Beruflich Dokumente
Kultur Dokumente
A
A
(
A
A
=
A
A
k
k
k
k
k
k
k
k
k x
k
B
Y
x
Y
B
T
B
x
B
T
T
Y
x
Y
T
= ....................................................... (9)
Equation 5, decomposes any tax system as the product of elasticity of tax-to-base and of
base to-income.
One potential hindrance to the use of this method is the non-availability of required data.
This is the problem that compelled Omoruyi (1983) into adopting an aggregative measure
of tax buoyancy for Nigeria. These problems gave rise to a consideration of other
techniques suggested by Singer (1988), usually referred to as the Dummy Variable
Technique (DVT). This technique introduces a dummy variable into Equation 2 for each
year in which there was an exogenous tax policy change. The resulted to a model as:
log T
R
= a
0
+ a
1
log Y + E
i
Dum
i
+ e
r
...............................................................(10)
where,
Dum
i
(i = 1, 2) takes a value of 1 for each year in which there is an exogenous change in
tax policy and a value of zero (0) otherwise. According to Singer, a potential major
problem with this approach relates to inadequate number of observations when
exogenous tax policy changes are too frequent.
2.5.4 Tax Stability
Haughton (1998) noted that the revenue from different taxes varied from year to
year. He opined that taxes whose revenue is relatively stable or whose revenue is
negatively correlated with the revenue from other taxes are likely to be particularly
61
helpful in giving stability to the over stream of revenue. The stability of revenue is
desirable at least from the governments perspective in that it makes it easier to put
together plausible spending and borrowing plans for the year head.
Haughton, stated that a simple measure of the stability of tax revenue is the
coefficient of variation (CV). The coefficient of variation is defined as the standard
deviation of the tax revenue as a fraction of GDP divided by its mean. It is
mathematically expressed as:
.. (11)
He further noted that the coefficient of variation may be calculated for tax revenue as a
whole or for individual sources of revenue and that, the measure is most useful when
comparing across countries.
Mean
Deviation Standard
CV =
62
CHAPTER THREE
RESEARCH METHODOLOGY
3.0 Introduction
This Chapter presents the methodology used in analysing the data collected as well as the
description of the nature and types of data. In addition, it presents the models to be used
and the justification of the variables used.
3.1 Materials and Methods
The data used for this study were collected from the secondary sources. As a result, the
research gathered materials from the Central Bank of Nigeria (CBN), the Federal Offices
of Statistics (FOS), and National Bureau of Statistics (NBS) for the period under
consideration. The researcher employed econometrics technique to establish the
regression equations and the E-view Econometric Package (Augmented Dunkey Fulley
and the Philips Perron) to establish the stationarity of the data with the objective of using
co-integration technique rather than only the simple Ordinary Least Square (OLS)
method.
In carrying out this study, data on the following variables were collected from the
publications of the Central Bank of Nigeria (Various issues), The Federal Office of
Statistics and the National Bureaux of Statistics. The variables are:
GTR = Total Tax Revenue
GDP = Gross Domestic Product
NGDP = Non-Oil Gross Domestic Product
NOR = Non-oil Total Revenue
63
CEXD = Custom and Excise Duties
PPT = Petroleum Profit Tax
TOR = Total Oil Revenue
CIT = Company Income Tax
TEXD = Total Export Duties
The researcher has adopted the regression analysis in analysing the productivity of the
Nigeria Tax System. In view of that, the multiple regression technique was employed as
numerous variables are involved. But considering the environment under which research
is being conducted, it was desirable to isolate the impact of some variables on the
productivity of the Nigerian tax system. As a result, we first performed a regression
analysis of the Gross Domestic Product (GDP) and the yield of Aggregate Tax-Based
Revenue (ATBR), as well as by each tax source, over the 1960 2010 periods. This
according to Ariyo (1997), provides an index of the buoyancy of the tax system as a
whole and for each tax source. Secondly, Ariyo further opined that, the oil boom would
have affected the productivity of the tax system in view of the behavioural explanation
discussed in the context of the Dutch Disease Syndrome.
Hence, a priori, the oil boom is expected to affect negatively the yield from non-oil tax
sources. The extent of this effect, however, depends upon the perception of policy makers
regarding the permanence or otherwise of the oil wealth. Third, the Federal Government
of Nigeria (FGN) commenced the implementation of a SAP in 1986 that amounted to a
significant structural change in the macroeconomic management framework for the
country. One of the core objectives of the SAP is to enhance the degree of self-reliance
64
within the economy. Of equal importance is the need to diversify the countrys revenue
base in order to minimize the extent of dependence on oil as the major source of revenue.
All these have potential implications for the yield of non-oil tax revenue sources. For
example, one major consequence of SAP is the rekindled interest in export of cash crops
such as cocoa. Ordinarily, this should have resulted in a significant upsurge in revenue
from export duties, but as part of the reform, the Federal Government of Nigeria scrapped
export duties as an element of the package of incentives meant to promote exports. There
were significant downward revisions in tax rates and import tariffs as well. The corporate
tax rate was reduced from 45% to 40% in 1987 in order to encourage reinvestment
activities by existing organizations and to encourage new investments.
Similarly, import duties on certain categories of imports were reviewed. Among these
was the elimination of duties on trucks and commercial vehicles to ease the transportation
problem in the country. Also, duty exemptions were granted on items required on some
public sector projects. Generous tariff concessions were also allowed on machinery and
raw materials that could not be sourced locally, at least not in the short run.
Several policies have implications for the yields of specific tax sources were also initiated
to mitigate the negative effects of SAP on the populace. For example, tax reliefs and
allowances were granted on personal income tax to enhance the real income of workers,
although this particular tax source is not covered in this study for reasons stated earlier.
The introduction of SAP generated several changes in tax-related policies, so that any
65
growth in GDP during this period might not necessarily translate into higher tax yield.
The determination of the net effect is therefore an empirical question.
Therefore, the analysis was arranged to highlight developments during each of the
following periods.
1970 2010, for an overall trend analysis.
1974 1985 represents Oil boom period.
1986 2010, Sap and Post-Sap period.
In addition, we disaggregated the analysis as much as possible for a number of reasons.
For example, we are aware of the dominance of the oil sector on total government
revenue, and its inclusion in the GDP may distort developments in the non-oil sector.
Consequently, we also regressed non-oil government revenue against non-oil GDP. This
modification based on non-oil GDP was extended to excise duties since we have no
authentic data on the relevant tax base such as total value of production or manufacturing
activities. The same applies to company income tax, because we had no reliable data on
corporate profits. Since this constraint does not apply to imports and sales of petroleum
oil, we regressed import duties against imports, while oil revenue was regressed against
reported oil sales.
3.2 Method of Data Analysis
The data that used for this study were analysed starting with the Ordinary Least Square
(OLS) method after checking for Unit Root Test, with the Augmented Ducker Fuller
(ADF) and the Philip Peron (PP).
66
3.2.1 Model Specification
Based on our data availability, the following multiple regression equation is specified for
analysis to determine the productivity of the Nigeria Tax system. The OLS method was
used to analyze the various equations.
GTR = f( GDP, NGDP, TOS, NOR, CEXD, PPT, CIT, TEXP ) ............................. (12)
The regression equation for our trend analysis for the period under consideration is as
shown below:
GTR = a
0
+ a
1
GDP + a
2
TOR + a
3
NOR + a
4
CEXD + a
5
PPT + a
6
CIT+a
7
TEXP..........(13)
However, in order to assess the revenue elasticity for individual variables used in this
study, we also specified one variable model, which was also estimated by OLS method.
Based on that, the following equations were formulated and analyzed.
log GTR = a
0
+ a
1
log GDP (i) ............................................... (14)
log NOR = b
0
+ b
1
log NGDP (ii)
log CEXD = c
0
+ c
1
log GDP (iii)
log CEXD = d
0
+ d
1
log NGDP (iv)
log PPT = e
0
+ e
1
log TOR (v)
log PPT =f
0
+f
1
log GDP (vi)
log TOR = g
0
+ g
1
log GDP (vii)
log CIT = h
0
+ h
1
log GDP (viii)
log CIT = i
0
+ i
1
log NGDP (ix)
log TEXP = j
0
+ j
1
log GDP (x)
67
In order to capture the period of the oil boom and SAP, the slope dummy equations were
used for this period. The reason for adopting the slope dummy are expressed in Gujarati
and Porter, (2009), and Wonnacott and Wonnacott, (1980) that over long periods of time
or under unusual circumstances (like the oil boom and SAP in Nigeria), not only do the
functions (intercept) change but also their slopes may as well be expected to change. We
believe this situation might have applied to the Nigerian situation for both the oil boom
and SAP. However, the empirical evidence relating to this will be more reassuring,
hence the desirability of using the slope dummy function for our analysis.
3.2.2 The Slope Dummy Functions
Dummy variables refer to the technique of using two variables usually coded 0 or 1 to
represent the separate categories of a nominal level measure. The term dummy appears
to refer to the fact that the presence of the trait indicated by the code of 1 represents a
factor or collection of factors that are not measurable by any better means within the
context of our analysis. The Dummies can affect the model in two ways, first it either
shift the intercept up or down, or shift the slope shallower or deeper.
Interpretation of the dummy variable is usually quite straightforward. The intercept term
represents the intercept for the omitted category. The slope coefficient for the dummy
variable represents the change in the intercept for the category coded 1. Now, lets use
the GTR equation to demonstrate the difference between the shift (intercept) and the
slope dummy functions. For example, the shift (intercept) dummy variable function is
represented by:
logGTR = a
0
+ a
1
logGDP + a
2
Dum
1
......(15)
68
where
Dum = the dummy variable taking on values (0, 1).
