion from various econometric models employed by these
authors indicates that inflation in Nigeria is determined largely by
developments in the external sector, but complemented by internal influences.
Specifically, their findings demonstrate that the openness of the economy is
highly correlated with inflation. For Pinto (1987), the monetization of the
foreign exchange earnings trom crude oil export, that vastly expanded the
growth of Ml, constituted the single most important factor to explain
movements in the general price levels in the 1970s to the early 1980s.
12 RESEARCH PAPER 26

Elsewhere, the causes of inflation have been the preoccupation of several

studies, particularly in recent years. It is important to document the findings
of those related to the present study. In this respect, the work of Aghevli and
Khan (1978) is illuminating. These authors developed structural equations to
demonstrate the two-way causation between budget deficit and inflation in
developing countries". Empirical estimates from their study indicate that
government expenditures respond faster to inflation than revenue, thereby
generating an enlarged budget deficit which further engenders inflation.
Chhibber et al. (1989) developed a detailed econometric model which takes
into account both monetary and structural factors in the course of inflation in
Zimbabwe. Their investigation shows that monetary growth, foreign prices,
exchange and interest rates, unit labour cost and real income are the
determinants of inflation in this country. A similar macro-economic model of
inflation was employed for Ghana by Chhibber and Shafik (1990a). This study,
which covers 1965—88, suggests that the growth of money supply is one key
variable explaining the Ghanaian inflationary process. Such variables as official
exchange rates and real wages could not exert any significant influence on
inflation. However, a significant positive relationship was found between the
parallel exchange rate and the general price level. Perhaps one policy
implication arising from this is that recent price movements in Ghana have
little relationship with the recent exchange rate policy implemented by the
government.
Still on the issue of inflation, Chhibber (1991) posits that there is no one-
and-only-one relationship between exchange rate and price inflation. Basing his
argument on empirical studies of some African countries, one of his main
conclusions is that devaluation could exert upward pressure on the general
price level through its increased cost of production in the short-run. For
Chhibber, the extent to which devaluation of a local currency engenders
inflation is largely a function of the impact of such policy measures on the
revenues and expenditures (budget) of government, together with the monetary
policy that is simultaneously pursued.
Probably motivated by the findings of Chhibber and Shafik (1990a), Sowa
and Kwakye (1991) also undertook a study of inflationary trends and control
in Ghana. A highly simplified econometric model was employed to determine
the relative effects of monetary and structural factors on the general price
level. Their results show that monetary expansion exerted some influence on
inflation. On the impact of the exchange rate, this variable could not have a
significant direct relationship with price movement, a confirmation of one of
the findings of Chhibber and Shafik. From their findings, the conclusion of
Sowa and Kwakye is that the Ghanaian inflation is structural in character.
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 13

Focusing on Uganda, Elbadawi's (1990) research revealed that rapid

monetary expansion and the precipitous depreciation of the parallel exchange
rate were the principal determinants of inflation during 1988—89. He concluded
from the comprehensive review of exchange rate and price movements that
devaluation of the official exchange rate is not inflationary. Obviously, this
conclusion is consistent with the findings of Chhibber and Shafik (1990a) and
Sowa and Kwakye (1991) with respect to Ghana.
The work of Tegene (1989) cannot be ignored. His method of analysis
departs from others, as he does not utilize econometric techniques to
investigate the role of domestic money supply in the course of inflation in six
African countries. Rather, he employs the Granger and Piece causality tests'2.
Evidence from this study demonstrates a unidirectional causality, from
monetary growth to inflation, in the sample countries. A similar analytical
methodology was employed by Canetti and Greene (1991) to evaluate the
relative contributions of exchange rate movements and monetary expansion to
price inflation in ten African countries during The broad conclusion
that emerged from this comprehensive investigation is that exchange rate
movement and monetary growth explain the inflationary trend in the study
countries. In countries such as Sierra Leone, Tanzania and Zaire, the bivariate
and trivariate Granger tests point out that the exchange rate has significant
causal influence on inflation. With respect to the role of money supply, the
statistical test identified causation from money to prices in Gambia, Sierra
Leone and Uganda. As for Nigeria and Zambia, the various tests performed
could not identify any significant causal relationship between money supply,
exchange rate and inflation.
Earlier, London (1989) had examined the role of money supply and
exchange rate in the inflationary process in 23 African countries14. The pure
monetarist model of the Harberger type was employed and the results revealed
that in the period between 1974 and 1985 the growth of money supply,
expected inflation and real income were significant determinants of inflation
in the sample countries. London, however, argued that because the results
obtained give account only of the period averages of the countries studies, they
should be seen as suggestive rather than definitive. In a related sense, the
coefficients of the regressors may not adequately reflect the developments in
a particular country; hence, the results should be interpreted with respect to a
'typical country on average over the period' (London, 1989, p. 95). The
exchange rate was later introduced as one of the explanatory variables in the
pure monetarist model. The results of this indicate that exchange rate
movements had a significant impact on the inflationary process in the 1980s.
Conversely, the growth of the money supply played a decreasing role in the
14 RESEARCH PAPER 26

course of inflation on the continent. This possibly suggests that structural

elements have been the proximate cause of inflation in recent years.
IV Conceptual links and the model

Analytical underpinning

The conceptualization of the linkages between exchange rate and budget

deficit, on the one hand, and between each of these variables and inflation, on
the other, is relatively straight-forward. This section sketches only the principal
channels, beginning with the channels of transmission between exchange rate
and inflation.
The 'pass through' argument is usually the basis on which the inflationary
tendencies of exchange rate devaluation are premised. This is predicated on the
assumption that the induced increases in the prices of imported inputs and fmal
goods, following a devaluation of the local currency, will be passed on to
domestic prices. From this it is obvious that this proposition has its roots in the
cost-push theory of inflation. This is not to deny that theoretical and empirical
strand of the literature which is predicated on the purchasing power parity
(PPP) doctrine which is discussed in several papers (e.g. Leith, 1991; Balassa,
1964; Dombush, 1987; and Edwards, 1989).
In general, there is high dependence on imported inputs for production by
most developing countries. When there is devaluation, the domestic prices of
imported inputs are raised and production costs are accordingly affected.
Profits plus indirect taxes are usually mark-ups on producer prices to obtain
ex-factory prices (and ex-factory prices plus distributor's margins equals
market prices). In this sense, prices are formed on the basis of mark-up over
cost of production. A highly simplified model of this is often represented as15:

P = M(W + eP*)

where P represents the output price, M stands for 1 plus the fixed mark-up
rate, W is the wage rate, e denotes exchange rate and is the foreign price
of imported inputs.
16 RESEARCH PAPER 26

