ArticlePDF Available

Fiscal Policy and Economic Growth in South Africa

Authors:

Abstract

The purpose of the paper is to examine the effect of fiscal policy variables on economic growth in South Africa. The fiscal policy variables considered in the study include government gross fixed capital formation, tax expenditure and government consumption expenditure as well as budget deficit. The study covered the period 1990 to 2004. Quarterly data was used in the estimation with the aid of vector regressive modeling technique and impulse response functions. The outcome supports four key conclusions. First, government consumption expenditure has a significant positive effect on economic growth. Gross fixed capital formation from government also has a positive impact on output growth but the size of the impact is less than that attained by consumption expenditure. Tax receipts also have a positive effect on output growth. However, the size of the deficit seems to have no significant impact on growth outcomes.
Fiscal Policy and Economic Growth in South Africa
Matthew Kofi Ocran1
1 Department of Economics, University of Fort Hare, Alice (Main) Campus, Private Mail Bag, X 1314,
Alice 5700. Tel: +27 (40) 602 2128; Cell: 0729077251; Email:
A paper presented at the Centre for the Study of African Economies “Conference on
Economic Development in Africa”, St. Catherine’s College, Oxford University, UK. March
22-24, 2009.
Abstract
The purpose of the paper is to examine the effect of fiscal policy variables on economic
growth in South Africa. The fiscal policy variables considered in the study include
government gross fixed capital formation, tax expenditure and government consumption
expenditure as well as budget deficit. The study covered the period 1990 to 2004. Quarterly
data was used in the estimation with the aid of vector regressive modeling technique and
impulse response functions. The outcome supports four key conclusions. First, government
consumption expenditure has a significant positive effect on economic growth. Gross fixed
capital formation from government also has a positive impact on output growth but the size
of the impact is less than that attained by consumption expenditure. Tax receipts also have a
positive effect on output growth. However, the size of the deficit seems to have no
significant impact on growth outcomes.
JEL Classification: C22, C33, E62, E66
Key words: Revenue, expenditure, growth, VAR, South Africa
MOcran@ufh.ac.za/mkocran@gmail.com.
2
1. Introduction
The intent of fiscal policy is essentially to stimulate economic and social development by
pursuing a policy stance that ensures a sense of balance between taxation, expenditure and
borrowing that is consistent with sustainable growth. However, the extent to which fiscal
policy engender economic growth continue to attract theoretical and empirical debate
especially in developing countries. In recent time the debate has been given more impetus in
South Africa as the unions and the communist party argue that the fiscal policy direction
since the new democracy in 1994 has to make way for a more interventionist approach. The
basis of that agitation is due to the argument that the seemingly steady growth that has been
recorded over the past decade has provided a low number of jobs. Indeed, over the past 10
decade (1995 -2004) real GDP grew at a rate of 3.0% per year. This period has also been
characterised with a more open economy as the country shed its pariah-state status and
deepened her integration in the global market. In addition to the fiscal restraint pursued the
economic policy was largely pro-market with reduced intervention and a considerable
measure of deregulation as compared to the pre-1994 era.
On the theoretical front, however, there are two main strands of literature regarding the role
fiscal policy play in fostering economic growth. One view is that government’s support for
knowledge accumulation, research & development, productive investment, the maintenance
of law and order and the provision of other public goods and services can stimulate growth
in both the short-run and the long run (Easterly and Ribero, 1993; Mauro, 1995; Folster and
Henrekson, 1999). On the other hand, there is also the view that governments are inherently
bureaucratic and less efficient and as a result they tend to hinder rather than facilitate
growth if they get involved in the productive sectors of the economy. Thus government
fiscal policy is thought to stifle economic growth by distorting the effect of tax and
inefficient government spending.
In the light of debate the question that comes to the fore is what has been the effect of fiscal
policy on economic growth in the country over the years? The objective of the present paper
therefore is to contribute to the debate by investigating the effect of fiscal policy on economic
growth in South Africa over the past few decades. The motivation for the paper is that thus
far debate on the efficacy of fiscal policy in stimulating growth seems to have received scant
attention. This paper therefore seeks to contribute to the public discourse on the matter but
from a South African focused empirical effort.
The rest of the paper is structured as follows. Section two provides a brief review of the
evolution of fiscal policy stance over the study period. Previous studies of relevance to the
study are discussed in Section three. The methodology adopted for the estimation and data
issues are considered in Section four. In chapter five we provide the outcomes of the
estimations and conclude in Chapter 6.
3
2. Overview of Fiscal Performance in South Africa
Public financing in South Africa has gone through diverse changes over the past four
decades. Significantly, the post-1994 years witnessed a raft of reforms among these was the
introduction of the medium term expenditure framework programme (MTEF). The MTEF
was undertaken between 1997 and 2000, as part of the programme tax reforms and
administration capacity improvements were carried out. Now, fiscal policy performance in
post-1994 South Africa has been mixed. For instance debt as a percentage of GDP has
increased marginally over the past four decades. After falling from a high of 40% in the
1960s to 32% in the 1980s it started climbing up in the first four years of the 1990s. The post
1994 period returned fiscal management to 1960s debt levels. Similarly, the deficit level that
peaked at -5.4% as a proportion of GDP between 1990 and 1993 was halved by the 1994-
2004 sub-period, a level comparable to average deficit to GDP ratio in the 1960s. Expenditure
as a share of GDP has generally increased over the years from a low of 18.4% in the 1960s to
26.4% in the first decade after 1994.
Table 1. Trends in fiscal policy variables and GDP, 1960 - 2004
Period average 1960-1969 1970-1979 1980-1989 1990-1993 1995-2004
As p
ercentage of
GDP
Debt 44.7% 39.7% 32.4% 39.0% 45.2%
Tax 15.9% 19.2% 22.3% 23.3% 23.8%
Deficit -2.5% -4.5% -3.3% -5.4% -2.5%
Consumption 11.1% 14.2% 17.4% 20.0% 18.7%
Expenditure 18.4% 23.7% 25.6% 28.7% 26.4%
Gross fixed capital
formation
21.1% 26.4% 23.1% 16.7% 15.9%
Real GDP growth 5.8% 3.3% 2.2% -0.6% 3.0%
Source: Data from IFS, 2005; author’s computations.
Consumption expenditure by government has also increased significantly; this was partly
driven by the higher non-wage consumption expenditure mostly on education, health and
their associated supplies and equipment. Real growth in general government spending on
wages and non-wages (this spending on education, health, e.t.c. including transfers to poor
households) are also accountable for the increased levels. With regard to the considerable
increase in consumption expenditure following the end of apartheid, it is worthy to note that
since 1994 government has been providing social grants to a considerable proportion of the
population. Presently, more than 13 million of the population (27%) receives one form of
4
social grant or the other. These transfers include child support, old age pensions, disability,
dependence care and foster care grants respectively. The size of the grants is estimated to
account for 12% of total government spending is reckoned to account for 12% of the total
government expenditure in the 2009/10 financial year2
The body of literature dealing with the effect of fiscal policy on economic performance
outcomes is essentially anchored on two broad positions
(Budget Statement, 2009). While
consumption expenditure is largely seen as unproductive by the literature on fiscal policy
and economic growth it is also asserted in a section of the literature social grants in South
Africa given its history has clear welfare implications. These positive welfare implications
can have a possible favourable effect on long term growth.
