Content uploaded by Christopher A. Erickson
Author content
All content in this area was uploaded by Christopher A. Erickson on Apr 24, 2017
Content may be subject to copyright.
Public Debt and Economic Growth in Ghana
Victor Owusu-Nantwi and Christopher Erickson
Abstract:This study uses Johansen cointegration and the vector error correction model to examine the long-term and causal
relationship between public debt and economic growth in Ghana. Annual time series data were gathered from the World Bank
Development Indicators and IMF Economic outlook data from 1970 to 2012. The findings from the study reveal a positive and
statistically significant long-run relationship between public debt and economic growth. Also, in the short run a bidirectional
Granger causality link exists between public debt and economic growth. The study recommends that Ghana should acquire
public debt for very high priority projects and programs that are well appraised and self-sustained that could contribute positively
to economic growth.
1. Introduction
Over the last half century, most countries in the world have seen significant development in their economies and Ghana is no
exception. In many countries, the underlying key issue behind economic development —but which has mostly been ignored
by empirical research —is the issue of public debt. Economic development requires investment in infrastructure, education,
social welfare, health and other sectors of the economies. The huge expenditures associated with such investments make it
challenging for countries to fund them from tax revenues, and mostly leads to budget deficits. Budget deficits caused by such
investments have to be financed, but the key question that confronts policy makers and most government economists is; how
should the budget deficit be funded? Public finance provides three alternative sources of financing deficits; taxes, debts and
user fees (Rosen and Gayer, 2008). Developing nations, faced with weak tax regimes and low incomes, opt for debt as the
best option for financing government budget. Given this, it is not surprising that public debt plays a particularly important
role in developing countries. Public debt enables fiscal authorities to play their role in stabilizing their economies and
stimulate aggregate growth (Caribbean Development Bank, 2013). In this regard, Ghana is no exception to borrowing. Its
current lower middle income economic status, makes it relevant for the country to continuously invest in the productive
sectors of the economy to ensure sustained growth. Ghana has a very weak tax system and, as a result, the country does not
generate enough tax revenues to fund its expenditures. Therefore, taxes are not considered a good option for funding budget
deficits. Additionally, the informal sector of Ghana’s economy is excluded from the tax base of the country due to the lack of
an accurate database of that sector, which makes it difficult for the tax system in Ghana to track their economic activities and
tax them accordingly (Bagahwa and Naho, 1995).
The stated policy of Ghana is to pursue prudent macroeconomic policies to improve the socioeconomic conditions and well-
being of its people through sound fiscal policies such as government spending in the areas of education, health, defense,
infrastructure and other social services. This requires that the government borrows in both local and international financial
markets, also through the World Bank and International Monetary Fund to finance development projects. Ghana’s total debt as of
2013 was $15.83 billion which is about 33 per cent of total GDP (IMF Economic Outlook, 2013).
Some studies have emphasized the significance of public debt to economic growth and notable among them are Modigliani
(1961), Buchanan (1958), Meade (1958) and Diamond (1965). However, few empirical studies have examined the impact of
public debt on developing economies. This study attempts to fill this gap in the literature and provide some insights into the fiscal
Victor Owusu-Nantwi (corresponding author), Department of Economics, Applied Statistics and International Business New Mexico State
University, PO Box 30001, MSC 3CQ, Las Cruces, NM 88003-8001; e-mail: owuvic@nmsu.edu. Christopher Erickson, Department of Economics,
Applied Statistics and International Business, New Mexico State University, Box 30001, MSC 3CQ, Las Cruces, NM 88003-8001, USA.
African Development Review, Vol. 28, No. 1, 2016, 116–126
© 2016 The Authors. African Development Review © 2016 African Development Bank. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. 116
policy formulation and implementation. Given this background, this study assesses empirically the impact of public debt on
economic growth in Ghana between 1970 and 2012. The paper is organized as follows: the next section presents a review of the
literature; Section 3 presents the data sources, theoretical model and research methodology; Section 4 presents and discusses the
empirical results. We conclude and summarize our findings in Section 5.