The above equation assumes a constant mark-up. But the argument has been
advanced that the size of the mark-up is largely a function of excess demand
in the economy. (Chhibber et al., 1989, p. 7).
The rise in the general price level engendered by devaluation usually
triggers a series of developments that often fuel the inflationary process. Even
the rise in domestic prices without a corresponding increase in nominal wage
rate reduces the real wage and a household has to spend more in order to
maintain the same living standard. There is the tendency for labour to agitate
for increases in wages and benefits. When such demands are granted,
production costs and, therefore, market prices are affected accordingly. Of
course, this is not inevitable if labour productivity increases correspondingly.
For most developing countries, the government is the major employer of labour
and as such, increases in wages have a tendency to raise public outlays: the
rise in government expenditure generates budget deficit when the revenue
collected is inadequate to meet the expansion in expenditures. This leads us to
budget deficit-inflation links.
The channels of transmission between budget deficit and inflation are easily
appreciated when the deficit is financed, in part, from increased Central Bank
credit to the government. Two immediate direct effects of this can be
identified. The increase in bank credit to the public sector expands aggregate
demand and enhanced public expenditure has a tendency to raise private sector
income and, therefore, demand for goods and services via the multiplier
process. For a developing country that has a domestic limited supply of goods
and services, the expansion in aggregate demand tends to exceed supply and
a sustained rise in the general price level inevitably results.
The second effect is easily understood and domestic money supply is central
to this. The growth of bank credit directly influences the growth of the money
base (i.e., high-powered money) which, in turn, expands the growth of the
money supply. The direct correspondence between money and price inflation
is well known. Even within this monetary framework, it has been argued that
the existence of an excess supply of real money balances directly stimulates
real private expenditure.
We now explore the relationship between the exchange rate and budget
deficits. This task has been simplified by the work of Chhibber et al. (1989).
Exchange rate movements simultaneously affect both the revenue and
expenditure sides of the budget. When there is devaluation, for instance,
external debt and debt service payments in domestic currency rise. Thus, a
component of the overall public expenditure is increased.
Theoretically, because devaluation stimulates the production of exports, it
raises the income of exporters and subsequently increases revenue generated
from taxes (Corden, 1989). Revenues from export duties rise with increased
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 17

export production consequent upon devaluation of the exchange rate. The

extent to which the volume of import decreases is a function of the price
elasticity of demand. But Corden (1989) argued that a devaluation of the
exchange rate raises the domestic currency value of government imports.
Changes in the level of imports affect revenue realised from customs duties.
It follows from this that the net effect of exchange rate movements on the
budget is largely a function of tax structures and expenditures. Should the net
effect be fiscal deficit that is monetized, inflation results. Both government
expenditure and revenue adjust to inflation disproportionately and an enlarged
budget deficit is generated if expenditure responds faster than revenue. Perhaps
this discussion presents the two-way causality between inflation and budget
deficit, as they feed on each other through what is referred to in the literature
as the Keynes-Olivera-Tanzi effect (Tanzi, 1977)16.

Model description

The macroeconomic model developed here takes account of the principal

linkages and reflects some of the indications from preliminary observations
presented earlier. It follows that the central issue of determining the
quantitative impact of exchange rate adjustments on budget deficit and the
price level is considered. Specifically, the macro model consists of a number
of equations explaining price behaviour, domestic money supply and
government budgetary developments. These are discussed in turn.

Price equation
The price equation considers the monetarist variables in addition to the
exchange rate. Thus, price is functionally related to money supply, real output,
expected rate of inflation and exchange rate, so that:

(1) in = a0 + a1 in M, + a2 in a3 pe, + a4 e,;

a2<O; a3>O; a4>O

where P stands for inflation, M represents money supply (broadly defined),

YR is real output proxied by real GDP, and pe and e are expected rate of
inflation and change in exchange rate, respectively17.
18 RESEARCH PAPER 26

It is hoped that the results of this equation will help to determine the relative
contributions of monetary factors and exchange rate to inflationary
developments in Nigeria.

Revenue equations
For simplicity, total government revenue is broken down into three
components: petroleum revenue (also called oil revenue, (PR)), revenue from
import duties (MR) and other government revenues (OR). That is:

(2) = PR1 + MR, + OR1

Revenue from import duties is directly related to import level which, in turn,
is influenced by exchange rate. This function which is expressed in log-linear
terms is as follows:

(3) ln MR, = B0 + b1 in (MT.e)1

where MT stands for the value of imports in foreign currency (expressed in

US dollars).

Next are other government revenues. These are determined by nominal income
(Y) and other government revenues lagged one year, so that:

(4) ln OR1 = c, + c, ln Y1 + c2 in OR,1

c1>O; c2>O.

Exchange rate movements affect the value of a country's exports and,

therefore, revenue generated from export duties. But an export revenue
equation could not be specified along with the revenue from import taxes —
as customs duties — simply because revenue from the former source paled
into insignificance from the mid—1970s. Moreover, from the 1987 fiscal year,
the Federal Government had eliminated export taxes as part of the open-
handed policies to boost non-oil exports. For these reasons, income from
export duties is considered under other government revenues.
Oil revenue is exogenously determined. This is because crude oil exports
and the price are determined by the Organisation of Petroleum Exporting
Countries (OPEC). However, it is recognised that with the devaluation of the
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 19

naira, revenue from crude oil export in local currency has increased
considerably; and, by definition:

(5) PR1 = (PRF.e)1

where PRF is the revenue from crude oil exports in US dollars.

Expenditure equations
Total government expenditure is divided into two for convenience. These are
interest charges on external debt plus capital repayment (DS) and other
government expenditures (GO).

(6) GE1 = DS1 + GO1

Interest charges on external debt and capital repayment depend on the

magnitude of the debt in local currency (and its local equivalent depends on
the prevailing exchange rate) and foreign interest rate. Thus, the estimated
equation is specified as:

(7) in DS1 = d0 + d1 in (FD.e)1 + d2 if

where FD is the magnitude of foreign debt in US dollars and rf represents
foreign interest on the debt.

Other government expenditures in real terms are explained by real income and
the magnitude of real government revenue. The lag value of other government
expenditures is included in this specification (as an adjustment lag) which tests
the responsiveness of government expenditures to inflation. Thus, the estimated
equation is as follows:

(8) ln (GOR)1 = e0 + e1 in YR, + e2 In (GRR), + e3 ln (GOR)11

The justification for the inclusion of government revenue in Equation (8) is

easily appreciated. Since the oil boom of the early 1970s, government
expenditure has been influenced largely by the earnings from crude oil. Indeed,
the Keynesian public expenditure-led growth pursued vigorously in this decade
(1973—83) was precipitated by the size of government revenue. Even since
1984, emphasis on reduced public spending through rationalization has been
sharpened by reduced government revenue, brought about by the collapse of
crude oil prices in the world market.
20 RESEARCH PAPER 26

Budget deficit, domestic credit and money supply

It will be recalled from the introduction that government budget deficit and
money supply are linked via the method of financing the deficit. Money supply
is related to money base and is of the form:

(9) M, = mMB1

where m and MB stand for money multiplier and money base, respectively.
The money base, on the other hand, is the sum total of budget deficit (BD),
change in net foreign assets (NFA) and changes in other assets (OA) of the
Central Bank.