However, when gross fixed capital formation is considered, it is noted that government has
been investing at lower levels as compared with say the 1960s. Indeed for most of the pre-
1994 years investments averaged more than 20% of GDP rising to a high of 26.4% in the
1970s. However, the post-1994 years saw a decline to an average of about 16%.
3. Review of the Literature
3
Fundamental to the discussion is the question of representation of fiscal policy. Here, the
literature shows that there are different views as to what variable best captures fiscal stance.
Out of the three standard fiscal policy variables; spending, taxation and deficits the
literature do not single any one of these as the most representative in terms of fiscal policy.
While many papers have made use of tax rates as a proxy for fiscal policy (Lucas, 1990;
Rebelo, 1991; Xu, 1994; Stokely and Rebelo, 1995; Engen and Skiner, 1996) others such as
Martin and Fardmanesh (1990) and Easterly and Rebelo (1993) have used deficits to account
for the fiscal policy in their estimations. Yet other researchers have also used expenditure.
Some of the papers that use expenditure to account for fiscal policy stance include Barro
(1990), Aushauer (1989), Easterly and Rebelo (1993). In a study by Levine and Renelt (1992)
. First, we have the classical
economic view which says that “with every dollar increase in real government spending is
offset by a dollar reduction in private spending, so crowding out is complete (Dornbusch et
al, 1998 as quoted in Kukk, 2008). On the contrary the Keynesian view as represented in
Blinder and Solow (2005) suggest that consumption has a positive effect on the economy.
The proponents of the classical view assert that the effect of government spending is
temporary and not effective particularly in the long-run when prices adjust and output and
employment are at their optimum levels.
2 http://www.treasury.gov.za/documents/national%20budget/2009/guides
3 It is however important to note that within each of the broad views there are other variant positions
though fundamentally similar but with some refinements.
5
the authors argue that none of the three policy variables has a robust association with
economic growth when examined individually. Fu et al (2003) suggest that the inadequacy of
any one of the identified fiscal policy indicators (as pointed by Levine and Renelt, 1992) but
disputed in the mainstream growth literature could be due to the inability of any one fiscal
policy factor to adequately account for a given fiscal policy position.
When expenditure is considered it is observed that while certain studies have considered
aggregate government expenditure as a single variable others have said that the variable
ought to be decomposed into several categories. These categories should then be analysed
separately. What has become increasingly acceptable is the division of government
expenditure into investment and consumption. It is reckoned that the former stimulates
growth while the latter impedes growth. In recent time however, people have gone a step
further to disaggregate consumption into what has been called productive government and
unproductive government expenditure following Devarajan et al (1996). The argument here
is that while certain consumption expenditures particularly those on health, education and
infrastructure4
4 Glomm and Ravikumar (1997) demonstrate that investment and education spending by government
impact directly and positively on the growth rate of the economy.
could foster growth or types of consumption spending can be growth
distorting. Nonetheless, Zagler and Durneker (2003) concede that while certain public
consumptive expenditures may not directly impact on long-term growth they may well have
positive welfare implications in the economy, an argument that is also made in Turnvosky
(1996). When it comes to research and development (R&D) expenditures provided by the
public sector it is expected that R&D spending would stimulate output growth but in the
literature the empirical outcomes are not unanimous in that view (Grossman and Helpman,
1991; Jones, 1995; Morales, 2001).
On the role of taxation the assertion is that tax induced distortions affects private agent’s
allocative decisions unfavorably in terms of factor accumulation and supply and hence may
affect growth. This position is due to the assumption that all taxes save lump-sum taxes are
non-neutral and distortionary. There is also debate about taxation as a short-run fiscal policy
instrument and its effect on long-term growth (Zagler and Durneker, 2003). Here again,
while one group of taxes such as those on savings, R&D, profits, raw capital and labour are
deemed to have direct impact on the growth prospects of an economy all other tax forms are
regarded as inconsequential to growth. The net effect of taxes however, is understood to be
the difference between the positive effects from productive government spending and the
growth distorting (negative) effect of taxation on growth. Indeed there is a vigorous debate
when it comes the decomposition of taxes and how individual tax components impact
economic growth (see Engen and Skinner, 1996; Milesi-Ferreti and Roubini, 1998;
Turnvosky, 2000).
6
The size of the public debt and its effect on growth is also explained by a number of
competing theories. The point here is that when government runs a deficit it tends to draw
on resources that the private sector could have used to accumulate private physical capital.
If government engages in any spending that is less productive as compared to that of the
private sector then we are faced with an overall negative growth effect (Araujo and Martins,
1999). A contrary view is espoused by Lin (2000) and others who take the position that
public debt do not necessarily reduce growth.
In a study by Levine and Renelt (1992) it is observed that none of the three conventional
fiscal policy variables on its own adequately captures the fiscal policy stance of any give
economy. Consequently, a third-generation strand of the literature on fiscal policy and
economic development has emerged that attempts to examine at least two fiscal policy
variables in simultaneously. Some of these studies include, Kocherlakota and Yi (1997), de la
Fuente (1997), Kneller et al (1999), Bleaney et al (2001) and Fu et al (2003).
The literature review amply demonstrates that no single indicator sufficiently represent
fiscal policy stance, however, as suggested by Fu et al (2003) pair-wise combinations of the
fiscal indicators better illustrate a series of fiscal policy actions. Fu et al (ibid) show that pair
wise analysis using a various specifications produces a plausible stable result that lends
itself to interpretation. Consequently, we follow the authors in adopting the vector
regressive (VAR) technique in the empirical analysis.
4. Methodology
The literature on the endogenous growth theory maintains that fiscal policy impact on both
the level and growth rates of output. Barro (1990) and Barro and Sala-i-Martin (1995) are
some of the papers that provide theoretical arguments behind this assertion. The analytical
framework underlying this position is fashioned along the line of the standard Cobb-
Douglas production function (see Amanja and Morrissey, 2005). The assumption here is that
government supported goods and services serves as an input to demonstrate the positive
impact of productive government spending and the adverse outcome due to distortionary
taxes. The production function is given as;
αα
gAky
=1,
10
α
(1)
where,
y
stands for output per capita,
k
denote per capita private capital and
A
is a
measure of productivity. Given that government budget is balanced by imposing a
proportional tax on output at a given rate,
τ
and lump sum taxes,
the budget constraint
can be written formally as;
7
nyLCg
τ
+=+
(2)
The number of producers in the economy is given by,
n
while
C
represents government
consumption which is deemed unproductive. Theoretically, an output linked tax impact on
private incentive to invest but a lump sum tax does not (Barro, 1990). Drawing on a specified
utility function, Barro (1990) and Barro and Sala-i-Martin (1992) derive a long-run growth
rate,
h
model;
µατλ
αα
=
1)1/(1
)/()1)(1( ygAh
(3)
where,
λ
and
µ
denote parameters in the utility function. It is observed from (3) that
growth rate (
h
) is a decreasing function of distortionary tax rate (
τ
) and an increasing
function of the productive government spending (
). Also notable is the fact that,
unproductive consumption (
C
) and non-distortionary taxes (
L
) have no role in equation 3.