To derive the slope dummy function, we introduce a second dummy variable D
2
equal to
the product of the explanatory variable and the first dummy D
1
. Applying this to the total
government revenue function (eq. 10) as an example, we have the slope dummy variable
equation as follows:
Log GTR = a
0
+ a
1
logGDP + a
2
Dum
1
+ a
3
Dum
2
.....(16)
where
Dum
2
= Dum
1
x TOR.
This function was applied to SAP and oil boom variables for all the equations in this
thesis.
Additional modifications were made to the preceding equations to enable us reasonably
capture the budgetary process in Nigeria as it relates to each of the revenue sources. In
practice, policy proposals in the annual budget are based on the performance of each
revenue source in the preceding period. For example, revenue sources that performed
above expectation in the out-going fiscal year are given more ambitious targets in the
new fiscal year, and are put under greater surveillance. This practice became prevalent
with the steady decline in oil revenue.
Administrative lag is another major factor. New policy guidelines announced in the
budget speech may not be implemented until the relevant circulars are issued. It may take
up to six months, however, from budget announcement before the content of such
69
circulars are implemented. This scenario applies particularly to customs and excise
duties. For company taxation, most companies do not discharge their tax liabilities until
long after the annual general meeting. To capture the potential effects of these issues, a
one-year lag of the explanatory variable was added to each equation. This will show not
only the relevance of this lagged value, but also its relative influence compared with
current years values. If there are pronounced administrative lags or delayed remittances,
for example, the lagged value will be more significantly associated with the dependent
variable in each equation.
Hence, the following represent the final equations used for non-dummy and dummy
based scenarios:
logGTR
t
= a
0
+ a
1
log GDP
t
+ a
2
log GDP
t-1
................................................... (17)
and
logGTR
t
= a
0
+ a
1
log GDP + a
2
log GDP
t-1
+ a
3
Dum
1
+ a
4
Dum
2
............... (18)
This is consistent with the logarithmic autoregressive model suggested by Pindyck and
Rubinfeld, (1998).
70
CHAPTER FOUR
PRESENTATION AND ANALYSIS RESULTS
4.0 Introduction
This chapter deals with the estimation and presentation of the regression result. It
therefore involves an estimation of the short run dynamic and long run relationship
models. Apart from that, the rest of the chapter will deal with presentation of regression
estimate and the policy implication of the results.
4.1 The Absolute Regression Model
The absolute values for the general model and the individual models are presented below.
The absolute regression model specified is given as:
GTR = a
0
+ a
1
GDP + a
2
TOR + a
3
NOR + a
4
PPT + a
5
CEXD + a
6
CIT + a
7
NGDP + a
8
TEXP
Thus, when the above regression equation was analyzed using the E-views, the resulting
model of the absolute values of all the variables for the entire period is presented below
as:
GTR = - 0.00337 + 0.051GDP + 0.140TOR - 0.124NOR - 0.141CEXD + 0.330PPT +
0.024CIT + 0.421NGDP + 0.285TEXP
t = -0.1358 0.9946 3.9812 -0.8100 -1.7675 7.1438
0.3098 2.6773 5.0190
S.E = 0.0249 0.0514 0.0351 0.1525 0.0798 0.0461 0.0760
0.1573 0.0568
R
2
= 0.939219 F = 59.87820 DW = 2.035461
Similarly, the regression equation for the oil boom period as:
71
GTR = -0.0291 + 0.0242GDP + 0.305TOR + 0.755NOR - 0.0335CEXD + 0.019PPT +
0.028CIT - 0.0002NGDP + 0.440TEXP + 0.319DUM1 - 0.000DUM2
t = -0.5500 0.5052 5.0244 8.9426 -1.4437 0.7458
1.0895 -5.5385 5.0350 1.5645 -1.2323
S.E = 0.5283 0.0480 0.06 06 0.0845 0.0232 0.0257
0 .0259 0.0000 0.0000 0.2042 0.0002
R
2
= 0.916687 F = 53.87820 DW = 2.247095
Also, the regression model obtained for the sap period is also presented below as:
GTR = -0.5316 + 0.0756GDP + 0.110TOR - 0.00007NOR - 0.0724CEXD - 0.0010PPT +
0.05197CIT + 0.000005NGDP + 0.5883TEXP 0.0042DUM1 + 0.0000DUM2
t = -1.3100 0.8267 2.6412 -0.3037 -0.3353 6.8050
0.8944 0.9108 2.6532
S.E = 0.0654 0.1275 0.0413 0.5214 -0.1243 0.0579
0.1865 0.4924 0.2408
R
2
= 0.857858 F = 83.1345 DW = 1.699191
From the regression models above, it is revealed that for the entire period, GDP, TOR,
CIT, PPT, NGDP and TEXP have positive impact on government total revenue (GTR)
while NOR and CEXD have negative influence on GTR. The regression also revealed
that GDP and CIT have the most insignificant influence of 0.050 and 0.024 on GTR,
which may be due to low level of economic activities in the country. From the result
also it is observed that, a percentage change in TOR, PPT, NGDP and TEXP will
increase GTR by 0.140, 0.330, 0.421 and 0.285 respectively while a percentage
72
change in NOR and CEXD will decrease GTR by 0.124 and 0.141. The t-values show
that four of the estimated parameters (TOR, PPT, NGDP and TEXP) are statistically
significant, while the other four (GDP, NOR, CIT and CEXD) are not significant and
they are statistically assumed to be zero because; their t-calculated is less than t-
tabulated 2.00. Therefore, the null hypothesis is accepted for each of the parameters of
the explanatory variables. However, the result of the F-statistic reveals that the
estimated parameters are simultaneously significant since their F-calculated value
(59.87820) is greater than F-tabulated. The coefficient of determination (R
2
) for the
entire period shows that the data fit the model well because about 85% of the total
variation in GTR is explained by the variation in all the variables used for the
analysis. The DW statistic value of 2.035461 shows that there is the no presence of
positive autocorrelation. The dependent variable (GTR) and the explanatory variables
are highly correlated and the overall regression is significant except for the NOR and
CIT.
The oil boom period in model two also revealed a positive linear relationship between
the variables and GTR except PPT which ironically exhibit a negative impact,
contrary to the a priori that a percentage change in PPT should lead to a positive
change in GTR. The t-statistic revealed that only TOR and TEXP that are statistically
significant while all other variables are not. However, an R
2
value shows that about
91% of the variation in GTR is explained by the variables used for the regression. On
the other hand, the DW statistic of 2.096311 shows that there is the absence of
73
positive autocorrelation and the F-statistic of 54.71380 indicates that the parameters
are simultaneously significant.
The model for the sap period presented above also did not reveal anything different,
as only CEXD have a negative (-0.042) and insignificant impact on GTR while all
other variables have positive impact on GTR. The R
2
value shows that about 96% of
the variation in GTR is explained by the variables used for the regression. Similarly,
the DW statistic value of 1.699191 which approximate 2 shows that there is the
absence of positive autocorrelation and the F-statistic of 39.77586 indicates that the
parameters are simultaneously significant.
4.2 The Unit Root Test Result
In order to assess stationarity and investigate the time series characteristics of the
variables (TGR, GDP, NGDP, TOR, NOR, CEXD, PPT, TEXP and CIT) were
subjected to the Unit Root Test using the Augmented Ducker Fuller (ADF) and the
Phillip Perron. A variable is stationary when it has no unit root and is denoted by 1(0)
while a non-stationary variable can have one or more unit roots. Stationarity is often
attained through differencing.
The Unit Root and the Philip Perron Test results summary for the various
categories of variables used for this study are therefore presented in tables 4.1 and 4.2
respectively below:
74
Table 4.1: ADF Unit Root Test Result with Intercept and Trend
ADF t-Statistic Value at Level ADF t-Statistics Value at First Difference
Variable None Intercept Trend & Intercept None Intercept Trend & Intercept No. of
Lag
Order of
Integration
GTR -10.29653 -10.88665 -11.63914 -17.85301 -17.57916 -17.31226 0 1(0)
GDP -3.940024 -4.505510 -4.776480 -6.866134 -6.767508 -6.660519 0 1(0)
TOR -10.64548 -10.96877 -11.27708 -20.27067 -19.98805 -19.75980 0 1(0)
NOR -4.893444 -5.627478 -7.988800 -17.33985 -17.26769 -17.26626 0 1(0)
CEXD -5.357864 -6.133567 -7.370466 -11.00532 -10.85922 -10.70750 0 1(0)
PPT -11.57182 -11.76901 -11.96985 -17.98466 -17.74717 -17.56840 0 1(0)
CIT -0.424396 -0.853814 -1.995121 -7.308456 -7.598434 -8.607207 0 1(1)
TEXP -4.848081 -5.600022 -7.420805 -12.19534 -11.99642 -11.96112 0 1(0)
NGDP -3.954616 -4.654526 -6.991366 -17.15379 -17.16474 -17.28982 0 1(0)
ADF Unit Root Test Critical Values
At Level At 1
st
Difference
Percent None Intersect Trend & Intercept None Intersect Trend & Intercept
1 -2.6227 -3.6067 -4.2092 -2.6243 -3.6117 -4.2165
5 -1.9495 -2.9378 -3.5279 -1.9498 -2.9399 -3.5312
10 -1.6202 -2.6069 -3.1949 -1.6204 -2.6080 -3.1968
75
Table 4.2: Philip Perron Test Result with Intercept and Trend
ADF t-Statistics Value at Level ADF t-Statistics Value at First Difference
Variable None Intercept Trend & Intercept None Intercept Trend & Intercept
TGR -10.29653 -10.88665 -11.63914 -17.85301 -17.57916 -17.31226
GDP -3.940024 -4.505510 -4.778480 -6.866134 -6.767508 -6.660519
TOR -10.64548 -10.96877 -11.27708 -20.27067 -19.98805 -19.75980
NOR -4.893444 -5.627478 -7.988800 -17.33985 -17.26769 -17.26626
CEXD -5.357864 -6.133597 -7.370466 -11.00532 -10.85922 -10.70750
PPT -11.57182 -11.76901 -11.96985 -17.98466 -17.74717 -17.56840
CIT -0.424396 -0.853814 -1.995121 -7.308456 -7.598434 -8.607207
TEXP -4.848081 -5.600022 -7.420805 -12.19534 -11.99642 -11.96112
NGDP -3.954616 -4.654526 -6.991366 -17.15379 -17.16474 -17.28982
Philip Perron Test Critical Values
At Level At 1
st
Difference
Percent None Intersect Trend & Intercept None Intersect Trend & Intercept
1 -2.6227 -3.6067 -4.2092 -2.6243 -3.6117 -4.2165
5 -1.9495 -2.9378 -3.5279 -1.9498 -2.9399 -3.5312
10 -1.6202 -2.6069 -3.1949 -1.6204 -2.6080 -3.1968
Source: Researchers Computation 2010.