2. Literature Review
Numerous studies have analyzed the impact of public debt on economic growth. This is demonstrated through a large body of
theoretical and empirical literature on the impact of public debt on economic growth. The empirical evidence provides mixed
and inconsistent predictions about the effect of public debt on economic growth. Reinhart and Rogoff (2010a) analyzed the
relationship between economic growth and public debt for the period 1949 to 2009 for 20 developed economies. Their study
found that high levels of debt and economic growth are negatively correlated; however, they found no link between debt and
economic growth when public debt is below 90 percent of GDP. The study also showed no link between inflation and growth.
Patillo et al. (2002) analyzed the external debt effect on per-capita GDP growth for a period between 1969 and 1998 using a
panel dataset of 93 developing countries. Their empirical results indicate that the effect of external debt on per capita GDP
growth is negative for the net present value of debt levels above 35–40 percent of GDP. Clements et al. (2003) reported a
negative correlation between external debt and growth for a panel of 55 low-income countries for a period that spanned from
1970 to 1999. Checherita and Rother (2010) evaluated the effect of government debt on economic growth for 12 European
countries over the period of 1970–2010 using a panel fixed effects estimation technique. The study reported a non-linear
impact of debt on economic growth, indicating that the government debt-to-GDP ratio has a negative effect on long-term
growth when debt is about 90–100 percent of GDP. Panizza and Presbitero (2012) examined the impact of public debt and
economic growth for a sample of OECD countries using the instrumental variable approach. The study concluded that there
is a negative relationship between debt and growth. Kumar and Woo (2010) studied the long-run effect of public debt on
economic growth using time series data that spans four decades of some developed and emerging countries. They concluded
that there is a long-run negative relationship between debt and growth. El-Mahdy and Torayeh (2009) investigated the debt
and growth relationship for Egypt’s economy using data spanning 1981–2006 and the study revealed a robust negative
relationship between debt and growth. Tchereni et al. (2013) analyzed the effect of foreign debt on Malawi’s economic
growth using time series data over the period 1975–2003. They reported a statistically insignificant negative relationship
between foreign debt and economic growth in Malawi.
Schclarek (2004) assessed the impact of gross government debt on economic growth for a sample of 24 industrial
countries over the period 1970–2002. The study found no robust relationship between debt and growth. Additionally,
Ogunmuyiwa (2011) evaluated the effect of external debt on Nigeria’s economic growth using time series data from
1970–2007. The study used a vector error correction model for the estimation and the results revealed a weak and
insignificant relationship between debt and growth. Also, the study reported no causality between debt and growth. Shah and
Shahida (2012) investigated the effect of the public debt burden on economic growth of Bangladesh for a study period of
1980–2012 using a vector autoregression model and found no impact of debt on economic growth. This indicates that the
public debt has no effect on economic growth.
Baum et al. (2012) investigated the relationship between public debt and economic growth using the dynamic threshold
panel methodology for 12 European countries for the period 1990–2012. The study reported a positive and high statistically
significant impact of debt on GDP when the debt-to-GDP ratio was less than 67 percent; after which point, there was no
relationship between debt and GDP. Egbetunde (2012) examined the impact of public debt on economic growth in Nigeria
between 1970 and 2012 using a vector autoregression model. The findings revealed a positive relationship between public
debt and growth. Also, the study reported a bidirectional link between public debt and economic growth in Nigeria and this
indicates that changes in public debt will cause variation in Nigeria’s economic growth and vice versa. Amin and Audu
(2006) reported a positive relationship between debt and economic growth when they examined the impact of external
indebtedness on Nigeria’s economic growth from 1990 to 2004. Maghyereh (2003) studied the impact of external debt on the
economic growth of Jordan using the endogenous growth model and time series data from 1970 to 2000. The empirical
results of the study revealed that at a debt level of 53 percent of GDP, there is a positive and statistically significant
relationship between external debt and economic growth. Anyanwu and Erhijakpor (2004) studied the impact of domestic
debt on economic growth of Nigeria over the period 1970–2003. The study reported that current debt has a significant
Public Debt and Economic Growth in Ghana 117
© 2016 The Authors. African Development Review © 2016 African Development Bank
negative impact on economic growth and this is attributed to a high domestic interest rate. Additionally, the study revealed
that past domestic debt has a significant and positive effect on Nigeria’seconomicgrowth.