(10)

It is assumed in this equation that government budget deficit is financed by

borrowing from the Central Bank of Nigeria. Therefore, that proportion of the
deficit which is financed by the Central Bank is represented by BD. Budget
deficit is endogenised in the system and it is defined as the difference between
government expenditure and revenue, so that:

(11) BD1 = GE, -

A comment on the inherent logic of the model is imperative. Exchange rate

movements affect the general price level and some components of revenue and
expenditure. For instance, exchange rate depreciation, which raises the interest
charges and capital repayment on external debt in local currency, will
inevitably increase the overall government expenditure. Assuming the
depreciation of the naira significantly engenders price inflation, this has
implications for expenditure. Moreover, the net effect of exchange rate
movements on the budget is a function of the structure of taxes and
expenditure, and their corresponding response.
Assuming revenue lags behind expenditure, the overall effect is fiscal deficit
and this constitutes an important component of the money base. The growth
of the money base expands domestic money supply which, in turn, accelerates
the inflationary process. Note, however, that the expansion in government
revenue, enhanced by depreciation of the naira, will boost public expenditure,
which can also exert upward pressure on the general price level through
several other channels. A complete list of the model variables is provided in
Appendix A.
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 21

Data sources

The model is estimated with the ordinary least squares (OLS) method, utilising
annual data covering the period between 1973 and 1989. Relevant statistics are
collected from various sources. The following variables money supply, real
GDP and its deflator, consumer price index, and official exchange rate — are
obtained from the International Monetary Fund (IMP) International Year Book
1991. Government revenues and expenditures, imports and external debt and
debt service payments are collected from the Central Bank of Nigeria's
publications: 1. Economic Review and Financial Review, various issues; and
2. Annual Report and Statement of Accounts, various years. The parallel
market exchange rate is obtained from World Currency Year Book.
V Model estimation and solution

This section focuses on the empirical estimates of the structural equations

developed in the preceding sections and also contains the model solution.

Model estimation
In line with recent developments in time series modelling, unit root tests on
the relevant economic variables in the model were performed to determine
their time series characteristics. These tests are basically required to ascertain
the number of times a variable has to be differenced to arrive at stationarity'8.
The syllogistic reasoning here is that there is the problem of 'spurious
regression' when non-stationary series are estimated at their levels in a
stochastic equation. It follows, therefore, that knowing the order of integration
of macroeconomic variables helps in a model specification. Following the
general classification, economic variables that are stationary are called 1(0)
series and those that have to be differenced once to obtain stationarity are
called 1(1) series. These are also called random walk. There are those that have
to be differenced more than once to achieve stationarity, however.
Fairly sophisticated methods are available to evaluate the time series
characteristics of macro variables. Some of the currently employed methods are
the Dickey-Fuller (DF) and Augmented Dickey-Fuller and Sargan-
Bhargava Durbin-Waston (SBDW) tests'9. The DF test is a test against the null
hypothesis that there is a unit root of 1(1) of the series and the test utilises the
equation of the form:

= aX, + > hi AX1 + W1,

The test employs the t-statistic on the coefficient of the lagged independent
variable. Thus, the null hypothesis is rejected if the t-value (which is expected
to be negative) is significantly different from the critical value, say at the 5%
level of significance, for a particular sample size. The ADF test is virtually the
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 23

same as the DF test, except the lag length has to be long in order to reflect the
additional dynamics that could not be captured by the DF test and also
possibly to ensure that the error term is white noise. Despite the usefulness of
these tests, it is often suggested that they should not be considered as final in
determining the time series characteristics of economic variables.
Beyond this, researchers also test whether there is a long-run relationship
between a dependent variable and its regressor(s). Indeed, this is the pre-
occupation of cointegration analysis with error correction model (ECM) that
is gradually gaining popularity among economists and econometricians20. The
processes for testing for the existence of cointegrating relationship are twofold,
using the Engle-Granger procedure (see Engle and Granger, 1987). First,
conduct unit root tests on the individual series and if the variables of interest
are 1(1), a static model is estimated for the cointegrating regression.
The second stage is to evaluate the order of integration on the residuals
generated from the static model. The t-statistic of the coefficient of the
regressor using the DF and ADF tests determines whether we should accept
cointegration or not. When the absolute value of the t-statistic is greater than
the critical values, then the existence of cointegration with respect to such
macro variables cannot be rejected. What this suggests is that an error
correction specification would provide a better fit than would be the case
without it.
In appreciation of the above, both the DF and ADF tests were conducted on
all the economic variables required in our structural equations. The results of
these are in Appendix B. Evidently, we could not reject the null hypothesis
that these variables are non-stationary — 1(1). But, for such variables —
external debt (ED) and debt service payment (DS) are all 1(2) series. These
tests revealed that inflation is 1(1) series. This implies that it has to be
differenced once to obtain 1(0). Thus, the second natural logarithmic difference
of the consumer price index (CPI) represents this rate of inflation.
In addition to this, the existence of cointegration between the regressands
and relevant regressors were conducted for other equations. The results of the
long-run solutions of the cointegrating relationships are shown in Table 2. A
statement on each of these equations is useful here. The positive coefficient of
money supply in Equation (a) is consistent with theoretical expectation and
could be interpreted to mean the long-run relationship between price inflation
and money supply. Evidently, in the long-run, revenue collected from taxes in
real terms is a positive function of real imports, as indicated in Equation (b).
It is revealed in Equation (c) that the long-run relationship between other
government revenue and income in real terms is positive. It should be noted
that the theory of tax structure development argues that as economic
development proceeds, more and more revenue tends to be internally generated
24 RESEARCH PAPER 26

(Musgrave, 1969). Other government revenues here are mainly from internal
sources (direct taxes mainly) and hence, the result of this static regression
confirms the suggestion by economic theory. In regard to Equation (d), debt
service payments are proportionally related to the magnitude of external debt
in the long-run. The results of our behavioral equations, utilising the error
correction term computed from these static regression models, are reported
below (see Table 3).

Table 2 Long-run solution of the cointegrating relationship

(a) = 0.61 32 M, - 1.7623

(b) = 0.4384 MTR1 + 6.072 1

(c) = 0.6181 YR1 - 3.7506

(d) = 0.9883 - 2.41 04

Note: Logarithmic specification was attempted in all the equations. The results of
such statistics — R2 and t-value — could not be reported, as they do not follow
the expected distribution since the variables being modelled are non-stationary.