Other papers have suggested the need to relax the balanced-budget assumption. Following
the work of Kneller et al (1999) and Bleaney et al (2000) equation 2 is restructured by
including a budget deficit/surplus term,
to the left-hand side of equation is equation. We
then have;
nyLdCg
τ
+=++
(4)
The growth model of the present study is specified in the spirit of Kneller et al (1999) by
considering both fiscal and non-fiscal variables;
itjt
m
jjit
k
itt XYy
εγβα
+++= == 11 (5)
where,
t
y
is the growth rate of output,
jt
X
is a vector of fiscal variables, it
Y, vector of non-
fiscal variables and t
ε
white noise error terms. Assuming that all elements of the budget are
included, in that case we have a balanced budget;
0=
m
jt jt
X,
One element of
X
is dropped to avoid a problem of perfect collinearity. The eliminated
variable is the compensating factor in the budget constraint. Therefore, (5) can be
reformulated as;
itjtm
m
jj
k
iitiit XYy
εγγβα
+++=
== )(
1
11 (6)
8
By testing the hypothesis of a zero coefficient of jt
X
what one is actually is testing is the
null hypothesis that the term 0)( =mj
γγ
rather than 0=
j
γ
. Consequently, the coefficient
on each of the fiscal policy variables is best interpreted as the effect of a unit change in the
related variable offset by a unit change in the omitted category (Kneller et al, 1999: 175)
Analytical framework
The challenge for the estimation process is what variable adequately captures fiscal policy
stance so that it may be included in the empirical equation? As demonstrated in the
literature no single variable appear to be the best variable to estimate in ascertaining the
impact of fiscal variables. Therefore following, Fu et al (2003) we adopt a vector
autoregression (VAR) model for estimating simultaneous shocks to more than one variable
and to use that to investigate unexpected and equivalent structural shocks
for simultaneous, unanticipated and equivalent structural shocks to pair-wise combinations
of selected fiscal variables.
Following, Yu et al (2003) we include both fiscal and monetary policy variables in the VAR.
Thus in addition to the growth indicator for South Africa, t
SA and fiscal policy (
tt
FF 2,1
-
pair wise combinations of government spending, taxes, debt, and deficit). The other
variables considered are the interest ( t
M) a proxy for monetary policy, and the GDP
deflator, (
t
P
). The inclusion of monetary policy variables also allows for a broad
consideration of the range of economic policy variables that may be driving growth or
otherwise. The VAR approach ensures initial shocks are of the same magnitude and again, it
also allows the system to play itself out.
Consequently, we specify a VAR model in its moving-average representation with
cumulative effect on the current level of the variables of current and past structural
shocks/impulses;
tt LAY
υ
)(= (7)
where,
)(LA
denotes matrix lag polynomial with all the parameters that account for the
response over time of the variables of the system to preceding economic disturbances.
and where, T
tttttt
SAFFPMfY ),2,1,,(=
and
T
SAttftfptmtt),,,,( 21
υυυυυυ
=
(8)
In order to identify the set of structural parameters contained in 0
A, the VAR is initially
estimated in its reduced form given by;
9
tktkttt
YBYBYBY
µ
++++= ...........
1211 (9)
the number of lags in the VAR is given by
k
and the variance-covariance matrix of the
reduced-form residuals is also given by,
Σ=),(
T
tt
E
µµ
. The reduced form’s moving
average designation connects the current values of the variables to contemporaneous and
past reduced form residuals;
tt LRY
µ
)(=
(10)
Since the reduced-form residuals do not lend itself to economic interpretation following Yu
et al (2003) equation (1) and (3) are represented as a linear combination of the economic
disturbances as;
tt A
υµ
0
= (11)
Therefore, the variance-covariance matrix of the reduced form shocks may be linked to the
structural shocks as follows;
Σ== TT
tt AAE 00
)(
µµ
(12)
where,
)(
T
tt
E
υυ
=
stands for the variance-covariance matrix of the structural disturbance.
First a four variable VAR is used to the simultaneously assessing the impact of shocks to
more than one variable in the system. The standard Choleski decomposition is used to
identify the structural shocks. The shocks are then normalized to one standard deviation of
the structural form disturbances in the VAR system.
Data Issues
The data used in the study cover the period 1990 to 2004. The data is quarterly in frequency.
The fiscal policy variables considered are; aggregate government expenditure is used to
account for government spending (GOVSPEND); gross fixed capital formation from
government is used to account for investment expenditure (GOVIN); DEFICIT stands for
government deficit or surplus. Total revenue and grants is used as a proxy for taxes (TAX),
this path was used as it was difficult to obtain time series data of consolidated government
tax collections. However, given that the level of grants as share GDP in South Africa is
insignificant, total revenue and grants is considered as a good approximation of total tax
that accrues to government. The GDP figures are seasonally adjusted. The 30-day Treasury
bill rate is used as a proxy for monetary policy, also considered is the GDP deflator to reflect
price level in the economy. Even though, the 30-day Treasury bill rate is not necessarily the
10
policy rate, it has been closely linked with the policy rate historically. All data series were
drawn from the International Monetary Fund’s International Financial Statistics CD Rom for
2006.
All the data series were tested for stationarity to forestall the possibility of drawing
conclusions based on statistically spurious relationship. The unit root test results are
presented in Table 2.
The first stage of the empirical analyses involved examination of the statistical properties of
the natural logarithms of all the variables under consideration, i.e., real GDP, consumption,
tax, deficit and investment. The other variables are 90-day Treasury bill rate and the GDP
deflator. The results of the ADF unit root tests are summarised in Table 1 below. The results
suggest that the null hypothesis of the presence of unit root in the variables in levels could
not be rejected indicating that all the variables are non-stationary in levels. However, after
first-differencing the variables the null hypothesis of the unit root in each of the series was
rejected at the 1% level of significance. Therefore it can be inferred that all the variables are
integrated of order 1, I (1).
5. Main Findings
The Generalized Impulses as defined by Pesaran and Shin (1998) generates an orthogonal set
of innovations that does not depend on the VAR ordering. The generalized impulse
responses from an innovation to the j-th variable are derived by applying a variable specific
Cholesky factor estimated with the j-th variable at the top of the Cholesky ordering.
Consequently, we adopt the Generalised Impulses in the estimation of the responses of the
identified policy variables as a result of shocks from one another.
The results of the impulse response analyses are presented in charts 1 to 5. Charts 1 to 3
represent VAR models with two fiscal policy variables, one monetary policy variable as well
as output. What we have in chart 1 to 3 is therefore various permutations of interest rate,
investment and consumption with output entering all the models. In charts 4 and 5 we
introduce deficit and drop tax as one of the two fiscal variables.
We draw inferences following the interpretation of transmission effects of impulse response
functions (IRF) due to Lutekpol and Reimers (1992). The said authors distinguish between
permanent and transitory one-time impulse response as a result of shock from one variable
to the other or to the variable itself. If a given shock generates a response path that returns
to its previous equilibrium value of zero after some period then it is referred to as temporary
and permanent if the response path does not return to the initial equilibrium.
11
Model 1
The first policy simulation considered is represented by chart 1. A generalised impulse from
each of the variables (output, interest rate, tax and consumption) to the other is evaluated. It
is observed that the response to own shock (self response) for all the variables flattens after
15 quarters with consumption having the shortest time path as equilibrium is attained after
about 15 quarters. However, the impact of tax on itself takes a relatively longer period to
play out as compared to the other own shocks. The response of output changes to own
shocks takes as twice the time required for own response in consumption to die out.
Considering the innovations from interest rate shocks, the response of output is largely
negative. Impulse response from output growth decline reaches a trough in a relatively
short time (3rd quarter) and then moves towards the initial equilibrium after 30 quarters. The
shocks generated by innovations in tax increases produce a positive response from output
growth. The growth in output hit the highest point by quarter 3 and then returns to
equilibrium over the next 27 quarters. In the case of consumption growth shocks, output
growth response is ambiguous nevertheless the response returns to equilibrium by the 20th
quarter.