While public debt is relevant to economic growth, high public debt could adversely impact medium and long-term growth
through several channels (Kumar and Woo, 2010). Specifically, excessive public debt could have an inauspicious effect on
capital accumulation and growth through interest rates (Gale and Orszag, 2003; Baldacci and Kumar, 2010), higher future
discretionary taxation (Barro, 1979; Dotsey, 1994), and inflation (Sargent and Wallace, 1981; Barro, 1995; Cochrane, 2010)
(cited in Kumar and Woo, 2010). Aghion and Kharroubi (2007) also point out that high public debt could trigger banking and
currency issues which may result in higher volatility and lower growth.
3. A Simple Model of Public Debt and GDP
In this section we develop a simple model of the macroeconomy that relates GDP growth to debt, as well as to other factors.
Production is given by:
Yt¼FK
t;Zt;Lt;St
ðÞ ð1Þ
where Ytis aggregate output, which is a measure of GDP, Ktis private capital, Ztis public capital, Ltis labor, and Stis a vector of
other state variables that affect output, and subscript indicates the time period. FK
t;Zt;Lt;St
ðÞis a production function meeting
the standard assumptions that Fx>0 and Fxx <0, where x¼{K, Z, L, S}.
To abstract from household decision making, it is assumed that savings is a fixed fraction of income (i.e., as in Solow, 1957).
Savings is used to finance private and public investment or to pay taxes:
sYt¼IK;tþpzIZ;tþTtð2Þ
where IK;tis investment in private capital, IZ;tis investment in public capital, Ttis taxes, which for simplicity, are assumed to be
lump sum. The variable pzis a parameter that indicates the relative efficiency of public investment in terms of output and is
determined by institutional factors. The larger is pzthe less efficient is the public investment. Increased public corruption, for
example, would increase pz.
The transition equation for private capital is given by:
DKtþ1¼IK;tdKKtð3Þ
where 0 <dK<1 is the depreciation rate of private capital. Setting the initial level of private capital to zero, Equation (3) implies
that:
Ktþ1¼X
t
s¼0
IK;sdKKs
ð4Þ
Similarly, the transition equation for public capital is given by:
DZtþ1¼IZ;tdZZtð5Þ
where 0 <dZ<1 is the depreciation rate of public capital. Setting the initial level of public capital to zero gives:
Ztþ1¼X
t
s¼0
IZ;sdZZs
ð6Þ
Solving (2) for IK;tand substituting into (4) gives:
Ktþ1¼X
t
s¼0
sYtpzIZ;tTtdKKs
ð7Þ
© 2016 The Authors. African Development Review © 2016 African Development Bank
118 V. Owusu-Nantwi and C. Erickson
Substituting (6) and (7) into (1) gives:
Ytþ1¼FX
t
s¼0
sYtpzIZ;tTtdKKs
;X
t
s¼0
IZ;sdZZs
;Ltþ1;Stþ1
!
ð8Þ
The government budget constraint is given by:
GtþIZ:t¼DDtþTtð9Þ
where Gtis government consumption, and DDtis the change in public debt (i.e., the government deficit). Solving (9) for DDt,
substituting into (8) and noting that debt (D) is given by DtP
t
s¼0
DDs
ðÞgives:
Ytþ1¼FX
t
s¼0
sYsTsþGsdKKs
ðÞpzDDt;DtþX
t
s¼0
TsGsdZZs
ðÞ;Ltþ1;Stþ1
!