We can begin our interpretation with the results of inflation. Both the
coefficients of growth in money supply and real output have their hypothesized
signs and are statistically significant at the 5% level. Lagged rate of inflation
used as a proxy for expected rate of inflation has a negative sign, a result
inconsistent with theoretical expectation. Following this result, an alternative
definition of expected rate of inflation, as in Olopoenia (1991), was employed
(see footnote 17). Although the coefficient of this is positive, it is not
statistically significant even at the 10% level. This is an indication, perhaps,
that inflation expectation is not an important element in explaining the
inflationary process during the period under study. The coefficient of lagged
exchange rate (official) is highly significant, an indication that depreciation of
the exchange rate exerts upward pressure on inflation — but it takes a
minimum period of one year before this is reflected on price inflation.
Table 3 Estimate of behavioral equations1

1. = -0.0724 + 0.4370 - 1.9400 + 0.1 153 + 0.2215 -

(2.0824) (2.7334) (3.7361) (0.6400) (2.5329) (2.3076)
R2 = 0.7929 F(5,9) = 6.89 (0.0066) DW = 2.12

2. = -0.0790 + 0.6192 -0.3869

(1.3464) (3.7448) (2.1697)
R2=0.6195 F(2,14)= 11.39(0.0012) DW= 2.15

3. = -0.0190 + 0.3869 + 0.4521 - 1.1367ECM11

(0.5240) (1.9705) (2.1003) (4.9024)
R2 = 0.6581 F(3,13) = 8.34(0.0024) DW = 2.16

4. = -1.1073 + 0.6650 + -
(1.2203) (1.9202) (1.5294) (4.6404)
R2 = 0.6249 F(3,12) = 7.22(0.0043) DW = 1.98

5. = -0.0525 + 0.3881 + 3.7833 + 0.1358

(0.9042) (1 .0982) (2.8752) (0.8126)
R2 = 0.6646 F(3,13) = 8.58(0.0021) DW = 2.32

Note: 1.Variables were estimated using natural logarithmic first difference, except ED and DS which were estimated using natural logarithmic
second difference. Variables used in equations (2-3 and 5) are in real terms. All the equations were estimated using OLS and the
econometric software employed is PC-GIVE (see Hendry, 1989).
26 REsEARCH PAPER 26

Acceptance of this result implies acceptance of the fact that the country's
price inflation is caused by both monetary and structural factors. It is pertinent
to note that the long-run relationship between inflation and money supply is
reflected in the coefficient of the ECM variable. The coefficient of the ECM
indicates the speed of adjustment of inflation to money supply in the long-run.
Thus, the feedback effect between inflation and money supply is 0.2.
The specification of revenue from imports as a function of the value of
imports, and an error correction variable, yields a fairly good result. It is
obvious that the volume of imports plays an important role in the
determination of income generated from import duties (see Equation (2)). Also
of interest is the elasticity of import taxes with respect to the volume of
imports that is 0.62, a value less than unity. Clearly, it suggests that revenue
from import duties responded slowly to imports during the estimation period.
One presumed factor that accounted for this is the sharp decline in the
country's imports in 1981—86, a development conditioned largely by
inadequate foreign exchange earnings. Even the role of the import
liberalization of 1974—77 cannot be ignored. However, the error correction
coefficient in this equation indicates that the long run effect of imports on
revenue through the feedback mechanism is relatively high, as it recorded a
value of almost 0.39.
Other government revenues are explained in Equation (3). As indicated
previously, the cointegrating statistics between this variable and real GDP
could not reject the existence of cointegration; and hence, the need for an error
correction specification, to capture the long-run dynamics between these series.
Obviously, other government revenues are, to some extent, dependent on the
growth of the economy.
Next is debt service payment. The performance of this equation is
acceptable. Evidently, the rate of growth of external debt determines the
growth in debt service payments. In turn, the expansion in the stock of external
debt is directly related to the depreciation of the local currency. Although the
coefficient of foreign interest rate on external debt has a positive sign, it is not
statistically significant at the 5% level. It is pertinent to note that the
coefficient of the ECM in this equation captures the short-run impact which
is also tied to the long-run proportionality between external debt and debt
service payment through the feedback mechanism. This is exceedingly low
compared with what was recorded in each of the other equations.
The results of other government expenditures are contained in Equation (5).
Evidence from this equation suggests that total government revenue is not an
important determinant of other government expenditures. Instead, it responds
significantly to the growth of the economy proxied by growth in real GDP.
The value of the adjustment coefficient in this equation is almost 0.14 and it
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NtGERIA 27

is far from being statistically significant at the conventional level. In general,

the descriptive statistics (R2, F test and DW statistic) of this model are quite
acceptable. It might be necessary to elaborate on the coefficient of income,
however. This is because the elasticity of government expenditures with
respect to income is greater than unity (the value is 3.78). It is possibly a
reflection of Wagner's law of ever-increasing state activity which states that
as the economy grows, government outlay tends to grow, at an even faster rate
than GDP (Egwaikhide, 1984).
Overall, the model fits the data reasonably well. The results of all the
equations confirm theoretical expectation, though, not all coefficients are
statistically significant at the 5% level. Interestingly, however, the model is
very dynamic through the use of lags, natural logarithmic difference and an
error correction model. Moreover, the network of feedback mechanisms also
makes the model more powerful than any of its behavioral equations. Yet, until
the complete model is solved, its internal logic and consistency may not easily
be appreciated.

Model solution

The system of equation was solved using Time Series Processor (TSP)
econometric software (version 4.0) developed by Hall (1983). This software
has two methods for solving a macroeconometric model of this nature. These
are the Gauss-Seidel (see Ortega and Rheinboldt, 1970) and Fletcher-Powell
(Fletcher and Powell, 1963) algorithms. The Gauss-Seidel iterative technique
is employed here and it is quite simple, as it solves the equations sequentially,
with each endogenous variable evaluated in turn.
A dynamic simulation of the model was conducted over the period 1975 to
1989. The idea is to assess the extent to which the model can be relied upon
for short-run policy forecasting. The argument has been posited in econometric
literature that even when all the individual equations of a large macro model
fit the data reasonably well, using various statistical tests (e.g. DW-statistic, t-
statistic, standard error of the regression, F test, R2, etc), there is no guarantee
that the model, when simulated, will be able to track the historical data very
closely. Thus, this exercise is important as it allows us to gauge the internal
consistency of the macro model developed.
We only present one important criterion for macroeconometric model
evaluation often discussed in the literature (see, for example, Pindyck and
Rubinfeld, 1981). This is the Theil' s inequality and its decomposition2' and the
results are contained in Table 4.
28 RESEARCH PAPER 26

Table 4 Theil's inequality coefficients and their decomposition for some key
variables

Variable Theirs Bias Variance Co-variance

inequality proportion proportion proportion

P 0.03914 0.02660 0.2459 0.72481

GOR 0.06458 0.03357 0.02999 0.93644
2DS 0.30282 0.14990 0.24855 0.60155
ORR 0.06161 0.00519 0.14433 0.85047
MRR 0.07818 0.01014 0.09282 0.89597
M 0.11252 0.02360 0.08311 0.89329
GE 0.10306 0.04162 0.12702 0.83136
GR 0.09207 0.06480 0.05347 0.88173

In general, the values of the coefficients of the computed endogenous variables

are small with the largest being 0.30 (registered by debt service payments).
One plausible interpretation derivable from the smallness of the mcii's
inequality coefficients is that there is an absence of any systematic bias and,
consequently, the model may not need any major revision for the current
purpose of it development.
Overall, it is evident that at least 88% of the variations in the variables
presented in Table 4 were predicted by the model, except debt service
payments. The inability to explain more than 70% of the variations in this
variable may be due to its volatility, particularly in recent years.
It may be necessary to comment on the decomposition of the Theil's
inequality coefficient. This decomposition, as demonstrated in Table 4, is
nearly optimal, with a large part of the errors skewed in favour of the co-
variance proportion for the variables, a result consistent with the requirements
of a 'good' model.
VI Policy simulation results