76
The result from both the ADF unit root and the Phillip Perron tests revealed that all
the variables are stationary at the levels except CIT which was stationary after
taking the first difference. It is also revealed that all the variables are stationary are
1%, 5% and 10% critical levels while CIT observed to be stationary at 1%, 5% and
10% only after first difference at none. Since the variables used are characterised as
1(0), the OLS regression analysis was at levels.
4.3 Regression Result for the Entire Period
Table 4.3
Dependent Variable: LOG(GTR)
Method: Least Squares
Date: 12/29/11 Time: 21:21
Sample(adjusted): 1971 2010
Included observations: 40 after adjusting endpoints
Variable Coefficient Std. Error t-Statistic Prob.
LOG(GDP) 0.051125 0.051400 0.994647 0.3276
LOG(TOR) 0.139909 0.035142 3.981217 0.0004
LOG(NOR) -0.123532 0.152504 -0.810024 0.4241
LOG(CEXD) -0.141003 0.079773 -1.767549 0.0870
LOG(PPT) 0.329586 0.046136 7.143812 0.0000
LOG(CIT) 0.023551 0.076029 0.309768 0.7588
LOG(NGDP) 0.421156 0.157304 2.677344 0.0118
LOG(TEXP) 0.285136 0.056811 5.019032 0.0000
C -0.003374 0.024855 -0.135762 0.8929
R-squared 0.939219 Mean dependent var 0.233796
Adjusted R-squared 0.923533 S.D. dependent var 0.346518
S.E. of regression 0.095821 Akaike info criterion -1.657557
Sum squared resid 0.284633 Schwarz criterion -1.277559
Log likelihood 42.15113 F-statistic 53.87820
Durbin-Watson stat 2.035461 Prob(F-statistic) 0.000000
Sources: E-View Output 2010
The table shows that all the variables used for the analysis except NOR and CEXD
have positive linear relationship with GTR. This is indicated by the positive
coefficients exhibited by the variables. The coefficient of determinations given as
0.939219 suggests that about 94% of the variations in GTR is explained by the
77
behaviour of the explanatory variables. The Durbin-Watson statistic is 2.035461 is
an evidence of the absence of serial correlation among the residuals in the model.
4.3.1 Regression Result for the Oil Boom Period
Table 4.4
Dependent Variable: LOG(GTR)
Method: Least Squares
Date: 05/07/12 Time: 22:45
Sample: 1970 2010
Included observations: 41
Coefficient Std. Error t-Statistic Prob.
C -0.029054 0.528267 -0.054998 0.9565
LOG(GDP) 0.024258 0.048016 0.505211 0.6171
LOG(NOR) 0.755226 0.084452 8.942645 0.0000
LOG(NGDP) -1.52E-06 2.75E-07 -5.538483 0.0000
LOG(CEXD) -0.033536 0.023229 -1.443681 0.1592
LOG(TOR) 0.304507 0.060606 5.024384 0.0000
LOG(PPT) 0.019174 0.025709 0.745830 0.4616
LOG(CIT) 0.028200 0.025883 1.089483 0.2846
LOG(TEXP) 2.21E-07 4.39E-08 5.034968 0.0000
DUM1 0.319209 0.204161 1.563517 0.1284
DUM2 -2.13E-05 1.73E-05 -1.232192 0.2274
R-squared 0.926687 Mean dependent var 11.49892
Adjusted R-squared 0.905583 S.D. dependent var 2.851569
S.E. of regression 0.189523 Akaike info criterion -0.264398
Sum squared resid 1.077573 Schwarz criterion 0.195341
Log likelihood
Durbin-Watson stat
16.42016
2.047095
F-statistic
Prob(F-statistic)
283.1345
0.000126
Sources: E-View Output 2010
Similarly, the ADF result shown in table 4.2 above indicated that all the variables
used for the analysis except NGDP and CEXD have positive linear relationship
with GTR. This is indicated by the positive coefficients exhibited by the variables.
The coefficient of determinations given as 0.926687 suggests that about 99% of the
variations in GTR is explained by the behaviour of the explanatory variables during
the oil boom period. The Durbin-Watson statistic of 2.047095 is an evidence of the
absence of serial correlation among the residuals in the model.
78
The ADF result for the SAP period shown in table 4.3 below reveals the same result
with the entire period, in which all the variables used for the analysis except NOR,
CEXD and PPT have positive linear relationship with GTR. The coefficient of
determinations (R
2
) given as 0.897769 suggests that about 100% of the variations in
GTR is explained by the behaviour of the explanatory variables. The Durbin-
Watson statistic of 1.855296 is an evidence of the absence of serial correlation
among the residuals in the model. From the ADF also, the variables show that they
are statistically significance as indicated by their low probability except PPT.
4.3.2 Regression Result for the SAP Period
Table 4.5:
Dependent Variable: LOG(GTR)
Method: Least Squares
Date: 05/07/12 Time: 22:23
Sample: 1970 2010
Included observations: 41
Coefficient Std. Error t-Statistic Prob.
C 0.531644 0.574738 0.925020 0.3623
LOG(GDP) 0.075602 0.038631 1.957047 0.0097
LOG(NOR) -5.64E-06 2.50E-06 -2.255959 0.0115
LOG(NGDP) 5.41E-06 2.58E-06 2.097136 0.0445
LOG(CEXD) -0.072434 0.019578 -3.699700 0.0009
LOG(PPT) -0.000984 0.022192 -0.044320 0.9649
LOG(TOR) 0.309532 0.079822 3.877792 0.0005
LOG(CIT) 0.051967 0.022852 2.274110 0.0003
LOG(TEXP) 0.588321 0.093006 6.325627 0.0000
DUM1 -0.004189 0.115214 -0.036358 0.9712
DUM2 8.12E-08 3.88E-08 2.093812 0.0448
R-squared 0.897769 Mean dependent var 11.49892
Adjusted R-squared 0.867025 S.D. dependent var 2.851569
S.E. of regression 0.155541 Akaike info criterion -0.659607
Sum squared resid 0.725788 Schwarz criterion -0.199868
Log likelihood
Durbin-Watson stat
24.52195
1.855296
F-statistic
Prob (F-statistic)
178.37503
0.000000
Sources: E-view Output 2010
79
4.4 Productivity of Nigerian Tax System (1970-2010)
Table 4.4 below shows the productivity indexes for the overall and individual
revenue sources for the period 1970 2010. From the table productivity indexes
for the whole and the individual revenue sources for the period 1970-2010 are
presented. The results indicate that out of the ten equations, 7 equations are
statistically significant with a t-value greater than 2. But the elasticity indexes for
all the tax sources lies below unity, as NOR (Eq.2) has elasticity coefficient of
0.927 in relation to NGDP while TOR (Eq. 7) has elasticity coefficient of 0.595 in
relation to GDP. However, since all equations shows positive elasticity coefficient,
it implies that all revenue sources can contribute positively to government total
revenue if properly managed.
The lagged values (LV) of the explanatory variables (column 4) also showed that
the 70% of the ten equations had lower elasticity indexes of the tax sources relative
to their respective tax bases, while the other three equations had elasticity indexes
relatively higher than what was obtained in their current values (column 3). The
finding from the lag values also indicates that 70% of the ten equations are
statistically significant with a t-value greater than 2. Thus, from the finding, it could
be ascertained that there should be need to capture the effect of policy lags on tax
yield. This was confirmed by Ariyo, (1997) when he opined that apart from
administrative lags, the enabling regulations allow for grace periods between the
due date of a tax liability and the actual time to remit.
80
The reports from the oil boom period as presented in table 4.7 indicate significant
variation from what obtained during the entire period. Only three of the ten
equations that are statistically significant both for current and lag values. This
means that for all variables, the t-values indicate that only NOR, CEXD and PPT
are statistically significant at the 95% confidence level. The coefficients of
determination (R
2
) show that, NOR, CEXD and PPT could explain over 80% of the
pattern of the behaviour of each dependent variable. This is not unexpected as the
neglect of the non-oil revenue (NOR) sources, had significant impact on GTR. This
also significantly affects CEXD and PPT respectively.
Table 4.6: Productivity Indexes of the Nigerian Tax System
Equations Constant Elasticity
Coefficient
Elasticity
Coefficient
(Lag Values)
F-stat R
2
DW SER
1 0.159
(2.800)
0.428
(2.855)
0.442
(2.972)
81.489 0.87 2.377406 0.150
2 0.015
(0.766)
0.927
(17.215)
0.927
(16.976)
256.364 0.88 3.101980 0.054
3 0.156
(3.375)
0.098
(0.806)
0.094
(0.763)
36.499 0.72 2.366269 0.122
4 0.063
(1.637)
0.527
(5.041)
0.526
(4.971)
225.409 0.80 2.282466 0.105
5 0.119
(1.776)
0.494
(4.742)
0.493
(4.657)
224.847 0.77 2.637976 0.104
6 0.197
(2.299)
0.291
(1.287)
0.299
(1.304)
165.611 0.74 2.106599 0.226
7 0.156
(1.533)
0.595
(2.219)
0.607
(2.240)
149.209 0.97 2.409109 0.268
8 0.216
(5.660)
0.136
(1.347)
0.133
(1.304)
94.068 0.82 2.532980 0.101
9 0.248
(5.842)
-0.060
(2.077)
0.221
(2.040)
127.071 0.60 2.459585 0.114
10 0.163
(2.368)
0.414
(2.267)
0.417
(2.253)
51.413 0.71 2.393816 0.182
Source: Researchers Computation, 2010
The lagged values (LV) of the explanatory variables for oil period also revealed
that only one of the equations (10%) had elasticity indexes greater than 1, while the
81
other 9 of the equations had elasticity indexes relatively below unity with equations
2 and 4 having elasticity indexes of 0.861 and 0.725 respectively.