On the other hand the response of changes in interest rates to own shock, output growth,
increases in tax and consumption are small in size and tend to move towards the
equilibrium within 10 quarters. Nonetheless, the response from interest rate changes is
negative with respect to unexpected increases in consumption growth (see row 2 of chart 1).
Unlike the responses from the variables in the VAR to shocks from the others, in the case of
changes in tax the response to shocks from the other variables is largely cyclical but moves
towards equilibrium only after 50 quarters (long term). The response of changes in
consumption to own shocks and shocks from the other variables in the VAR system is also
cyclical and dies out completely by the 15th quarter. Significant, though is the quick negative
response from consumption growth due to increases in the interest rate changes. The
negative consumption growth bottoms out within the second quarter but returns to
equilibrium after 15 quarters.
Model 2
The second model focuses on a VAR with output, interest rate, investment and consumption
as components. The own response of output shocks is persistent and lingers for almost 40
quarters. In the case of shocks emanating from interest rate innovations the response of
output growth is negative. The impulse response reaches a peak within 5 quarters and dies
out after a period of 45 quarters. The impulse response returns to the initial equilibrium 25
quarters after over-shooting. In terms of output growth response to shocks from innovations
due to changes in consumption is almost immediate hitting a peak by the second quarter
and then eases slowly until the effects dies out by the 10th quarter (see row 1 chart 2).
12
Generally, the impulse response of interest rate to innovations from all the other variables
included in the model is relatively smaller and lasts for less than 15 quarters. The response
due to shocks from increases in gross capital formation is positive. While the own shock
response of interest rate change is negative the impulse response returns to equilibrium
within 10 quarters. The direction of the impulse response from interest rate changes due to
unexpected increases in investment expenditure has an immediate negative effect that drops
to the initial equilibrium before the 10 quarter. The impulse response of interest rate changes
to shocks from government consumption expenditure is negative and bottoms out after
quarter 2 and dies out after a relatively shorter period (see row 2 of chart 2).
The impulse response of government investment to shocks from increases in output growth
is positive but small with an impulse response of less than 1%. Again, while the impulse
response of changes in interest rate and consumption to increase in government investment
expenditure are negative the size of the response is less than 1%. While the impulse response
of investment to shocks from interest rate changes is negative it is also protracted lasting for
50 quarters.
The impulse response of changes in government consumption expenditure to shocks from
output, interest rate, investment as well as own shocks is negative. The response is also very
small and dies out within a very short time horizon.
Model 3
In the third model the variables that enter the VAR are output, interest rate, investment and
tax. Here, the impulse response of output growth to shocks from changes in interest rate is
negative. The impulse response reaches a trough at quarter 3. However, the size of the
impact is huge as compared to the impulse responses from the other policy variables in the
VAR. On the contrary, the impulse response of output growth to shocks from innovations in
changes in investment is positive. The impulse response dies out 8 quarters after over-
shooting. The response of output growth to increases in tax is also positive but returns to
equilibrium after the 40th quarter.
The impulse response of interest rate changes to output growth shocks is small and cyclical,
it measures about 1%. The responses of interest rate changes to innovations from changes in
investment and interest rate itself are immediate measuring 8% and 4% new respectively.
The time horizon over which the effect pans out is short (i.e., less than 10 quarters, see row 2
of chart 3). The impulse response of changes in tax receipts to shocks from changes in
interest rate is small in magnitude and short in time horizon.
13
The response of changes in investment to the other set of policy variables is generally drawn
out albeit temporary. For instance, the impulse response of changes in investment
expenditure to output growth shocks is positive but small in size as the highest point of the
over-shooting is around 1%. The impulse response drops after the 5th quarter and finally
tending to zero by the 40th quarter. The impulse response of changes in investment to shocks
from innovations from interest rate changes is negative and less than 1%. The response
nonetheless moves back to equilibrium after 35 quarters. Both the response to own shock
and changes in tax are small and momentary.
Generally, the impulse response of changes in tax to output growth, investment, interest rate
and own shocks is cyclical. The impulse response reaches the highest point (4%) in all
instances; though the impulse response is temporary it is persistent. Equilibrium in the
impulse response is restored in all cases only after 50 quarters.
Model 4
The fourth VAR model used in the analysis of fiscal policy effects on economic growth has
the following variables in the system; changes in output, interest rate, investment and
deficits respectively. Significantly, in the present case unlike the VARs without changes in
the deficit as a component, the time-path of all the impulse response outcomes following
innovations from one variable or the other are permanent in nature but the size of the
response is very small. However, when the individual cumulative effects are considered we
see significant impulse response of own shocks regarding output growth. The impulse
response of interest rates to output growth shocks is small but permanent. We also see a
relatively bigger impulse response of investment to output growth shocks (see column one
of chart 4, second panel).
Model 5
Model five also is implemented with a four variable VAR involving changes in output,
interest rates, investment and deficit. Like model 4 the impulse responses are quite weak
and they do deviate marginally if at all from equilibrium. This was the case for all the
impulse response outcomes obtained. The results from the cumulated impulse responses
were not any different. It appears therefore that changes in the deficit on its own thus far is a
bland policy variable regarding its effect on output growth.
14
6. Conclusion
The paper sought to examine the relationship between a selection of fiscal policy variables
and economic growth. Vector autoregression model models were used to estimate
individually the effects of government consumption and investment expenditure, deficit and
tax receipts on economic growth respectively. The analyses covered the period 1990 to 2004
using quarterly data.
The outcome supports four broad conclusions. First, government consumption expenditure
has a significant positive effect on economic growth. Gross fixed capital formation from
government also has a positive impact on output growth but the size of the impact is less
than that attained by consumption expenditure. Tax receipts also have a positive effect on
output growth. However, the size of the deficit seems to have no significant impact on
growth outcomes.
The results of the study stand in sharp contrast to studies on developed economies such as
the work of Fu et al, 2003. While many of the papers on developed economies are consistent
with the orthodoxy as they usually observe that growth in the size of government impedes
economic growth in the present study the outcome is otherwise. Nevertheless, the result of
the present study has an intuitive appeal as well given by the important role of government
in correcting the massive socio-economic imbalances that continue to hold back a large
section of the population. The policy lesson that can be distilled from the findings is that a
continued sensible use of consumption and investment expenditure as policy tools can
speed up growth as compared to a reduction in the size of government.
15
List of References
Araujo, J.T. and Marins, M. A. C., (1999), “Economic growth with finite lifetimes”, Economic
Letters 62, 377-381.
Aschauer, D. A., (1989) Is Public Expenditure Productive?, Journal of Monetary Economics 23,
177200.
Barro, R.J. and Sala-I-Martin, X. (1995), Economic Growth, McGraw-Hill.
Barro, R.J. (1990), “Government spending in a simple model of endogenous growth model”,
Journal of Political Economy, 98(2):103-25
Barro, R.J and Sala-i-Martin, X. (1992), “Public finance in models of economic growth”,
Review of Economic Studies, 59:645-661.
Bleaney, M. Gemmel, N. and Kneller, R. (2000), “Testing the endogenous growth model:
Public expenditure, taxation and growth over the long-run”, Discussion Paper No. 00/25,
University of Nottingham.