Counterfactual simulation analysis addresses a number of questions of the

'what if' variety. Such questions are usually answered under two approaches:
ex-post and ex-ante impact simulations (see Challen and Hagger, 1983). Our
preoccupation here, however, is on ex-post simulation of the effects of
exchange rate depreciation on specific economic variables of interest. To
conduct this, we have to assume a particular percentage level of exchange rate
depreciation. This task has been simplified by the exchange rate policy adopted
on 5 March 1992, by the Central Bank of Nigeria (CBN). It is important to
note that this policy attempts to close the gap between the official exchange
rate and the parallel market rate, by devaluing the former. Thus, the
assumption is made: supposing the Central Bank had allowed the exchange rate
to float from early 1984, soon after the overthrow of the civilian government
in December 1983, and in the process the naira depreciated, what would have
happened to the relevant variables?
Available statistics indicate that Nigeria's official exchange rate in 1984 was
NO.7642:US$l.O and the parallel market rate recorded about N3.226:US$1.O.
It follows, therefore, that the Central Bank needed to devalue the official
exchange rate by about 76% (then) to bridge the gap between these rates; and
so we adopt this percentage depreciation of the local currency in the simulation
experiment during the six-year period 1984—89. A simple method is employed
to present the results. First, the mean annual growth rates of the variables
generated by the model solution are computed — this is called base line
(control run) solution. Second, the official exchange rate was assumed to
depreciate by 76% from 1984 (a once-and-for-all exercise) and the result of
this is labelled as disturbed (shocked) solution22
The results of this experiment are shown in Table 5. It is obvious that this
hypothetical situation will raise average annual growth rates of the key
variables in the model, except real growth of other revenues which declined
in the reference period. Specifically, price inflation would increase by an
annual average of about 6% over the control solution and domestic money
supply by almost 2.0%. The results of the model also specify that the average
revenue (GRR) in the base line solution were 17.90 and 13.07%, respectively.
30 RESEARCH PAPER 26

The difference between these figures was 4.83%. By the assumed percentage
depreciation of the naira, real government expenditure seemed to have grown
faster than state revenues, as the difference between these variables in the
simulation experiment recorded 8.11% (8.11 > 4.83).

Table 5 Effects of exchange rate depreciation on budget and inflation: a

simulation experiment (mean annual growth rate, 1984-89)

Variable Base line solution Disturbed solution

P 23.06 29.13
M 13.70 15.62
GOR 12.35 14.25
DS 56.14 65.28
MRR 11.78 13.40
OAR 2.58 1.13
GER 17.90 24.03
GRR 13.07 15.92

Note: Natural logarithmic first difference (multiplied by 100%) was employed to

generate the growth rate.

This is possibly an indication of enlarged budget deficit (as it already exists)

engendered by exchange rate reforms. On the basis of the model result, the
increase in total government expenditure appears to have been fostered by the
growth in debt service payment, with depreciation of the exchange rate raising
the stock of external debt (a significant explanatory variable of debt service
payments) in local currency enormously. Perhaps it is through the monetization
of the increased revenues from the exchange rate liberalization that the
expansion in money base became enhanced, with rapid growth in money
supply and inflation as some of the derivatives. This is not surprising, as we
found earlier from the results of the price equation that domestic money supply
has both short- and long-run effects on inflation.
Following the depreciation of the naira, net foreign assets in local currency
could expand. A direct result of this is increased money base and money
supply which in turn feeds into price inflation. This could create budget deficit
if expenditure responds to inflation faster than revenue. In this sense, the link
between changes in net foreign assets and budget deficit is indirect23.
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 31

The policy lesson for the above is easily discerned. It is clear from the
simulation experiment that exchange rate depreciation affects both sides of the
budget significantly, with total expenditure responding faster than total
revenues. This development could generate budget deficit if not properly
managed. This is particularly true when it is realised that exchange rate reform
of this nature could be undertaken primarily to mobilise more revenues for
government. If that is the case, the monetization of such earnings is likely to
boost aggregate demand and possibly the money base and, consequently, fuel
price inflation. We should also learn from the results that depreciation of the
exchange rate has a tendency to expand debt service payments, since it raises
the size of external debt remarkably. This is reflected significantly in
government expenditure. Of course, this development can have adverse effects
on investment which constitutes the motor of economic expansion — at least,
in the short-run since increased debt service payments tends to crowd-out other
expenditure categories, particularly capital outlay.
VII Conclusion

This study examined the quantitative effects of exchange rate depreciation on

inflation, government revenues and expenditures, and money supply in Nigeria.
Our objective was achieved through the use of a macroeconometric model that
captures the key aspects of the linkages between the above variables. In the
empirical estimates of the structural equations, we drew on recent
developments on cointegration and error correction model which are rapidly
gaining popularity among economist and econometricians. In addition to this,
we relied on the use of trends in the relationship between inflation and other
economic variables.
So far, our findings highlight several points of importance. Evidence from
trend analysis suggests that domestic money supply, real output, the shadow
price of exchange rate — the parallel market exchange rate — and, more
recently, official exchange rate, cannot be ignored in evaluating the proximate
causes of inflation in Nigeria. With particular reference to exchange rate,
graphical representation reveals that the parallel market exchange rate appears
to correlate with inflation more when compared with the official rate.
Statistical estimates of the structural equations are also quite revealing. The
results of inflation equation demonstrate that official exchange rate is a
significant determinant of price inflation, with a lag period of one year.
Inadequate output and monetary expansion also featured prominently in this
equation. But the coefficient of the expected rate of inflation, though positive,
is not statistically significant even at the 10% level. Imports, influenced by
exchange rate, significantly explain revenue from import taxes. The magnitude
of external debt in local currency, which also depends on the exchange rate,
has a tendency to increase debt service payments and, therefore, total
expenditures, through a feedback mechanism using an error-correction term.
All this has implications for budget deficit and growth in domestic money
supply and thus inflation.
After testing for the internal consistency of the complete model, it was
applied to evaluate the impact of exchange rate depreciation on such variables
as money supply, revenues and expenditures, and inflation. Evidence from the
model results, in which we assumed a floating exchange rate that eventually
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 33