The result of the SAP period reported in table 4.6 below also revealed that all the
variables have positive linear relationship with GTR. The result revealed that equations
1, 2, 4, 5, 7, 9 and 10 are statistically significant while equations 3, 6 and 8 with t-
values of 1.890, 1.031 and 1.429 are not statistically significant for both current and
lagged values.
Table 4.7 revealed the effect of the oil boom on the productivity of these revenue
sources, using a slope dummy function. The result is not significantly different
from the general result except that the oil boom led to a significant shift in non-oil
revenue base due to interaction effect of the oil boom and the non-oil sector. The
elasticity index for GTR during the oil boom was very significant for both current
and lagged values of GDP. This same was revealed for CEXD, TOR and TEXP for
both the current and lagged values of GDP.
The result of the Sap period as reported in table 4.6 revealed that all the variables
have positive linear relationship with GTR except for PPT on TOR and CIT on
NGP which exhibits negative relationship. The result revealed that equations i, ii,
iv, v, vii and x are statistically significant while equations iii, vi and viii are
statistically significant for both current and lag values. Since the non-oil revenue
(NOR) is statistically significant in explaining GTR, it further suggests that during
82
the SAP period, the administration and collection of taxes from the non-oil revenue
sources had some positive effect. This is in line with Ariyo (1997) work when he
opined that, during the SAP period the government emphasizes the efficiency in the
administration and collection of non-oil revenue tax sources, which led to the
positive impact
Table 4.7: The Oil Boom and the Tax Yield (1970 2010)
Equa-
tion
Constant Elasticity
Coeff.
Elasticity
Coefficient
(Lag
Values)
Slope
Dummy
Shift F-Stat R
2
DW SER
1 -3.8634
(-1.933)
1.267
(8.253)
0.339
(0.570)
-0.00006
(-2.0351)
-0.1152
(-0.1674)
42.27 0.89 1.6338 0.153
2 0.018
(0.308)
0.998
(195.682)
0.414
(2.294)
0.00002
(-1.1712)
-0.0569
(-0.9723)
119.13 0.95 2.5547 0.005
3 9.470
(3.837)
0.319
(1.685)
0.472
(0.528)
0.000005
(0.1397)
-0.1502
(-0.1758)
270.59 0.93 2.2072 0.101
4 14.323
(9.206)
-0.066
(-0.486)
0.266
(1.384)
0.000575
(1.1786)
-2.5414
(-1.6194)
134.53 0.87 2.4741 0.630
5 13.675
(11.366)
0.022
(0.225)
-0.097
(-0.789)
0.000122
(0.6970)
0.5573
(0.3858)
149.46 0.91 1.8830 0.823
6 15.570
(7.484)
-0.127
(-0.792)
0.159
(0.1665)
-0.0000007
(0.15053)
1.0711
(1.4867)
106.16 0.58 1.6216 0.433
7 -5.531
(-2.740)
1.366
(8.817)
0.220
(0.276)
-0.000007
(-2.2364)
0.2272
(0.3251)
114.33 0.74 1.9219 0.619
8 -11.362
(5.150)
0.125
(0.739)
2.165
(3.098)
0.0000003
(0.1014)
0.3454
(0.4523)
135.91 0.55 2.4122 0.046
9 3.014
(2.582)
-0.004
(-0.029)
0.801
(2.348)
0.000370
(0.8696))
-0.9197
(-0.6714)
126.71 0.80 2.0101 0.091
10 -4.081
(-2.003)
1.301
(8.320)
0.107
(0.123)
0.000007
(-2.2180)
-0.3401
(-0.4822)
63.59 0.72 1.8501 0.000
Source: Result Computed by Researcher (2011)
83
Table 4.8: SAP and Tax Yield 1970 2010
Equa-
tion
Constant Elasticity
Coeff.
Elasticity
Coefficient
(Lag
Values)
Slope
Dummy
Shift F-stat R
2
DW SER
1 -3.357
(-1.085)
1.151
(3.897)
0.930
(4.163)
-0.0000005
(-1.6559)
1.7887
(2.1597)
174.14 0.66 2.01776 0.061
2 0.004
(0.044)
1.000
(15.410)
1.014
(14.766)
-0.000000005
(0.628)
-0.0045
-(0.0992)
208.91 0.98 2.31187 0.020
3 6.191
(1.589)
0.616
(1.657)
0.328
(1.777)
0.0000006
(-1.5432)
0.0632
(0.0605)
222.47 0.97 1.89794 0.306
4 13.100
(6.380)
0.945
(24.783)
0.375
(2.669)
-0.000002
(-1.7294)
2.2734
(2.1894)
230.98 0.88 1.76143 0.207
5 5.431
(1.450)
-0.091
(-0.421)
0.315
(2.329)
0.00000008
(0.3285)
-0.2401
(-0.2441)
248.12 0.96 1.7562 0.376
6 12.920
(4.044)
0.1694
(0.555)
0.402
(1.070)
-0.00000085
(-2.7865)
-0.0138
(-0.0161)
171.05 0.94 1.6923 0.051
7 -5.329
(-1.706)
1.299
(4.355)
1.185
(2.408)
-0.00000058
(-1.9245)
1.5676
(1.8746)
127.20 0.45 2.386 0.001
8 6.983
(2.326)
0.537
(1.874)
0.136
(1.311)
-0.00000011
(-3.6872)
0.3718
(0.4627)
107.39 0.51 2.3401 0.269
9 15.421
(8.095)
-0.383
(-1.526)
0.168
(2.091)
0.00000025
0.2786)
2.2727
(2.3587)
327.77 0.97 1.7467 0.136
10 -3.068
(-0.952)
1.119
(3.637)
1.090
(4.050)
-0.00000043
(-1.4029)
2.1026
(2.4381)
181.07 0.54 2.0706 0.001
Source: Result Computed by Researcher (2011)
The elasticity indexes for all the tax sources showed positive elasticity coefficient
except equations v and ix. This finding is also in line with Ariyo (1997) who
established a significant relationship between GTR and the various tax sources and
a leakage between the variables and the magnitude of the revenue involved. Also
revealed in the table is the effect of SAP using the slope dummy function. It is
revealed that there is no relative difference between the results obtained for the
84
periods. That is, in our analysis, 6 of the equations are statistically significant
indicating that the results obtained from the general equation are not different from
what is obtained during the SAP period.
4.5 Analysis of Tax Legislative Control in Nigeria
The table below shows that the federal government exercises legislative control
over the first 14 tax sources, while the states are in charge of the remaining 6
sources. It is noteworthy that the local government has no legislative power over
any revenue source, although it can initiate bylaws subject to the approval of the
state government. The Federal Government of Nigeria (FGN) also dominates tax
administration and collection. For example, the FGN directly collects revenue for
the first 7 items, which accounted for over 80% of total tax-based revenue in the
country.
From the table also, the state government is responsible for the collection of
revenue for items 8 to 18, which cumulatively account for an insignificant
proportion of the total tax-based revenue. The local government controls only two
items. It does appear that administrative efficiency is the overriding criterion
guiding the assignment of tax sources to the different tiers of government.
Consequently, all the major tax sources have been assigned to the federal
government. This observation provides a valuable guide as to the appropriate focus
for this study, in two respects. First, it is cost-effective to focus on tax sources that
are both legislatively and administratively under the control of the FGN. Second,
85
on a close study, only four or five of these sources account for about 80% of total
tax-based revenue: these are customs and excise duties, mining and royalties,
petroleum profits tax, and companies income tax. Consequently, the study focuses
on these major tax sources. We believe that the findings emanating from the study
are validly generalizable to the Nigerian environment within the context of this
studys objectives.
Table 4.9: Nigerias Major Taxes, Since 1990
Types of Tax Jurisdiction
Legislation Administrative
Collection
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Import Duties
Excise Duties
Export Duties
Mining Rent and Royalties
Petroleum profit tax
Company income tax
Personal income tax
Armed Forces, External Affairs officers and
Federal Capital Territory
Capital gain tax
Personal income tax
License fees on television and wireless radio
Stamp duties
Estate duties
Gift tax
Sales of purchase tax
Football pools and other betting taxes
Motor vehicle tax and drivers license fees
Entertainment tax
Federal
Federal
Federal
Federal
Federal
Federal
Federal
Federal
Federal
Federal
Federal
Federal
Federal
Federal
State
State
State
Federal
Federal
Federal
Federal
Federal
Federal
Federal
State
State
State
State
State
Stat
State
State
State
State
86
18
19
20
Land registration and survey
Property tax
Market and trading license fees
State
State
State
State
Local
Local
Source: Adopted from Ariyo (1997)
4.6 Structural Changes in Some Federal Tax Sources, 1989-2010
The table below revealed the changes in some tax sources between 1989 -2010. From
the table it is revealed that there is a decreasing trend in income from the non-oil
sector, as the oil sector contributed as high as 86.2% in 1992 and 88.3% in 2006. In
2010, the oil sector still contributes 73.9% of total revenue while the non-oil sector
contributes 26.1%. By this, it implies that the country depends on the oil sector for
almost all revenue generated as against other sectors. The trend of decreasing
income from non-oil sector and increasing income from the oil sector makes the
Nigeria economy to suffer a Dependence Syndrome. An examination of most
countrys sources of revenue profile implies that taxation generates a larger
proportion of their national revenue, which is efficiently used for economy planning
and regulates the economic activities of the country. This is not the case in Nigeria,
as the opposite operates.