Blinder, A.S. and Solow, R.M., (2005), “Does fiscal policy matter? In A. Bagchi, ed., Readings
in Public Finance. New Delhi:Oxford University Press, 283-300
de la Fuerente, A., (1997), Fiscal policy and growth in the OECD. London: Centre for Policy
Research.
Deverajan, D., Swaroop, V. and Zou, H. (1993), “What do governments buy? The
composition of public spending and economic performance”, Policy Research Working
Paper WPS1082. The World Bank.
Devarajan, S.V., Swaroop, V. and Zhou, H., (1996), “The composition of public expenditure
and economic growth”, Journal of Monetary Economics, 37, 313-344.
Dornbusch, R.S., Fischer, R. and Starz, R., (1998), Macroeconomic, USA: The McGraw-Hill
Companies, Inc.
Easterly, W. and Rebelo, S. (1993), “Fiscal policy and economic development: An empirical
investigation”, Journal of Monetary Economics, 32:417-458.
16
Engen, E. and Skinner, J. (1996), “Taxation and economic growth”, National Tax Journal, 49
(December):617-41
Folster, S. and M. Henrekson (2001), ‘Growth Effects of Government Expenditure and
Taxation in rich Countries’, European Journal of Political Economy, 45: 1501
1520.
Fu, D., Taylor, L.L. and Yucel, M.K. (2003), “Fiscal policy and growth”, Research
Department Working Paper 0301, Federal Reserve Bank of Dallas.
Glomm, G. and Ravikumar, B. (1997), Productive government expenditures and long-run
growth”, Journal of Economic Development and Control, 21:183-204.
Grossman, G. M. and Helpman, E., (1991) Quality Ladders in the Theory of Growth,
Review of Economic Studies 58, 43–61.
IMF, (2005), International financial statistics, CD Rom, Washington, D.C.: International
Monetary Fund.
Jones, C., (1995) R&D-Based Models of Economic Growth, Journal of Political Economy
103, 759–784.
Kneller, R., Bleaney, M.F. and Gemmel, N. (1999), “Fiscal policy and growth: Evidence from
the OECD countries”, Journal of Public Economics, 74:171-190.
Kocherlakota, N., Yi, K-M. (1997), “Is there endogenous long-run growth? Evidence from the
US and the UK”, Journal of Money, Credit and Banking, 29 (2), 235-262.
Kukk, K. (2008), Fiscal policy effects on economic growth: Short-run vs long run, TTUWPE
No. 167, Department of Economics, Tallinn University of Technology, Estonia.
Levine, R. Renelt, D. (1992), “A sensitivity analysis of cross-country growth regressions”,
American Economic Review, 82(5), 942-963.
Lin, S., (2000), “Government debt and economic growth in an overlapping generations
model, Southern Economic Journal, 66(3), 754-763.
Mauro, Paolo (1995), ‘Corruption and Growth’, Quarterly Journal of Economics, 110: 681
712.
17
Martin, R. and Fardmanesh, M. (1990), “Fiscal variables and growth: A cross-sectional
analysis”, Public Choice, 64 (3): 239-252
Miles-Ferretti, G. M. and Roubini, N., (1998), “Growth effects of income and consumption
taxes, Journal of Money”, Credit and Banking 30(4).
Morales, M. F., (2001) Research Policy and Endogenous Growth, Universitat Auto`noma
de Barcelona and Universitat de Murcia.
M’Amanja, D. and Morrisey, O., (2005), Fiscal policy and economic growth in Kenya,
CREDIT Research Paper, No. 05/06, Centre for Research in Economic Development and
International Trade, University of Nottingham.
Republic of South Africa, (2009) National Budget Statement, National Treasury
http://www.finance.gov.za/documents/mtbps/2005/mtbps/chapter%203.pdf, accessed on
March 2009.
Rebelo, Sergio (1991). “Long-Run Policy Analysis and Long-Run Growth,” Journal of Political
Economy, 99 (June).
Stokey, N. and Rebelo, S., (1995), “Growth Effects of Flat-Rate Taxes,” Journal of Political
Economy, 103 (June).
Trostel, P. A., (1993), “The Effect of Taxation on Human Capital”, Journal of Political
Economy, 101, No. 21.
Turnovsky, S. J., (1996) Optimal Tax, Dept and Expenditures Policies in a Growing
Economy, Journal of Public Economics 60, 21–44.
Turnovsky, S. J., (2000) Fiscal Policy, Elastic Labour Supply, and Endogenous Growth,
Journal of Monetary Economics 45, 185–210.
Xu, Bin (1994). “Tax Policy Implications in Endogeneous Growth Models,” IMF Working
Paper 94/38, Washington, D.C: International Monetary Fund.
Zagler, M and Durnker, G., (2003), “Fiscal policy and economic growth”, Journal of Economic
Surveys, 17(3)
18
Appendix Table 2. The results of various unit root tests
Variable Levels First difference
Real output 0.308 -6.561***
Consumption -1.881 17.161***
Tax -1.677 -8.935***
Deficit -1.115 -16.329***
Investment -0.996 -6.174***
Interest rate -2.339 -7.542***
19
Chart 1: Impulse Responses VAR (4) Model Output, Interest rate, Tax, Consumption
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DRGDP
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLTBILL
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLTAX
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLCONSUMPT ION
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DRGDP
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLTBILL
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLTAX
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLCONSUMPTION
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Response of DLTAX to DRGDP
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Response of DLTAX to DLTBILL
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Response of DLTAX to DLTAX
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Response of DLTAX to DLCONSUMPTION
-.01
.00
.01
.02
.03
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DRGDP
-.01
.00
.01
.02
.03
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DLTBILL
-.01
.00
.01
.02
.03
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DLTAX
-.01
.00
.01
.02
.03
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DLCONSUMPTION
Response to Generalized One S.D. Innovat ions ± 2 S.E.
-200
-100
0
100
200
300
400
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DRGDP
-200
-100
0
100
200
300
400
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLTBILL
-200
-100
0
100
200
300
400
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP t o DLTAX
-200
-100
0
100
200
300
400
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP t o DLCONSUMPT ION
-.15
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBI LL to DRGDP
-.15
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLTBILL
-.15
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL t o DLTAX
-.15
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLCONSUMPT ION
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTAX to DRGDP
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTAX t o DLTBILL
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTAX to DLTAX
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTAX to DLCONSUMPTI ON
-.02
-.01
.00
.01
.02
.03
.04
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTIO N to DRGDP
-.02
-.01
.00
.01
.02
.03
.04
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTIO N to DLTBILL
-.02
-.01
.00
.01
.02
.03
.04
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTION t o DLTAX
-.02
-.01
.00
.01
.02
.03
.04
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTION t o DLCONSUMPT ION
Acc umulated Respons e to Generali zed One S.D. Innovations ± 2 S.E.
20
Chart 2: Impulse Responses VAR (4) Model Output, Interest rate, Investment, Consumption
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DRGDP
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLTBILL
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLGFCF
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLCONSUMPT ION
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DRGDP
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLTBILL
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLGFCF
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLCONSUMPTION
-.01
.00
.01
.02
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DRGDP
-.01
.00
.01
.02
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DLTBILL
-.01
.00
.01
.02
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DLGFCF
-.01
.00
.01
.02
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DLCONSUMPTION
-.02
-.01
.00
.01
.02
.03
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DRGDP
-.02
-.01
.00
.01
.02
.03
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DLTBILL
-.02
-.01
.00
.01
.02
.03
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DLGFCF
-.02
-.01
.00
.01
.02
.03
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DLCONSUMPTION
Response to Generalized One S.D. Innovat ions ± 2 S.E.