lead to a depreciation of the naira, indicate that exchange rate depreciation can
be inflationary. This works via its direct impact on inflation, and through
budgetary and monetary effects. On average, the depreciation of the naira (by
about 76%) seems to raise the growth of total expenditure more than total
revenue.
This is an indication of an enlarged budget deficit (where budget deficit
already exists, as in the case of Nigeria) or it can generate budget deficit over
time. The monetization of the deficit expands the money base, resulting in
inevitable growth in money supply.
This is contrary to the result obtained from trend analysis which indicates
that inflation has already adjusted to the parallel market rate; and consequently
devaluing the naira may not necessarily be inflationary. One important policy
lesson is therefore obvious from the simulation experiment: namely that
exchange rate depreciation significantly affects both the revenue and
expenditure sides of the budget in Nigeria and it could enlarge the existing
budget deficit if not properly managed. A restrictive monetary policy may be
implemented to complement the exchange rate policy adopted.
Two important limitations of this research should be mentioned. First, is
that the model employed is highly aggregative, particularly the revenue and
expenditure components. In the disaggregated model, net foreign assets could
be endogenized. Moreover, the effects of exchange rate reform on the
productive base of the economy which, in turn, influences government revenue
and inflation are not considered in the current research. To that extent, our
findings, particularly from the model results, are more suggestive than
definitive. A more detailed modelling of the interactions between exchange
rate, budget, inflation and production is therefore required.
The second weakness has to do with the cointegration and error correction
technique explored in the estimation of the structural equations. This
methodology generally requires a large sample size to draw solid inferences
for policy simulations. The sample size of the current estimate does not permit
us to have great confidence in the results obtained.
Appendix A
Definition of the estimated model variables

Endogenous

BD Budget deficit: defined as that part which is financed

by the Central Bank of Nigeria.
DS Debt service payments (annual).
GER Total government revenue (real).
GOR Other government revenue (real).
GRR Total government revenue (real).
M2 Money supply, broadly defined.
MB Money base (BD + NFA + OA).
MRR Revenue from import duties (real).
MTR Merchandise imports (real).
ORR Other government revenue (real).
P Inflation.
PR Revenue from crude oil in local currency (nominal).
YR Real income.

Exogenous

e Nominal exchange rate (official).

ED External debt in local currency.
NFA Net foreign assets.
QA Other assets of the Central Bank.
PRF Revenue from crude oil in US dollar (nominal)
if Foreign interest rate on external debt.
Appendix B
Unit root tests

Variable Test4 a SE1 t-value

Y DF -0.0039 0.0204 -0.1912

ADF(3) -0.0062 0.0232 -0.2672
YR DF -0.0690 0.0470 -1 .4681
ADF(3) -0.0750 0.0572 -1.3112
M2 DF -0.0066 0.0185 -0.3568
ADF(2) -0.0080 0.0199 -0.4020
MTR DF -0.0128 0.0459 -0.2789
ADF(3) -0.0175 0.0529 -0.3308
ORR DF -0.0099 0.0296 -0.3345
ADF(2) -0.0010 0.0266 -0.0376
GOR DF -0.0239 0.0381 -0.6273
ADF(2) -0.0358 0.0404 -0.8738
GRR DF -0.0037 0.0350 -0.1057
ADF(2) -0.0153 0.0374 -0.3556
ED DF -0.0376 0.0477 -0.7883
ADF(2) -0.0228 0.0489 -0.4663
DS DF -0.0380 0.0793 -0.4792
ADF(2) -0.0245 0.0582 -0.4210
P DF -0.6743 0.2439
ADF(4) -0.3891 0.3437 -1.1321
MRR DF -0.0048 0.0572 -0.0839
ADF(2) -0.0002 0.0035 -0.0031

Notes: 1. SE = Standard error

2. Critical values at 1 and 5% level of significance are -1.95 and -2.66 for
sample of 25, respectively.
3. The null hypothesis is rejected at the 5% level of signiticance, an indication
that inflation is 1(0). But the low power of the ADF test reveals that inflation
is 1(1). This is more reliable, as it is more powerful than the DF test.
4. DF = Dickey-Fuller test. ADF = Augmented Dickey-Fuller test (the number
in parenthesis indicates the lag length).
Appendix C
Inflation rate and discount rate

40 — Inflation rate
Discount rate

30

C
20
0
U-

n
70 72 74 76 7880 82 84 86 88
Notes

1. For a detailed discussion of the objectives and policy instruments of

SAP, see Federal Republic of Nigeria (1986).

2. The key petroleum tax reforms implemented by the Federal

Government are contained in Iwayemi (1981).

3. On this, see Oyejide (1985a).

4. The article by Schatz (1984) presents an excellent discussion on

Nigeria's inert economy without oil.

5. A historical sketch of Nigeria's exchange rate policy are in Oyejide

(1985b), Ogun (1990), and Ajayi (1988).

6. An excellent discussion of this development is presented by Pinto

(1987). The theoretical underpinnings of his submission are in Morgan
(1979) and Olopoenia (1986), among others.

7. Pinto (1989: 333) noted earlier that inflation reflects movements in

Nigeria's parallel market exchange rate, on the basis of some
calculations.

8. Fitzpatrick and Nixson (1976) present a detailed review of the

monetarist-structuralist debate on inflation. A widely-cited empirical
study of structural causes of inflation is that of Argy (1970). On the
origin of structuralism, see Arndt (1985).

9. On some of the papers, see Adejugbe (1974), Adeyokunu and Ladipo

(1974), Ajayi and Teriba (1974), Ojo (1974), Owosekun and Odama
(1974).
38 RESEARCH PAPER 26

10. Four alternative measures of deficits were employed. These are: 1.

change in the narrow definition of money (Ml); 2. domestic credit
creation; 3. internal credit monetisation defined as increase in money
supply plus the reduction in net gold and foreign reserves; and 4. draw-
down of foreign reserves. For elaboration, see Oyejide (1972). A recent
study by Ariyo and Raheem (1990) utilizes a broad definition of deficit
financing and economic development in the country. On the sources of
financing budget deficit in Nigeria, see Mbanefoh (1982).

11. See Aghevli and Khan (1978) for the internal logic of the structural
equations developed.

12. On this, see Granger (1969).

13. Only Granger causality tests were, however, performed. Perhaps, it is

important to emphasize that the techniques of Vector Autoregressive
(VAR) analysis of bivariate and trivariate tests were employed in this
research.

14. An interesting study on the link between exchange rate and the general
price level in Botswana has been conducted by Leith (1991).

15. Sophisticated mark-up models demonstrating the impact of exchange

rate on prices are available. For example, see Ajakaiye (1990).

16. Empirical testing of this was attempted by Aghevli and Khan (1978).
The works by Chhibber et a!. (1989), and Chhibber and Shafik (1990a)
also provide some useful estimates with respect to Zimbabwe and
Ghana, respectively.

17. The expected rate of inflation is along the line employed in a recent
study by Olopoenia (1991). In this formulation, the expected rate of
inflation is related to its past values, so that:

pe is the expected rate of inflation and p is actual

inflation.
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 39

18. A simplified presentation of stationary and non-stationary processes of

time series variables and their relevance in econometric modelling can
be found in Yoshida (1990, pp. 20-24).

19. On the first two tests, see Yoshida (1990, pp. 24-26). For the Sargan-
Bhargava Durbin-Waston test (SBDW), see Sargan and Bhargava
(1983), and Engle and Granger (1987).