Table 4.10: Structural Changes in Some Federal Tax Sources, 1989-2010 (Percent)
Oil Non-OIL
YEAR PPT OTHERS TOTAL CIT DUTIES OTHERS TOTAL G/TOTAL
1989 21.1 61.1 82.2 3.9 11.8 2.0 17.8 100.0
1990 36.1 45.7 81.8 5.3 12.9 0.0 18.2 100.0
1991 38.2 43.6 81.9 6.8 11.3 0.0 18.1 100.0
1992 27.0 59.1 86.2 5.4 8.4 0.0 13.8 100.0
1993 30.7 53.4 84.1 5.0 8.0 2.9 15.9 100.0
87
1994 21.2 58.1 79.3 6.1 9.1 5.5 20.7 100.0
1995 9.3 61.2 70.6 4.8 8.1 16.6 29.4 100.0
1996 14.7 56.2 71.0 4.2 10.6 14.2 29.0 100.0
1997 11.8 59.8 71.5 4.5 10.8 13.2 28.5 100.0
1998 14.7 55.3 70.0 7.2 12.4 10.4 30.0 100.0
1999 17.3 59.0 76.3 4.9 9.3 9.5 23.7 100.0
2000 27.5 56.0 83.5 2.7 5.3 8.5 16.5 100.0
2001 28.6 47.9 76.5 3.1 8.0 12.4 23.5 100.0
2002 22.6 48.4 71.1 5.7 10.5 12.8 28.9 100.0
2003 26.5 54.0 80.6 4.5 7.6 7.4 19.4 100.0
2004 30.2 55.4 85.6 3.3 5.5 5.6 14.4 100.0
2005 34.3 51.5 85.8 2.9 4.2 7.0 14.2 100.0
2006 34.2 54.1 88.3 4.1 3.0 4.6 11.7 100.0
2007 26.3 51.8 78.1 5.7 4.2 12.0 21.9 100.0
2008 35.7 47.3 83.0 5.3 3.6 8.1 17.0 100.0
2009 45.6 20.3 65.9 11.7 6.1 16.2 34.1 100.0
2010 36.0 37.9 73.9 9.0 4.2 12.9 26.1 100.0
Source: Central Bank of Nigeria Annual Report and Statement of Account, various Issues.
Percentages Computed by Researcher.
For a clearer understanding, the percentages changes in revenue generated from the
oil and Non-oil sources of revenue are presented graphically below. It is clearer
from the graph that more revenue are collected from the oil sources while less revenue
is collected from the non-oil sources.
88
Fig.4.1: Percentage Changes in Revenue Generation in Oil and Non-Oil Sectors
(1989 -2010) (Minitab Output)
Where
Revenue from Oil Sources
Revenue from Non-Oil Sources
4.7 Analysis of Federal Government Fiscal Operation (1970-2010)
The analysis of Federal Government Fiscal Operation from 1970 2010) is
presented in the tables 4.10(a) to 4.10(d) below. From the tables, it is revealed that
Nigeria experiences recurrent fiscal deficit throughout the period under
consideration except for the years 1995 and 1996 that experiences fiscal surplus.
0
10
20
30
40
50
60
70
80
90
100
Source: Graphed by Researcher (2011)
89
Even at those years, Anyanwu (1991) opined the fiscal the fiscal surplus of 1995
and 1996 were mere arithmetical manipulations to conceive the international world
that Nigeria is financial buoyant to qualify them for another loan.
90
Table 4.10 (a ): Federal Governments Fiscal Operation (1970-1989) (N million)
Items/Year 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979
Total Federally
Retained Revenue
448.8 1,168.8 1,404.8 1,695.3 4,537.0 5,514.7 6,765.9 8,042.4
5,178.1
8,868.4
Total Federal
Govt. Expenditure
903.9 997.2 1,463.6 1,529.2 2,740.6 5,942.6 7,856.7 8,823.8
8,000.0
7,406.7
Overall Surplus
(+) /Deficit (-) (455.1) 171.6 (58.8) 166.1 1,796.4
(427.9) (1,090.8) (781.4)
(2,821.9)
1,461.7
Items/Year 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
Total Federally
Retained Revenue
12,993.3 7,511.6 5,819.1 6,272.0 7,267.2 10,001.4 7,969.4 16,129.0
15,588.6
25,893.6
Total Federal Govt.
Expenditure
14,968.5 11,413.7 11,923.2 9,636.5 9,927.6 13,041.1 16,223.7 22,018.7
27,749.5
41,028.3
Overall Surplus (+)
/Deficit (-)
(1,975.2)
(3,902.1)
(6,104.1)
(3,364.5) (2,660.4)
(3,039.7) (8,254.3)
(5,889.7)
(12,160.9)
(15,134.7)
Sources: Central Bank of Nigeria: Annual Report and Statements of Accounts (various years); Federal Office of Statistics;
Annual Digest of Statistics (various years).
91
Table 4.10(b): Federal Governments Fiscal Operation:1990-2010 (N million)
Items/Year 1990 1991 1992 1993 1994 1995 1996 1997 1998
Total Federally
Retained Revenue 38152 30829 53265 126071 90623 249768 369267 423215 353724
Total Federal Govt.
Expenditure 60268 66584 92797 233807 160893 248768 337218 428215 487113
Overall Surplus (+)
/Deficit (-) (22,116.1) (35,755.2) (39,532.5) (107,735.3) (70,270.6) 1,000.0 32,049.4 (5,000.0) (133,389.3)
Items/Year 2001 2002 2003 2004 2005 2006 2007 2008 2009
Total Federally
Retained Revenue 79696.7 716754.2 1023241.2 1253600.0 1660700.0 1836605.0 2333659.6 3193440.0 2646904.7
Total Federal Govt.
Expenditure 10108025.6 1018155.8 1225965.9 1426201.3 1822100.0 1938002.0 2450896.7 3240818.5 3456925.4
Overall Surplus (+)
/Deficit (-) (221,048.9) (301,401.6) (202,724.7) (172,601.3) (161,400.0) (101,397.5) (117237.1) (47,378.5) (810,.20.7)
Sources: Central Bank of Nigeria: Annual Report and Statements of Accounts (various years); Federal Office of Statistics;
Digest of Statistics (various years).
Note: All figures in parenthesis are negative indicating deficit for the years under consideration.
92
Table 4.10(c): Nigerian Fiscal Deficit Operation: 1990-2000 (N millions)
Item/Year 1990 1991 1992 1993 1994 1995 1996 1997 1998
TR 38,152.10 30,829.20 53,264.90 126,071.20 90,622.60 249,768.10 369,267.00 423,215.20 353,724.10
nB -181,352.02 -171,624.96 -118,597.50 -290,885.31 -91,351.78 2,200.00 108,967.96 -19,000.00 -306,795.39
Subtotal -143,199.92 -140,795.76 -65,332.60 -164,814.11 -729.18 251,968.10 478,234.96 404,215.20 46,928.71
Less (G - C) 36,219.60 38,243.50 53,034.10 136,727.10 89,974.90 127,629.80 124,291.30 158,563.50 178,097.80
Sub-Total -106,980.32 -102,552.26 -12,298.50 -28,087.01 89,245.72 379,597.90 602,526.26 562,778.70 225,026.51
Less D 6,092.60 32,112.40 46,716.70 91,136.00 60,247.60 7,102.20 -143,189.50 -60,637.10 103,885.70
Subtotal -100,887.72 -70,439.86 34,418.20 63,048.99 149,493.32 386,700.10 459,336.76 502,141.60 328,912.21
External Assets -187,900.00 2,972.60 -11,859.60 16,963.50 8,390.80 22,455.40 7,825.40 13,382.60 16,600.60
Overall Balance -288,787.72
-
67,467.26 22,558.60 80,012.49 57,884.12 409,155.50 467,162.16 515,524.20 345,512.81
Source: Computed from CBN Annual Reports and Statements of Accounts (1970 2009) various issues.
Interpretations of Symbols: TR = Federal Government Retained Revenue G = Total Government Expenditure
D = Total Domestic Debt n = Growth rate of gross Domestic Product C = Total Government Capital Expenditure
B = Budget Deficit NW = Net worth
93
Table 4.10(d): Nigerian Fiscal Deficit Operation: 2001-2010 (N millions)
Item 2001 2002 2003 2004 2005 2006 2007 2008
TR 796,976.70 716,754.20 1,023,241.20 1,253,600.00 1,660,700.00 1,836,605.00 233,659.60 3,193,400.00
nB -150,171.84 -130,457.25 -700,279.45 -463,785.96 6,200.00 221,140.86 -16,500.00 -853,691.52
Subtotal 646,804.86 586,296.95 322,961.75 789,814.04 1,666,900.00 2,057,745.86 217,159.60 2,339,708.48
Less (G - C) 579,330.10 696,777.70 984,277.60 1,074,901.30 1,302,600.00 1,385,616.70 1,691,573.70 2,117,362.50
Subtotal (NW) 1,226,134.96 1,283,074.65 1,307,239.35 1,864,715.34 2,969,500.00 3,443,362.56 1,908,733.30 4,457,070.98
Less D 118,720.00 149,026.70 163,746.40 46,481.30 143,500.00 45,146.10 212,300.00 150,700.00
Subtotal 1,344,854.96 1,432,101.35 1,470,985.75 1,911,196.64 3,113,000.00 3,488,508.66 2,121,033.30 4,607,770.98
External Assets
(NET) 1,472,101.80 1,414,330.70 1,502,358.50 2,962,522.20 4,277,730.10 6,258,033.00 7,482,040.00 8,616,524.60
Overall
Balance 2,816,956.76 2,846,432.05 2,973,344.25 4,873,718.84 7,390,730.10 9,746,541.66 9,603,073.30 13,224,295.58
Source: Computed from CBN Annual Reports and Statements of Accounts (1970 2010) various issues.