-400
-300
-200
-100
0
100
200
300
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DRGDP
-400
-300
-200
-100
0
100
200
300
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLTBILL
-400
-300
-200
-100
0
100
200
300
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLGFCF
-400
-300
-200
-100
0
100
200
300
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLCONSUMPTION
-.15
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBI LL to DRGDP
-.15
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL t o DLTBILL
-.15
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL t o DLGFCF
-.15
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLCONSUMPT ION
-.12
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Accumulated Response of DLGF CF to DRGDP
-.12
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Accumulated Response of DLGFCF to DLTBILL
-.12
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Accumulated Response of DLGFCF to DLGF CF
-.12
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Accumulated Response of DLGFCF t o DLCONSUMPTION
-.03
-.02
-.01
.00
.01
.02
.03
.04
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTIO N to DRGDP
-.03
-.02
-.01
.00
.01
.02
.03
.04
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTION to DLTBILL
-.03
-.02
-.01
.00
.01
.02
.03
.04
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTION to DLGFCF
-.03
-.02
-.01
.00
.01
.02
.03
.04
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTION to DLCONSUMPT ION
Accumul ated Response to Generalized One S.D. Innovations ± 2 S.E.
21
Chart 3: Impulse Responses VAR (4) Model Output, Interest rate, Investment, Tax
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGD P to DRGDP
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DR GDP to DLTBILL
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGD P to DLGFCF
-40
-30
-20
-10
0
10
20
30
40
50
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLTAX
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DRGDP
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLTBILL
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLGFCF
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLTAX
-.01
.00
.01
.02
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DRGD P
-.01
.00
.01
.02
510 15 20 25 30 35 40 45 50
Response of DLGFC F to DLTBILL
-.01
.00
.01
.02
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DLGFCF
-.01
.00
.01
.02
510 15 20 25 30 35 40 45 50
Response of D LGFCF to DLTAX
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Response of DLTAX to DRGDP
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Response of DLTAX to DLTBILL
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Response of D LTAX to DLGFCF
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Response of DLTAX to DLTAX
Response to Generalized One S.D. Innovations ± 2 S.E.
-300
-200
-100
0
100
200
300
400
510 15 20 25 30 35 40 45 50
Accumulated Response of DR GDP to DRGDP
-300
-200
-100
0
100
200
300
400
510 15 20 25 30 35 40 45 50
Accumulated Response of D RGDP to DLTBILL
-300
-200
-100
0
100
200
300
400
510 15 20 25 30 35 40 45 50
Accumulated Response of DR GDP to DLGFCF
-300
-200
-100
0
100
200
300
400
510 15 20 25 30 35 40 45 50
Accumulated Response of DR GDP to DLTAX
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of D LTBILL to DRGDP
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLTBILL
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLGFCF
-.10
-.05
.00
.05
.10
.15
.20
510 15 20 25 30 35 40 45 50
Accumulated Response of D LTBILL to DLTAX
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Accumulated Response of DLGFC F to DRGD P
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Accumulated Response of DLGFCF to D LTBILL
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Accumulated R esponse of D LGFCF to DLGFCF
-.08
-.04
.00
.04
.08
.12
510 15 20 25 30 35 40 45 50
Accumulated Response of DLGFC F to DLTAX
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTAX to DRGDP
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Accumulated Response of D LTAX to DLTBILL
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTAX to DLGFCF
-.06
-.04
-.02
.00
.02
.04
.06
.08
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTAX to DLTAX
Accumul ated Response to Generali zed One S.D. Innovation s ± 2 S.E.
22
Chart 4: Impulse Responses VAR (4) Model Output, Interest rate, Investment, Deficit
-1200
-800
-400
0
400
800
1200
510 15 20 25 30 35 40 45 50
Response of DRGDP to DRGDP
-1200
-800
-400
0
400
800
1200
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLTBILL
-1200
-800
-400
0
400
800
1200
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLGFCF
-1200
-800
-400
0
400
800
1200
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLDEFICIT
-3
-2
-1
0
1
2
3
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DRGDP
-3
-2
-1
0
1
2
3
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLTBILL
-3
-2
-1
0
1
2
3
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLGFCF
-3
-2
-1
0
1
2
3
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLDEFIC IT
-.6
-.4
-.2
.0
.2
.4
.6
.8
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DRGDP
-.6
-.4
-.2
.0
.2
.4
.6
.8
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DLTBILL
-.6
-.4
-.2
.0
.2
.4
.6
.8
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DLGFCF
-.6
-.4
-.2
.0
.2
.4
.6
.8
510 15 20 25 30 35 40 45 50
Response of DLGFCF to DLDEFICIT
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Response of DLDEFICIT to DRGDP
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Response of DLDEFICI T to DLTBILL
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Response of DLDEFICIT to DLGFCF
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Response of DLDEFICIT to DLDEFICI T
Response to Generalized One S.D. Innovations ± 2 S.E.
-800
-400
0
400
800
510 15 20 25 30 35 40 45 50
Accumulat ed Response of DRGDP to DRGDP
-800
-400
0
400
800
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLTBILL
-800
-400
0
400
800
510 15 20 25 30 35 40 45 50
Accumulat ed Response of DRGDP to DLGFCF
-800
-400
0
400
800
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLDEFICIT
-1 .6
-1 .2
-0 .8
-0 .4
0.0
0.4
0.8
1.2
1.6
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DRGDP
-1 .6
-1 .2
-0 .8
-0 .4
0.0
0.4
0.8
1.2
1.6
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLTBILL
-1 .6
-1 .2
-0 .8
-0 .4
0.0
0.4
0.8
1.2
1.6
510 15 20 25 30 35 40 45 50
Accumulat ed Response of DLTBILL to DLGFCF
-1 .6
-1 .2
-0 .8
-0 .4
0.0
0.4
0.8
1.2
1.6
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLDEFICIT
-.4
-.3
-.2
-.1
.0
.1
.2
.3
.4
510 15 20 25 30 35 40 45 50
Accumulat ed Response of DLGFCF to DRGDP
-.4
-.3
-.2
-.1
.0
.1
.2
.3
.4
510 15 20 25 30 35 40 45 50
Accumulat ed Response of DLGFCF to DLTBILL
-.4
-.3
-.2
-.1
.0
.1
.2
.3
.4
510 15 20 25 30 35 40 45 50
Accumulat ed Response of DLGFCF to DLGFCF
-.4
-.3
-.2
-.1
.0
.1
.2
.3
.4
510 15 20 25 30 35 40 45 50
Accumulat ed Response of DLGFCF to DLDEFICIT
-15
-10
-5
0
5
10
15
510 15 20 25 30 35 40 45 50
Accumulated Response of DLDEFICIT to DRGDP
-15
-10
-5
0
5
10
15
510 15 20 25 30 35 40 45 50
Accumulated Response of DLDEFICIT to DLTBILL
-15
-10
-5
0
5
10
15
510 15 20 25 30 35 40 45 50
Accumulat ed Response of DLDEFICIT to DLGFCF
-15
-10
-5
0
5
10
15
510 15 20 25 30 35 40 45 50
Accumulated Response of DLDEFICIT to DLDEFICIT
Accumulated Response to Generalized One S.D. Innovations ± 2 S.E.