20. Recent studies on the use of cointegration and an error correction

model are available. Among them are Hendry (1986), Domowitz and
Elbadawi (1987), Yoshida (1990), Chhibber and Shafik (1990b),
Hendry and Ericsson (1991), and Adam (1991).

21. The inequality was introduced by Theil (1961, pp. 30-37) and is of the
form:

J ni-i
I i-i
fl
+ I n

where P = predicted value

A = actual value;
n = sample size.

The numerator is the root-mean square error (RMSE) and the

denominator represents weights which constrained the value of T to
The closer the value of T to zero, the better are the results of
the dynamic historical simulation. On the derivation of the
decomposition, see Theil (1961; 1966).

22. Several variants of this approach have been used by various authors.
For example, see Salvatore (1983), and Nziramasanga and Obidegwu
(198 1).

23. In the simulation experiment, we assume neutrality of other government

policies. For this reason, we did not critically assess the impact of net
40 REsEARCH PAPER 26

foreign assets on budget deficit. Even in the development of the model,

the role played by net foreign assets was not central.
References

Adam, C.S., 1991, 'Financial innovation and the demand for M3 hi the UK,
1975—86', Oxford Bulletin of Economics and Statistics, Vol. 53, No. 4, pp.
40 1-24.
Adejugbe, M.O.A., 1974, 'The mechanics and dynamics of inflation in
Nigeria', in Onitiri, H.M.A. and K. Awosika (eds) Inflation in Nigeria.
Proceedings of a National Conference, Nigerian Institute of Social and
Economic Research, NISER, Ibadan, pp.46-70.
Adeyokunu, J. and 0.0. Ladipo, 1974, 'The causes of inflation in Nigeria: A
quantitative assessment', in Onitiri, H.M.A. and K. Awosika (eds) Inflation
in Nigeria. Proceedings of a National Conference, NISER, Ibadan, pp. 141-
55.
Aghevli, B. and M.S. Khan, 1978, 'Government deficits and the inflationary
process in developing countries', IMF Staff Papers, Vol. 25, pp. 383-416.
Ajakaiye, D.0., 1990, Public Enterprises Policies in Nigeria. A
Macroeconomic Impact Analysis, NISER, Ibadan.
Ajayi, S. Ibi, 1988, 'Issues of overvaluation and exchange rate adjustment in
Nigeria', prepared for Economic Development Institute of the World Bank,
Washington DC.
Ajayi, S. Ibi and K. Awosika, 1980, 'Inflation in Nigeria: Domestic or
Imported', commissioned paper for Productivity Prices and Incomes Board,
January.
Ajayi, S. Ibi and 0. Teriba, 1974, 'The inflationary process in Nigeria
1960—1972: Evidence from quarterly series', in Onitiri, H.M.A. and K.
Awosika (eds) Inflation in Nigeria. Proceedings from a National
Conference, NISER, Ibadan, pp. 112-25.
Argy, V., 1970, 'Structural inflation in developing countries', Oxford
Economic Papers, Vol 22, No. 1, pp. 73-85.
Ariyo, A. and M.I. Raheem, 1990, 'Deficit financing and economic
development: Empirical perspectives from Nigeria', final research report
presented at the African Economic Research Consortium Workshop,
Abidjan, December 9-14.
42 RESEARCH PAPER 26

Arndt, H.W., 1985, 'The origins of structuralism', World Development, No.

2, pp. 151-9.
Balassa, B. (1964), 'The purchasing power parity doctrine: A reappraisal', The
Journal of Political Economy, December, pp. 584-96.
Canetti, E. and J. Greene, 1991, 'Monetary growth and exchange rate
depreciation as causes of inflation in African countries: An empirical
analysis', mimeo, IMF, Washington DC.
Challen, D.W. and A.J. Hagger, 1983, Macroeconomic systems: Construction,
validation and applications, Macmillan, London.
Chhibber, A., 1991, 'Africa's rising inflation: causes, consequences and cures',
World Bank Working Papers, WPS 577.
Chhibber, A. and N. Shafik, 1990a, 'Exchange reform, parallel markets and
inflation in Africa: The case of Ghana', World Bank Working Papers, WPS
427, May.
Chhibber, A. and N. Shafik, 1990b, 'Does devaluation hurt private
investment? The Indonesian case', World Bank Working Papers, WPS 418,
May.
Chhibber, A., J. Cottani, R. Firuzabadi, and M. Walton, 1989, 'Inflation, price
controls and fiscal adjustment in Zimbabwe', World Bank Working Papers,
WPS 192, April.
Corden, W.M., 1989, 'Macroeconomic adjustment in developing countries',
World Bank Research Observer, Vol. 4, No. 1, pp. 5 1-64.
Domowitz, I. and I. Elbadawi, 1987, 'An error-correction approach to the
money demand: The case of Sudan', Journal of Economic Development,
Vol. 26, pp. 257-75.
Dornbush, R., 1987, 'Purchasing power parity', in Eatwell, J., M. Milgate and
P. Newman (eds) 1987), The New Paigrave: A Dictionary of Economics,
vol. 3 (London: Macmillan Press).
Edwards, S., 1989,'Real exchange rates, devaluation, and adjustment:
Exchange rate policy in developing countries', Cambridge, Mass.: MIT
Press.
Egwaikhide, Festus 0., 1984, 'Patterns of government revenue and expenditure
in Nigeria: A historical survey', an unpublished M.Sc. project, Department
of Economics, University of Ibadan, Nigeria.
Elbadawi, l.A., 1990, 'Inflationary process, stabilisation and the role of public
expenditure in Uganda', mimeo, July, World Bank, Washington DC.
Engle, R.F. and C.W.I. Granger, 1987, 'Co-integration and error correction;
Representation, estimation and testing', Econometrica, Vol. 55, No. 2, pp.
251-76.
Federal Republic of Nigeria, 1986, Structural Adjustment Programme, July
1986—June 1988, Government Printer, Lagos.
EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 43