Interpretations of Symbols: TR = Federal Government Retained Revenue G = Total Government Expenditure
D = Total Domestic Debt n = Growth rate of gross Domestic Product C = Total Government Capital Expenditure
B = Budget Deficit NW = Net worth
94
4.8 Discussion of Research Findings
To provide further evidence on the productivity of the tax system on the index of tax
buoyancy of the tax system was computed from equations i x. In the computation,
the data were divided into three groups and the results are recorded in table 5 below.
Group A shows the overall tax buoyancy, group B is for the Oil boom period while
Group C recorded the SAP period.
For the entire period covered by this study, there was a buoyancy of 1.230 for GDP
relative to GTR and 1.155 for CEXD relative to NGDP. In fact, contrary to our
expectations, the non-oil component however performed slightly lower with a
buoyancy of 1.000, 1.002 and 1.004 respectively for the three periods. The result
further revealed that the Nigeria tax structure is buoyant for about five of the tax
sources while it is not buoyant for other five tax sources for the period under
consideration. This fact was supported by Okpara (2010) when he opines that, a tax
structure is said to be buoyant if the buoyancy index is greater than unity implying
that as national income or the proxy base changes, tax revenue changes by a larger
proportion as a result of built-in elasticity and discretionary changes.
There was no significant change in the buoyancy for CEXD both in relation to GDP
and NGDP respectively as they exhibited a buoyancy of approximately 1 as shown in
table 4.12 below. Generally, the results show the effect of administrative lags and
lapses in the implementation of tax-related policies. This lag could have resulted from
the time in tax collection and remittance of proceed to government.
95
Column B of table 4.12 (Oil Boom Period) shows that a buoyancy of 1.401 and 1.338
for GDP and TOR, were recorded. Although, the coefficient of buoyancy for PPT in
relation to TOR remains high and this was attributable to improvement in the prices of
oil after 1985, and the deregulation of the oil sector.
Table 4.11: Index of Tax Buoyancy 1970 2010
Equation Overall Tax
Buoyancy
(A)
Oil-Boom
Period
(B)
Sap/Post Sap
Period
(C)
1 1.230 1.401 1.133
2 1.000 1.002 1.004
3 1.007 1.338 0.871
4 1.155 1.122 1.202
5 1.127 1.090 1.045
6 0.878 0.702 0.722
7 0.893 0.644 0.691
8 0.815 0.704 0.671
9 0.934 0.828 0.926
10 0.779 0.887 0.670
Source: Result Computed by Researcher (2011)
However, there was a downward trend when compared to Ariyo (1997) report, which
establish a buoyancy coefficient of 2.60 and 1.88 for the same components. This
downward trend may not be surprising phenomenon as the crises in the Niger Delta
96
may have significantly affected oil production. The buoyancy of 0.644 for TOR
during the oil boom was not encouraging compare with the previous, which has a
buoyancy of 0.893. This downward trend in the buoyancy for TOR however is a case
that required further investigation.
4.9 Policy Implications
The results of the findings suggest that deficit financing is a recurrent feature of
government fiscal activity not in Nigeria but in the world as a whole. Thus, while it is
inevitable to stop the government from engaging in the fiscal deficit option, there is
no gain saying that this has some policy implications on the economy. Firstly,
according to Obi, and Nurudeem (2009) deficit financing leads to huge debt stock and
tends to crowing-out of private sector investment, by reducing the access of investors
to adequate funds. This reduces investment in the economy, which creates
unemployment, reduces import and export as well as Company Income Tax (CIT).
Similarly, Anyanwu (1997) opined that fiscal deficit leads to unfavourable effect on
productive capital stock, and that persistent large government deficit would inevitably
result in increase government debt as a ratio of GDP.
97
CHAPTER 5
SUMMARY, CONCLUSION AND RECOMMENDATIONS
5.0 Introduction
In the preceding Chapter, a detailed discussion of methodological issues that are
essential for the achievement of the aim and objectives of this study was carried out.
Therefore, the chapter was devoted to refinement of findings from this research work
and discussion of the policy implications. In this concluding Chapter however, the
study endeavours to provide a closing summary of the research work, followed by
recommendations, to address some of the problems highlighted in the preceding
Chapter and conclusion. An attempt is also made to highlight relevant areas for future
research.
5.1 Summary
This study evaluated the productivity of the Nigerias tax system over a period 41
years (1970-2010). The overall work was based on the critical examination of some of
the key variables that are related to revenue generation and hence to determine the
productivity of the tax system. Some of these variables are government total revenue
(GTR), total oil revenue (TOR), company income tax (CIT), petroleum profit tax
(PPT), non-oil revenue (NOR) and the gross domestic product (GDP). Based on the
empirical analysis of the result, we summarized the findings and made some
recommendation.
The main objective of the research work is to assess the productivity of the Nigeria
tax system from 1970 2010. However, the following specific objectives were state:
1. To review the Nigeria tax system since the attainment of independence in 1960
98
2. To investigate whether an increase in tax revenue leads to the same
proportionate increase in Gross Domestic Product in Nigeria
3. To examine the index of tax buoyancy 1970 2009
4. To examine the deficit profile of the country
5. To examine the consequences of the neglect of the non-oil tax sources over
time.
6. To investigate the financing options of the Nigeria fiscal deficit
7. To highlight as much as possible the various tax reforms and their objectives.
8. To examine the structure of the Nigerian tax based revenue between
1970 2010.
In order to achieve the objectives, hypotheses were formulated and tested using the
Augmented Dickey Fuller to test the stationarity of the variables. The drift of the
variables form the normal were established with the cointegration test and the error
correction model.
Although the productivity level appears satisfactory overall, the results indicate wide
variations in the level of productivity by tax source. This was attributable to laxity in
the administration of non-oil tax sources during the oil boom periods, which suggests
the effect of the Dutch-diseases syndrome discussed earlier. Some of this laxness was
mopped up at the end of the honeymoon, however, especially with the
commencement of SAP in 1986.
In the context of this studys objectives, it appears the current revenue profile of the
nation is sustainable. Further, opportunities for improvement exist especially in
99
import duty collection. Better monitoring and transparency of operations within the
petroleum oil industry will also ensure a significant increase in total government
revenue. The government should also desist from revenue bursting activities such as
unbridled granting of both the prohibition and duty waiver for public sector projects
and few privileged individuals within the society.
5.2 Conclusion
This study was conducted to determine the productivity of the Nigerias tax system
during the 1970 2010. Ten models were formulated for the study and the ordinary
least square method was used to analyze the data. As a preliminary check, the ADF
unit root test was used to determine the presence of unit roots. The variables were
characterized as 1(0). The regression in logs was done using the data at levels.
To assess the productivity of the Nigeria tax system, two measures - index of tax
efficiency and index of tax buoyancy developed by Houghton (1998) were adopted. In
carrying out the analysis, the data set was disaggregated into sub-periods to enable us
trace the effects of changes in the macroeconomic environment. The analysis was
done at three levels of aggregation i.e. using the entire data set (Period A) at one level
(1970-2010), the oil boom period (B) at another level (1970-1984) and the SAP and
Post SAP period (C) at another (1986-2010).
The results of the analysis indicate that overall i.e. (Period A) 7 out of the 10
equations recorded elasticity index of less than 0.5 while the remaining 3 had
elasticity of between 0.5 and 0.9. The analysis for the oil boom period shows that the
PPT has elasticity of 1.01, which indicates that the petroleum profit tax (PPT) is
100
productive during the oil boom period. During this period, only 4 out of the 10
equations came out with elasticity of between 0.5 - 0.9 while the remaining 6
equations had elasticity of below 0.5. This suggests that even during the oil boom
period, the tax system was not productive. The analysis for the period of SAP was not
significantly different from the preceding two analyses. It therefore follows that the
Nigeria tax system is less productive irrespective of the level of data aggregation.
Therefore, of greater concern to this research work is the findings of a study reported
by Ariyo and Raheem, (1990) that shows that the level of fiscal deficit in Nigeria is
no longer sustainable, given the identified lack of co-integration between its revenue
and expenditure profiles. Ariyo (1993) also used the litmus test developed by Zee
(1988), Blinder and Solow (1983), and Buiter (1983), and modified by Rutayasire
(1990), to assess the sustainability of the Nigerian fiscal deficit between 1970 and
1990. Their findings indicate that the fiscal deficit profile in Nigeria has become non-
sustainable since 1970. They also provide policy and relevant information about the
causes, structure and severity of the deficit problem. Observation of table 2.2 shows
available evidence and researches indicates that, this deficit-prone policy orientation
continues unabated, in spite of its negative effect on the economy. In the light of the
above, the World Bank (1995) once opines that it is precisely a conviction that the
government is shortsighted in its policies, and that it is biased towards overspending
because of the nature of our political economy, and that makes sustainability an issue.
This fiscal deficit problem suggests the need for concrete steps to bring Nigerias
fiscal profile back on course. To achieve this, the country needs to either expand their
101
revenue generation sources in other to increase revenue or reduce the level of
expenditure or embark on an appropriate combination of both.
All over the world, a productive tax system provides sufficient revenue for
government spending. From our analysis, this is not the case for Nigeria. The options
open to government for closing the gap between revenue from the tax system and
expenditure is to embark on fiscal deficit financing. This study has established that the
persistent fiscal deficit financing in Nigeria is due to the unproductive tax system.