23
Chart 5: Impulse Responses VAR (4) Model Output, Interest rate, Consumption, Deficit
-1500
-1000
-500
0
500
1000
1500
510 15 20 25 30 35 40 45 50
Response of DRGDP to DRGDP
-1500
-1000
-500
0
500
1000
1500
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLTBILL
-1500
-1000
-500
0
500
1000
1500
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLCONSUMPTI ON
-1500
-1000
-500
0
500
1000
1500
510 15 20 25 30 35 40 45 50
Response of DRGDP to DLDEFICIT
-4
-3
-2
-1
0
1
2
3
4
5
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DRGDP
-4
-3
-2
-1
0
1
2
3
4
5
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLTBILL
-4
-3
-2
-1
0
1
2
3
4
5
510 15 20 25 30 35 40 45 50
Response of DLT BILL to DLCONSUMPTION
-4
-3
-2
-1
0
1
2
3
4
5
510 15 20 25 30 35 40 45 50
Response of DLTBILL to DLDEFICIT
-1 .2
-0 .8
-0 .4
0.0
0.4
0.8
1.2
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTI ON to DRGDP
-1 .2
-0 .8
-0 .4
0.0
0.4
0.8
1.2
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPT ION to DLTBILL
-1 .2
-0 .8
-0 .4
0.0
0.4
0.8
1.2
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTI ON to DLCONSUMPTION
-1 .2
-0 .8
-0 .4
0.0
0.4
0.8
1.2
510 15 20 25 30 35 40 45 50
Response of DLCONSUMPTION to DLDEFICIT
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Response of DLDEFICIT to DRGDP
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Response of DLDEFICIT to DLTBILL
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Response of DLDEFICIT to DLCONSUMPTION
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Response of DLDEFICIT to DLDEFICIT
Response to Generalized One S.D. Innov ations ± 2 S.E.
-3000
-2000
-1000
0
1000
2000
3000
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP t o DRGDP
-3000
-2000
-1000
0
1000
2000
3000
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLTBILL
-3000
-2000
-1000
0
1000
2000
3000
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLCONSUMPTIO N
-3000
-2000
-1000
0
1000
2000
3000
510 15 20 25 30 35 40 45 50
Accumulated Response of DRGDP to DLDEFICIT
-8
-4
0
4
8
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DRGDP
-8
-4
0
4
8
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLTBILL
-8
-4
0
4
8
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLCONSUMPTION
-8
-4
0
4
8
510 15 20 25 30 35 40 45 50
Accumulated Response of DLTBILL to DLDEFICIT
-2
-1
0
1
2
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTION to DRGDP
-2
-1
0
1
2
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTION to DLTBILL
-2
-1
0
1
2
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTIO N to DLCONSUMPTION
-2
-1
0
1
2
510 15 20 25 30 35 40 45 50
Accumulated Response of DLCONSUMPTION to DLDEFICIT
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLDEFICIT to DRGDP
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLDEFICIT to DLTBILL
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLDEFICIT to DLCONSUMPTION
-20
-10
0
10
20
510 15 20 25 30 35 40 45 50
Accumulated Response of DLDEFICIT to DLDEFICIT
Accumulated Response to Generali zed One S.D. Innovations ± 2 S.E.
... Total output will be reduced (Mdanat et al., 2018). On the other hand, taxation was found in numerous studies to have a favorable impact on output (Babatunde, Ibukun, & Oyeyemi, 2017;Ocran, 2011;Tosun & Abizadeh, 2005). ...
Article
Full-text available
Purpose: Governments rely on taxes, but a high tax rate can slow economic growth. Fiscal policy objectives can be achieved most effectively by lowering tax collection costs and boosting economic growth through efficient taxation. This study aims to find the impact of tax revenue on Pakistan's economic growth. Design/Methodology/Approach: The time series dataset spanning 1985-2021 is used for the current analysis. GDP is used as the dependent variable, while tax revenue and other fiscal policy variables like government spending, inflation, gross fixed capital formation and current account balance are used as the explanatory variables. The stationarity of the data is checked using the ADF test. The results of the ARDL bound test, which is used to determine whether there is a long-term link between the variables and a short-term relationship, indicate a long-term relationship. Findings: Current analysis reveals that tax revenue and inflation have a negative and significant impact while government expenditures and gross fixed capital formation have a positive and significant impact on the economic growth of Pakistan. Implications/Originality/Value: According to the study's results, Pakistan's government should ensure that tax rates are set at the right level to bring in enough money to pay for government spending that helps the economy grow.
... Total output will be reduced (Mdanat et al., 2018). On the other hand, taxation was found in numerous studies to have a favorable impact on output (Babatunde, Ibukun, & Oyeyemi, 2017;Ocran, 2011;Tosun & Abizadeh, 2005). ...
Article
Full-text available
Purpose: Governments rely on taxes, but a high tax rate can slow economic growth. Fiscal policy objectives can be achieved most effectively by lowering tax collection costs and boosting economic growth through efficient taxation. This study aims to find the impact of tax revenue on Pakistan's economic growth. Design/Methodology/Approach: The time series dataset spanning 1985-2021 is used for the current analysis. GDP is used as the dependent variable, while tax revenue and other fiscal policy variables like government spending, inflation, gross fixed capital formation and current account balance are used as the explanatory variables. The stationarity of the data is checked using the ADF test. The results of the ARDL bound test, which is used to determine whether there is a long-term link between the variables and a short-term relationship, indicate a long-term relationship. Findings: Current analysis reveals that tax revenue and inflation have a negative and significant impact while government expenditures and gross fixed capital formation have a positive and significant impact on the economic growth of Pakistan. Implications/Originality/Value: According to the study's results, Pakistan's government should ensure that tax rates are set at the right level to bring in enough money to pay for government spending that helps the economy grow.
... Government consumption expenditure (Govexpense): Some research discovered that the ratio of government consumption expenditure to GDP can be used to measure the size of a country's government when studying the relationship between trade openness and government size [56]. Furthermore, countries with higher fiscal expenditures have larger governments, with government consumption expenditure being a major indicator [57]. Inspired by previous studies, in this study, we used the ratio of government consumption expenditure to GDP to measure government size. ...
The expansion of government size will have dual effects on a country's green innovation. An appropriately sized government size increases marginal productivity and stimulates the development of green innovation by increasing government expenditure. On the contrary, an excessively sized government creates a huge administrative agency, which not only increases the tax burden but also damages social welfare by excessive intervention. Therefore, the effect of government size on green innovation is not linear. In order to prove this proposition, this study examines the impact of government size on green innovation in 166 countries between 1995 and 2018, using a two-way fixed effects model. The results reveal an inverted U-shaped relationship between government size and the level of green innovation, indicating that optimal government size may maximize a country's green innovation output. The results further suggest that this inverted U-shaped relationship is mainly influenced by environmental regulations and financial support. Finally, our heterogeneity analysis demonstrates that the inverted U-shaped relationship is more pronounced for countries with high organizational inertia and more R&D expenditure than for those with low organizational inertia and less R&D expenditure. This finding makes up for the research gap between government size and green innovation and provides a reference for countries to formulate the optimal government size to improve the level of green innovation.
... In South Africa, there are few studies on the relationship between fiscal policy and FDI. Until recently, most research has focused on the relationship between fiscal policy and economic growth (see, Kofi, 2011;Makhoba, Kaseeram and Greyling, 2019;Burger and Calitz, 2020). ...