Fitzpatrick, C.H. and F.!. Nixson, 1976, 'The origins of inflation in less
developed countries: A selective review' in Parking, M. and G. Zis (eds)
Inflation in Open Economies, pp. 126-74, Manchester University Press,
Manchester.
Fletcher, R. and M.F.D. Powell, 1963, 'A rapidly converging descent method
for minimization', Computer Journal, Vol. 6, pp. 163-8.
Granger, C.W.J., 1969, 'Investigating causal relations by econometric models
and cross-spectral methods', Econometrica, Vol. 37, pp. 424-38.
Hall, B.H., 1983, Times Series Processor Version 4.0, User's Manual, TSP
International, Stanford, California, July.
Hendry, D.F., 1986, 'Econometrics modelling with cointegrated variables: An
Overview', Oxford Bulletin of Economics and Statistics, Vol. 48, No. 3, pp.
201-12.
Hendry, D.F., 1989, PC-GIVE: An Interactive Econometric Modelling System,
Institute of Economics and Statistics, Oxford.
Hendry, D.F. and N.R. Ericsson, 1991, 'An econometric analysis of UK
money demand in Monetaiy trends in the United States and the United
Kingdom by Milton Friedman and Anna J. Schwartz', American Economic
Review, Vol. 81, No. 1, pp. 8-38.
Iwayemi, A., 1981, 'Planning and performance in the Nigerian Petroleum
Industry', proceedings of the Annual Conference of the Nigerian Economic
Society, Ilorin.
Leith, J.C., 1991, 'Exchange rate and the price level in a small open economy:
Botswana', Journal of Policy Modelling, Vol. 13, No. 13, No. 2, pp. 309-
315.
London, A., 1989, 'Money, inflation and adjustment policy in Africa: Some
further evidence', African Development Review, Vol. 1, No. pp. 87-111.
Mbanefoh, G.F., 1982, 'Sources and consequences of deficit financing in the
Nigerian economy', paper presented at the Workshop on Deficit Financing,
University of Ibadan, Nigeria.
Morgan, D.R., 1979, 'Fiscal policy in oil exporting countries', IMF Staff
Papers, Vol. 26, No. 1, pp. 55-86.
Musgrave, R.A., 1969, Fiscal Systems, pp. 125-136, Yale University Press
New Haven.
Nziramasanga, M.T. and C. Obidegwu, 1981, 'Primary commodity price
fluctuations and developing countries: An econometric model of copper in
Zambia', Journal of Development Economics, Vol. 9, No. 1, pp. 89-119.
Ogun, O.D., 1990, 'The monetary effects of an export boom: Nigeria, 1950-
85', an unpublished doctoral thesis submitted to the Faculty of the Social
Sciences, University of Ibadan, Nigeria.
_____

44 RESEARCH PAPER 26

Ojo, 0., 1974, 'Causes of inflation in Nigeria — Comments on the papers' in

Onitiri, H.M.A. and K. Awosika (eds) Inflation in Nigeria, Proceedings of
a National Conference, NISER, Ibadan, pp. 156-64.
Olopoenia, R.A., 1986, 'Some observations on the macroeconomic
implications of the budget in an oil-exporting developing economy', The
Nigerian Journal of Economic and Social Studies, Vol. 28, No. 3, pp. 285-
96.
1991, 'Fiscal response to oil wealth and balance of payments
performance in Nigeria, 1970—89', draft final report presented at the
Economic Research Workshop of the African Economic Research
Consortium (AERC), Nairobi, Kenya, December 7-11.
Ortega, J.M. and W.C. Rheinboldt, 1970, Iterative Solution of Non-Linear
Equations of Several Variables, Academic Press, New York.
Owosekun, A. and J.S. Odama, 1974, 'The causes of inflation — Some
empirical evidence' in Onitiri, H.M.A. and K. Awosika (eds) Inflation in
Nigeria, Proceedings of a National Conference, Nigerian Institute of Social
and Economic Research (NISER), Ibadan, pp. 98-ill.
Oyejide, T.A., 1972, 'Deficit financing, inflation and capital formation: The
analysis of the Nigerian experience, 1957—70', The Nigerian Journal of
Economic and Social Studies, Vol. 14, pp. 22-43.
1985a, 'The debt and development dilemma', an inaugural lecture,
University of Ibadan, Nigeria, April.
1985b, 'Exchange rate policy for Nigeria: Some options and their
consequences', proceedings of the Nigerian Economic Society, pp. 17-
32.
Pindyck, R.S. and D.L. Rubinfeld, 1981, Economic Models and Economic
Forecasts, McGraw Hill, Singapore.
Pinto, B., 1987, 'Nigeria during and after the oil boom: A policy comparison
with Indonesia', World Bank Economic Review, Vol. 1, No. 3, pp. 419-445.
1989, 'Black market premia, exchange rate unification, and inflation in
Sub-Sahara Africa', World Bank Economic Review, Vol. 3, No. 3, pp.
321-38.
Salvatore, D., 1983, 'A simultaneous equation model of trade and
development: Dynamic policy simulations', Kyklos, Vol. 36, pp. 67-89.
Sargan, J.D. and A. Bhargava, 1983, 'Testing residuals from least squares
regression for being generated by the Gaussian Random Walk',
Econometrica, Vol. 50, pp. 153-67.
Schatz, S.P., 1984, 'Pirate capitalism and the inert economy of Nigeria',
Journal of Modern African Studies, No. 1, pp. 45-57.
_____

EXCHANGE RATE DEPRECIATION, BUDGET DEFICIT AND INFLATION IN NIGERIA 45

Sowa, N.K. and J.K. Kwakye, 1991, 'Inflationary trends and control in
Ghana', final research report presented at the African Economic Research
Consortium Workshop, Nairobi, Kenya, May.
Tanzi, V., 1977, 'Inflation, lags in collection, and the real value of tax
revenue', IMF Staff Papers, Vol. 24, pp. 154-67.
Tegene, A., 1989, 'The monetarist explanation of inflation: The experience of
six African countries', Journal of Economic Studies, Vol. 16, pp. 5-18.
Theil, H., 1961, Economic Forecasts and Policy, Amsterdam, North-Holland.
1966, Applied Economic Forecasting, Amsterdam, North-Holland.
World Currency Year Book, International Analysis mc, Brooklyn, New
York (formerly Pick's Currency Year Book before 1984).
Yoshida, Tomoo, 1990, 'On the stability of the Japanese money demand
function: Estimation results using the error correction model', Bank of
Japan Monetary and Economic Studies, Vol. 8, No. 1, pp. 1-47.
 IDRC CR01

111111 11111 11111 11111 11111 liii liii

233685
 AFRICAN ECONOMIC RESEARCH CONSORTIUM

AERC
CREA The principal objective of the African Economic Research
Consortium (AERC), established in August 1988, is to
strengthen local capacity for conducting independent, rigorous
inquiry into problems pertinent to the management of
economies in Sub-Saharan Africa.
In response to special needs of the region, AERC has
P.O. BOX 62882 adopted a flexible approach to improve the technical skills of
NAIROBI, KENYA local researchers, allow for regional determination of research
priorities, strengthen national institutions concerned with
economic policy research, and facilitate closer ties between
researchers and policy makers.
Since its establishment, AERC has been supported by
TELEPHONE (254-2) 228057 private foundations, bilateral aid agencies and international
225234 215898 212359
organizations.
332438 225087
SPECIAL PAPERS contain the findings of commissioned
studies in furtherance of AERC's programmes for research,
training and capacity building.
RESEARCH PAPERS contain the edited and externally
TELEX 22480 reviewed results of research financed by the AERC.
It is AERC's policy that authors of Special and Research
Papers are free to use material contained therein in other
publications. Views expressed in the Special and Research
Papers are those of the authors alone and should not be
FAX (254-2) 219308
attributed to the AERC's sponsoring Members, Advisory
Committee, or Secretariat.
Further information concerning the AERC and additional
copies of Special and Research Papers can be obtained by
writing to: Afi-ican Economic Research Consortium, P.O. Box
62882, Nairobi, Kenya.

ISBN 9966-900-13-6