5.3 Recommendations
The findings from thus study shows that the situation of the buoyancy of the
Nigeria tax system is not too encouraging as there are opportunities for improvement
in administration and collection of taxes from the existing tax sources. The study
therefore recommends as follows:
First, the government has to broaden the tax base by providing the enabling
environment for private enterprises to thrive. In support of Ariyos (1997) report, the
government should withdraw from several economic activities that the private sector
is able and willing to provide more efficiently. This would control and significantly
reduce public expenditure, which is evidenced by the large number of projects that are
abandoned in the country. When this is done, huge sum of money would be saved to
reduce fiscal deficit in the country.
Secondly, the government should improve on tax administration in order to narrow
the gap between tax collection and remittance to government coffers.
102
Thirdly, government should also ensure that the leakages in the administration of tax
collection and tax revenue remittance are effectively checked.
Fourthly, government should deregulate system substantially so as to enable the
private sector to thrive. This will reduce the level of government expenditure such that
revenue and expenditure are synchronized. This was in line with Ariyo (1997) when
he recommends that government should design meaningful programmes that would
reduce significantly government expenditure thereby curbing the problem of fiscal
deficit and inefficient tax administration. The present administrative structure seems
to be too expensive hence the need for prudent management and productive use of the
nations financial resources.
5.4 Contribution to Knowledge
Several researches such as that of Ariyo and Raheem (1990), Ariyo, (1993), Ariyo,
(1997) have established the non-sustainability of the Nigeria tax System. In doing, so
eight variables (GTR, NOR, IMD, ED, PPT, TOS, NGDP and CIT) were employed to
explain the link between GTR and the various tax sources employed. But in this
research, nine variables (GTR, NOR, CEXD, PPT, TOR, GDP, NGDP, CIT and
TEXP) were used to establish the link between fiscal deficit and the productivity of
the Nigerias tax system. Secondly, none of these researches linked the non-
sustainability of the Nigerias tax system with the increasing fiscal deficit in spite of
the increasing revenue generation attempt by the government. Thus, these two cases
are additional knowledge contributed by this research.
Omoruyi (1983) so far represents the most comprehensive assessment of the
productivity of the Nigeria tax system. He evaluated the buoyancy of the tax system
103
as defined by Sahota (1981) and Ghai (1980) for the period 1960 to 1979. He focused
on both the indirect taxes such as import, export and excise duties, as well as direct
taxes such as personal income tax (federally collected) and petroleum profit tax. Our
study improves upon Omoruyi (1983) in the following respects. First, our study
captures the impact of the structural changes in the macroeconomic management
framework introduced since 1970.
Second, Omoruyi (1983) disaggregated his analysis in terms of decades (1960 1967,
1970 1980, etc.) but we believe that such disaggregation could not provide an
adequate guide for policy decisions, which are of interest to this study. Hence, we
disaggregated our analysis around notable economic events such as the pre-and post-
oil boom era, as well as the impact of SAP on the buoyancy of Nigerias tax system.
104
REFERENCES
Abiola, S. (2002) Tax Incentive and Reforms in Nigeria. A paper presented at a
Mandatory Professional Training Programme (MPTP) of the Chartered
Institute of Taxation of Nigeria.
Adekanola, O. (1997) Efficient Tax Collection and Effective Tax Administration in
Nigeria. (A Paper presented at a Seminar Organized by the University of
Lagos Consultancy Services Otta, Lagos.
Adesola, S.M. (1995) Income Tax Law and Administration in Nigeria, Ile-Ife,
University of Ife Press Ltd.
Aghevli, B.B. and Sassanpour, C. (1982) Prices, Output and the Trade Balance in
Iran. World Development, Vol. 10 791800, obtained online from
http://www.sciencedirect.com/science/article/B6VC6 ...
07bebabf24a662ae2f65
Ajakaiye, D. O., and Odusola, A. F. (1996) Price Effects of Value Added Tax in
Nigeria. NCEMA Policy Analysis Series, 2 (2): 48-68. National Centre for
Economic Management and Administration, Ibadan
Alade, S. (2003), Fiscal Adjustment in Nigeria: Issues in Capital Expenditure, The
Bullion, Vol. 27, Number 2, pp.8-16, Central Bank of Nigeria.
Aluko, R. L. (2004), Macroeconomic Activity and Fiscal Position in Nigeria,
http://www.google.com.ng/#sclient=psyab&hl=en&site=&source=hp&q=Over
view
Anyaduba, J.O. (2004) Partnership Taxation in Nigeria: Some Basic Issue. ICAN
Student Journal 9,2, 15-17.
Anyanwu, J. C. (1997); Nigerian Public Finance, Joanee Educational Publishers
Limited, Onitsha, Nigeria.
Anyanwu JC (1998). Do Large Fiscal Deficits Produce High Interest Rate?: The
Case of Nigeria, Ghana and The Gambia, 1987:31995:4. CBN Financial
Economic Review, Vol 36, No 1: 51-84. Nigeria.
105
Anyanwu JC, Oaikhenam H (1995). Macroeconomics: Theory and Applications.
(1st Ed), Joane Publishers, Onitsha.
Ariyo, A., and Raheem, M.I. (1990) Deficit Financing and Economic
Development: Empirical Perspectives from Nigeria. Project Report
Presented to the African Economic Research Consortium. Abidjan
Ariyo, A. and Raheem, M.I. (1991) Effect of Fiscal Deficit on Some
Macroeconomic Aggregates in Nigeria. Final report presented at the
Workshop of the African Economic Research Consortium. Nairobi,
Kenya.
Ariyo, A. (1993) An Assessment of the Sustainability of Nigerias Fiscal Deficit:
1970 1990". Journal of African Economies, Vol. 2, No. 2: 26382,
obtained online from http://www.oup.co.uk/journals
Ariyo, A. (1997) Productivity of the Nigerian tax system: 19701990. AERC
Research Paper 67. Nairobi: African Economic Research Consortium.
Retrieved June 9, 2009, from http://aercafrica.org/DOCUMENTS/RP67.pdf
Ariyo, A. (1997); Productivity of the Nigerian Tax System: 1970-1990, African
Economic Research Consortium, Nairobi.
Asher, M.G. (1989). Fiscal Systems and Practices in ASEAN: Trends, Impact and
Evaluation. Singapore: Institute of South-East Asian Studies. At a workshop
on Quality of African Data organized by the African Economic Research
Consortium, Nairobi.
Baily, M.J. (1980) Inflation and Tax-Induced Resource Misallocation. National
Tax Journal, vol. 33: 27578, National Tax Association, obtained online
from www.ntanet.org/tax-resources/ntj-full-text-articles.html
Blinder, A.S. and Solow R.M. (1983). Does Fiscal Policy Matter?. Journal of
Public Economics, Vol. 2: 31937, www.journals.elsevier.com/journal-of-
public-economics/
106
Buiter, W. (1988) The Theory of Optimum Deficits and Debt. NBER Working
Paper No. 1232 National Bureau of Economic Research, Inc. obtained online
from www.bostonfed.org/economic/conf/conf27/conf27b.pdf
CBN (Central Bank of Nigeria) (various years), Annual Reports and Statement of
Accounts 1996-2001.
Chibber, A. and Khalizadeh-Shirazi, J. (1988) Public Finance in Adjustment
Programs. World Bank, PPR Working Paper No. 128.
CITN (Chartered Institute of Taxation of Nigeria) (2002). CITN Nigerian Tax
Guide Statutes. Lagos: The Chartered Institute of Taxation of Nigeria.
Corden, W.M. and Nealy J.F. (1982) Booming Sector and De-Industrialization in
a Small Open Economy. Economic Journal, Vol. 92, 825848, obtained
online from academic.research.microsoft.com/Paper/2790554
Diejomaoh, V.P. (1986) Mobilizing Resources for Development: Problems and
Prospects of Import Taxation in Nigeria. In Teriba, O., and V.P. Diejomaoh
(Eds.), Money, Finance and Economic Development, University Press,
Ibadan.
Draft Document on the National Tax Policy, updated as at 7 June 2008.
(Presentation by the Presidential Committee on National Tax Policy).
Egwakhide, F.O. (1988); Analysis of Structural Shifts of Government Revenue in
Nigeria: The Nigerian Journal of Economic and Social Studies, Vol. 30, No.
2. http://books.google.co.za/books/about/NJESR.html?id=7PjTAAAAMAAJ
Ekpo, A.H. and Ndebbio J.E.U. (1992) Fiscal Operations in a Depressed Economy:
Nigeria, 19601990. Final report submitted to the African Economic
Research Consortium, Nairobi.
Ekuerhare, B.U. (1980) The Impact and Lessons of Nigerias Industrial Policy
Under the Military Government: 19661979". In Nigerian Economy
under the Military, Proceedings, 1980 Annual Conference of the
Nigerian Economic Society: 199-223.
107
Ezeabasili, V. N. and Mojekwu, J. N. (2011), Analysis of Fiscal Deficits and
Interest Rates In Nigeria; Journal Of Economics And International Finance,
Vol. 3(4), pp. 236-245, www.academicjournals.org/jeif/index.htm.
Fjeldstad, O. H. (2003). Fighting Fiscal Corruption. Lessons from the Tanzania
Revenue Authority. Public Administration and Development, Vol. 23, No. 2,
pp. 165-175 . Tanzania
Federal Government of Nigeria, (1989a) Finance (Miscellaneous Taxation
Provisions) Decree Nos. 31 and 55 of 1989. Federal Government Printer,
Lagos.
Federal Government of Nigeria, (1989b) Constitution of the Federal Republic of
Nigeria. Federal Government Printer. Lagos.
Federal Inland Revenue Service, (1990) Notes for Guidance on Finance
(Miscellaneous Taxation Provisions) Decree Nos. 31 and 55 of 1989.
Federal Government Printer, Lagos.