Article
Full-text available
A key objective of this study is to examine the symmetric effects of fiscal policy on foreign direct investment (FDI) inflows into South Africa using time series data from 1980 to 2020. The Augmented Dickey-Fuller (ADF) unit root test was conducted to establish stationarity of the variables. To scrutinize the symmetric link between FDI, tax revenue (TAX), government expenditure (GEX), gross fixed capital formation (GFCF), government debt (GDE), and debt service (DES), the linear autoregressive distributed lag (ARDL) bound testing method was applied, which exhibits the long-run and short-run links between the variables. ADF results stipulate that none of the variables are integrated beyond one, i.e. (1). Findings support the existence of cointegration between the variables. Linear ARDL results confirm that TAX and GDE have positive links with FDI in the long-run, whereas GEX, GFCF, and DES have negative links. In the short-run, all variables have a negative link with FDI according to the linear ARDL results. All diagnostic test on serial correlation, normality, and heteroscedasticity are passed with results obtained from the FDI model.
... Aggregate demand could be raised by public spending, and revenues. Hence, so as to cut back the inflation level contractionary financial policy ought to be applied by following budget surplus policy, whereas within the time of recession monetary expansion would be provided by a budget deficit [12]. Keynesian arguments indicate that, financial contraction leads to an impermanent contraction by the aggregate demand channel. ...
Article
Full-text available
The current study was carried out to empirically investigate the relationships between government expenditure, taxes, and debt in Jordan, using the annual time series data over the period 1980-2020. Despite the increased awareness of the importance of fiscal policy, Jordan is still seeing a drastic decrease in the volume of growth. Thus, this study was carried out to determine how government expenditure, taxes, and debt affect economic growth. For the empirical analysis, the study adopted Augmented Dicky Fuller, Phillips-Perron unit root tests, Zivot and Andrews and Lumsdaine and Papell unit root tests with structural breaks. It employs recently developed econometric techniques such as Maki cointegration tests allowing for an unknown number of breaks. The results of the cointegration test showed the presence of a cointegration relationship with structural breaks. It was found that the Jordanian crisis that occurred in 1989 harmed economic growth. However, the structural break in 2008 had an affirmative effect on economic growth. According to the study, tax revenues, public deficits, and debt can all be reduced and the economy can thrive if policymakers implement smart fiscal policies that boost gross fixed capital formation.
... It implies that the responses of economic performance to positive and negative changes in government revenue, in the long run, is asymmetric while in the short-run, it is symmetric. The study agreed with the study of Ravnik & Žilić [31] on Croatia, Mathewos [32] on Ethiopia, and Ocran [33] on South Africa. ...
Article
Full-text available
The study investigated the asymmetric effects of fiscal policy on economic growth in Anglophone countries of the Economic Community of West Africa States (ECOWAS). The annual panel data of Anglophone countries were used from 2005 to 2018. In doing so, the study employed a Structural Vector Autoregressive (SVAR) approach to extract the fiscal policy shocks series. The result showed that, in the long run, the asymmetric effect of government expenditure on economic performance follows a significant inverted U-shape while in the short run, the effect follows a non-significant U-shape. The government revenue follows a significant U-shaped in the long run while in the short run, it follows a non-significant linearity pattern.
Article
Full-text available
Agell et al. (2005) criticise our earlier findings (Fölster and Henrekson 2001) of a robust negative relationship between government size and economic growth for an extended sample of rich countries. In this short paper it is argued that their critique is unfounded. Most importantly, the critique does not deal with the main result of the original study, namely that the most complete specifications for assessing the relationship between government expenditure and growth in rich countries are robust even according to stringent extreme bounds criteria.
Article
Full-text available
Noting that the literature has focused on the link between the level of public expenditure and growth, we derive conditions under which a change in the composition of expenditure leads to a higher steady-state growth rate of the economy. The conditions depend not just on the physical productivity of the different components of public expenditure but also on the initial shares. Using data from 43 developing countries over 20 years we show that an increase in the share of current expenditure has positive and statistically significant growth effects. By contrast, the relationship between the capital component of public expenditure and per-capita growth is negative. Thus, seemingly productive expenditures, when used in excess, could become unproductive. These results imply that developing-country governments have been misallocating public expenditures in favor of capital expenditures at the expense of current expenditures.
Article
This paper surveys the literature on fiscal policy and economic growth. We present a unifying framework for the analysis of long run growth implications of government expenditures and revenues. We find that several tax rates and expenditure categories exhibit a direct impact on the growth rate of the economy. In a creative synthesis we have assigned the relevant literature to the twelve introduced policy variables. Due to the equivalence of some policy variables we are left with six degrees of freedom, where we need four to internalize the model’s intrinsic externalities, leaving two instruments to conduct short run fiscal policy.
Article
The purpose of this paper is to review some of the recent developments in endogenous growth models. Specifically, our focus is on the growth effects of productive government spending in dynamic general equilibrium models. We use a simple overlapping generations model as our basic framework and illustrate the role of taxes and spending. We then examine several related issues: nonrivalry in publicly provided goods, existence and uniqueness of competitive equilibrium, endogenous public policy, ways of financing public expenditures, composition of publicly provided goods and services, and private alternatives. Finally, we review some empirical results related to output elasticity of public capital and educational expenditures.
Article
A number of cross-country comparisons do not find a robust negative relationship between government size and economic growth. In part, this may reflect the prediction in economic theory that a negative relationship should exist primarily for rich countries with large public sectors. In this paper an econometric panel study is conducted on a sample of rich countries covering the 1970–1995 period. Extended extreme bounds analyses are reported based on a regression model that tackles a number of econometric issues. Our general finding is that the more the econometric problems are addressed, the more robust the relationship between government size and economic growth appears. Our most complete specifications are robust even according to the stringent extreme bounds criterion.
Article
Is the evidence consistent with the predictions of endogenous growth models that the structure of taxation and public expenditure can affect the steady-state growth rate? Much previous research needs to be re-evaluated because it ignores the biases associated with incomplete specification of the government budget constraint. We show these biases to be substantial and, correcting for them, find strong support for the Barro model (1990, Government spending in a simple model of endogenous growth. Journal of Political Economy 98 (1), s103–117, for a panel of 22 OECD countries, 1970–95. Specifically we find that (1) distortionary taxation reduces growth, whilst non-distortionary taxation does not; and (2) productive government expenditure enhances growth, whilst non-productive expenditure does not.
Article
This paper describes the empirical regularities relating fiscal policy variables, the level of development, and the rate of growth. We employ historical data, recent cross-section data and newly constructed public investment series. Our main findings are: (i) there is a strong association between the development level and the fiscal structure: poor countries rely heavily on international trade taxes, while income taxes are only important in developed economies; (ii) fiscal policy is influenced by the scale of the economy, measured by its population; (iii) investment in transport and communication is consistently correlated with growth; (iv) the effects of taxation are difficult to isolate empirically.
Article
This paper considers the relationship between aggregate productivity and stock and flow government-spending variables. The empirical results indicate that (i) the nonmilitary public capital stock is dramatically more important in determining productivity than is either the flow of nonmilitary or military spending, (ii) military capital bears little relation to productivity, and (iii) a ‘core’ infrastructure of streets, highways, airports, mass transit, sewers, water systems, etc. has most explanatory power for productivity. The paper also suggests an important role for the net public capital stock in the ‘productivity slowdown’ of the last fifteen years.