Feldstein, M. (1980) Fiscal Policies, Inflation and Capital Formation. American
Economic Review, Vol. 70, September: 63650, National Bureau of
Economic Research (NBER)
Ghai, D.P. (1980) Taxation for Economic Development: A Case Study of Uganda.
Nairobi, East African Publishing House, Kenya:
Gills, M. (1989) Comprehensive Tax Reforms: The Indonesian Experience, 1981-
1988. In Gills, M. (ed.) Tax Reform in Developing Countries Providence:
Duke University Press, Indonesia
Gills, M. (1990) Micro and Macroeconomics of Tax Reform: Indonesia. In
Bird, R.M., and O. Oldman (eds.) Taxation in Developing Countries, 4th ed.
John Hopkins, London.
Gordon, R. J. (2006); Macroeconomics Theory, 10
th
Edition, Addison-Wesley
Publications, Kuwait.
108
Gujarati, D. N, & Porter, D. C (2009); Basic Econometrics, McGraw-Hill
Companies Inc., New York, USA.
Haughton, J. (1998); Estimating Tax Buoyancy Elasticity and Stability,
Eager/PSGE- Excise Project, Methodological Note 1 obtained online from
pdf.usaid.gov/pdf_docs/PNACE024.pd
Herberger, A. C. (1983) Dutch Disease: How Much Sickness, How Much Boom?
Resource and Energy, Vol. 5: 120, obtained on line from
www.sciencedirect.com/science/article/pii/0165057283900154
Hinricks, H. (1986) A General Theory of Tax Structure Change During Economic
Development, Harvard Law School, England
Idachaba, F.S. (1984); Economic Structure and Public Revenue Instability; In
Ijewere, E.I. (1991). Past, Present and Future of Tax Legislation in Nigeria.
Paper presented at a National Workshop on Tax Structure and
Administration in Nigeria, 15-17 Lagos,
Ijewere, E. I. (1991) Past, Present and Future of Tax Legislation in Nigeria.
Paper Presented at a National Workshop on Tax Structure and
Administration in Nigeria, Lagos.
Iyoha, M. A. (2002) An Econometric Model of the Nigerian Economy. In Milton A
Iyoha and Chris O. Itsede (Eds.), The Nigerian economy: Structure, Growth
and Development (pp. 161 185), Mindex Publishers, Benin.
Iyoha, M. A. (2004); Macroeconomics, Theory and Policy, Mindex Publishing,
Benin-City.
Jhingan, M. L. (2005); Macroeconomics Theory, Vrinda Publications (P) Ltd, New
Delhi, India.
Johansen, S. (1988),Statistical Analysis of Cointegration Vectors. Journal of
Economic Dynamics and Control, 23154, obtained from
www.elsevier.com/journals/journal-of-economic-dynamics
109
Johansen, S. (1991). A Statistical Analysis of Cointegration for I(2) Variables,
University of Copenhagen, Copenhagen, Germany
Landau, D. (1983),Government Expenditure and Economic Growth: A Cross-
Country Study, Southern Economic Journal, vol. 49, no. 3, 78392.
econpapers.repec.org/article/sejancoe.
Lipumba, N, H, I and Mbelle, A. V. Y. (1990); Achievements and Setbacks of
Economic Recovery Programme and Policy Reforms for Sustained Growth, a
Paper presented at the 6
th
Economic Policy Workshop, Dares- Salaam.
Mansfield, C. Y. (1992). Elasticity and Buoyancy of Tax System: A Method
Applied to Paraguay. IMF Staff Papers, vol. 29, Paraguay
Massell, B. F., Pearson, S.R. and Fitch, J. B. (1982) Foreign Exchange and
Economic Development: An Empirical Study of Selected Latin American
Countries. Review of Economics and Statistics, Obtained from
www.jbnst.de/en.
Murithi, K.M and Moyi, E. D (2003) Tax Reforms and Revenue Mobilization in
Kenya, EERC Research Paper, 131, African Economic Research
Consortium, Nairobi.
Musgrave, R.A. and P.B. Musgrave, (1984), Public Finance in Theory and Practice,
McGraw-Hill: New York, Ch. 4, 2630, Nairobi
Ndekwu, E.C. (1991)An Analytical Review of Nigerias Tax System and
Administration. Paper presented at a National Workshop on Tax Structure
and Administration in Nigeria, 1517, Lagos.
Newberry, D. and Stern, N. (1987), The Theory of Taxation in Developing Countries,
(Ed)Oxford University Press: Oxford, England.
Nnadi, M. A. and Akpomi, Meg. (2008) The Effect of Taxes on Dividend Policy of
Banks in Nigeria, International Research Journal of Finance and Economics,
obtained online from http://www.eurojournals.com/finance.htm
110
Obi, B. and Nurudeen (2008); Do Fiscal Deficits Raise Interest Rates in Nigeria? A
Vector Autoregression Approach, Journal of Applied Quantitative methods,
Vol. 4 No 3. Otained online from: www.jaqm.ro
Odusola, A. F. (2003) Internally Generated Revenue at the Local Government:
Issues and Challenges. (A Paper presented at the Workshop on Revenue
Generation at the State Government Level, University Press, Ibadan.
Odusola, A. (2006) Tax Policy Reforms in Nigeria, United Nations University,
UNU- WIDER, obtained from www.wider.unu.edu
Okpara, G. C. (2010) Productivity and the Effect of Taxation on Economic Growth in
Nigeria, International Journal of Research in Commerce and Management,
Vol. No. 1, Issue No. 6. Obtained from www.ijrcm.org.in
Olopoenia, R.A. (1991). Fiscal Response to Oil Wealth and Balance of Payments
Performance in Nigeria, 1970-1989. Final report presented at the African
Economic Research Consortiums Workshop, Kenya.
Omoruyi, S.E. (1983) Growth and Flexibility of Federal Government Tax Revenue:
1960-1979. Economic and Financial Review, Vol. 21, No.1: 1119.
www.freedocumentsearch.com/pdf/review-of-international.
Onwioduokit, E. A, (1999) Fiscal Deficits and Inflation Dynamics in Nigeria: An
Empirical Investigation of Causal Relationships, CBN Economic & Financial
Review, Vol. 37 No2 1-16, obtained online from
http://www.cenbank.org/out/publications/efr/rd/1999/efrvol37-2-1.htm
Osoro, N.E., (1985), The Buoyancy and Elasticity of the Tanzania Tax Structure:
1969170-1980/81, Ph.D. Thesis presented to University of Illinois, USA.
Osoro, N.E. (1991) Tax Reform in Tanzania: Motivations, Directions and
Implications. A progress Report presented to the African Economic
Research Consortium, Kenya.
Osoro, N. E. (1993) Revenue productivity implications of tax reform in Tanzania,
African Economic Research Consortium, Kenya.
111
Osunibi, T. S and Olaleru, O. E (2006), Budget Deficits, External Debt and Economic
Growth in Nigeria, Applied Econometrics and International Development,
Vol. 6-3. Obtained from internationalx.net/APPLIED- ECONOMETRICS-pdf
Phillips, A. O. (1991) Formulation of an Effective Tax Policy. Paper presented at
a National Workshop on the Nigerian Tax System and Administration,
Lagos.
Pindyck, R.S. and Rubinfeld, D.L. (1998) Econometric Models and Economic
Forecasts, 2nd. Ed, McGraw-Hill, Japan.
Prest, A.R. (1992) 'The Sensitivity of Yield of Personal Income Tax in the United
Kingdom', Economic Journal, 52, 576-96, U.K
Quantitative Micro Software (1999) E-Views 3.1 Users Guide, 3rd Edition,
United States of America
Rao, V.G. (1990) The Responsiveness of the Indian Tax System, 1969/70 1973/74,
New Delhi, India
Rutayasire, L.W. (1990) Issues in Public Expenditure Policy in Africa: Evidence
from Tanzanian Experience. Paper presented at the World Bank
Conference on African Economic Issues. Nairobi
Sahota, G.S. (1980) Indian Tax Structure and Economic Development. London:
Asia Publishing House, London.
Singer, N .M. (1988); The Use of Dummy Variables in Establishing the Income
Elasticity of State Income Tax Revenues. National Tax Journal, vol. 21.
Lipumba, N. H. I. and Mbelle, A.V.Y. (1990); Achievements and Setbacks of the
Economic Recovery Programme and Policy Reforms for Sustained Economic
Growth. Paper presented at the 6th Economic Policy Workshop, Dares-
Salaam, Uganda
112
Thornton, D. L. (1990); Do Government Deficits Matter? Review, The Federal
Reserve Bank of St. Louis, Vol. 72, No. 5, 2539.
Veez-Zadeh, R. (1989). Oil Wealth and Economic Behaviour: The Case of
Venezuela, 19651981. International Monetary Fund, Staff Papers, Vol. 36,
No. 2: 343383. Washington, DC: The World Bank Group., USA
Wickens, M.R. and Uctum, M. (1990) National Insolvency: A Test of the US
Inter- Temporal Budget Constraint. Centre for Economic Policy Research,
Discussion Paper Series, No. 437, USA
Wonnacott, R.T. and Wonnacott, T.H. (1980), Econometrics, 2nd. ed. New York:
John Wiley and Sons, New York.
World Bank (1995) African Development Indicator, 1994 1995. Washington,
D.C:. World Bank, USA
World Bank (1995), Private Sector Development in Low-Income Countries.
World Bank, Washington, D. C., USA
World Bank (2002). Taxation and State Participation in Nigerias Oil and Gas
Sector, worldbank.org/curated/en/2004/08/5539607/
Yellen, J. L. (1996) Symposium on the Budget Deficit. Journal of Economic
Perspectives, Vol. 3, no. 2: 7393, obtained online from
www.researchgate.net/...Symposium_on_the_Budget_Deficit
Zee, H.H. (1988) The Sustainability and Optimality of Government Debt IMF
Staff Papers, Vol. 35: 65885, www.jstor.org/stable/3867115.