ð10Þ
Taking the derivative of Ywith respect to Dt:
dYtþ1
dDt
¼@F
@Zpz
@F
@Kð11Þ
The first term is the increase in production arising from public investment funded by debt. The second term is crowding out of
private investment. Note the second term depends on pz; the more efficient is the government sector, the less is crowding out.
Equation (11) is ambiguous, which is consistent with the varying results reported in the literature. Whether public debt
contributes or hinders GDP growth in Ghana, thus, is an empirical question.
4. Data Description, Model and Methodology
4.1 Data Description
The empirical analysis is carried out using annual time series data for Ghana that spans 1970 to 2012. A total of seven
macroeconomic variables were employed in the analysis. The definitions and sources of each of the variables are described in
Table 1.
4.2 Model Specification
The relationship between public debt and economic growth have been extensively studied and debated especially within the
framework of the neoclassical growth theory. The econometric model to be estimated for this study is as follows;
GDPt¼aþb1GOVDtþb2GOV Etþb3INFLtþb4INVtþb5OPEN tþb6POPGtþe1tð12Þ
where GDP
t
is the growth rate of GDP in period t, which we use as a measure of economic growth, GOVD
t
is a measure of public
debt, GOVE
t
is a government consumption expenditure, INFL is inflation, INV is investment spending, OPEN
t
is openness, and
POPG
t
is population growth. Population is used rather than employment because reliable estimates of employment are not
available for Ghana.
4.3 Research Methodology
Time series data was analyzed to test for its stationarity using the Augmented Dickey–Fuller (1979) and Phillips–Perron (1988)
unit root tests. The stationarity test is necessary as most time series data are non-stationary and if this is not checked, running
© 2016 The Authors. African Development Review © 2016 African Development Bank
Public Debt and Economic Growth in Ghana 119
regressions with it may yield to spurious regression (Owusu-Nantwi and Kuwornu, 2011). These two unit root tests determine
whether the variables are stationary or not and also indicate the degree of integration. Performing this test allows correction for
the spurious autocorrelation associated with time series data by adding lagged differenced terms on the right-hand side of the
equation as indicated in Equation (2) (Owusu-Nantwi and Kuwornu, 2011). The ADF test equation is:
DXt¼mþgTþdXt1þX
k
i¼1
liDXi1þetð13Þ
where X
t
represents the variable in question, Tis the trend, kis the lag length and e
t
is a random variable assumed to be white
noise.
A Johansen multivariate cointegration test is performed after the unit root test when the variables are found to have no unit
root or stationary at the first difference. Furthermore, if the variables are cointegrated then the relationship may be interpreted
as a long-run equilibrium relationship. Lastly, a Granger causality test using the vector error correction model is estimated to
determine the short-run causality among the variables. The short run granger causality test is measured by the Wald Test.
5. Empirical Results
5.1 Unit Root Test Result
The ADF and PP unit root tests results are presented in Table 2. The test results reveal that all the time series variables are
integrated of order one except for inflation which was stationary at levels. Even though inflation is stationary at levels, the study
used cointegration and vector error correction techniques for the estimate because most of the variables are stationary at first
difference.
5.2 Vector Error Correction Model (VECM) Analysis
Lag Length Selection Criterion
Table 3 presents the result for the lag length selection criteria. This study adopted four lag selection criteria (Final
Prediction Error (FPE), Akaike Information Criterion (AIC), Schwarz Information Criterion (SIC) and Hannan-Quinn
Information Criterion (HQ)) for the analysis. All four selected criteria suggested a lag length of 3, which is used in the
vector error correction model (VECM).
The Cointegration Test
Table 4 presents the Johansen multivariate cointegration test result. The result consists of both the trace and maximum
Eigenvalue tests. These tests determine the number of cointegration vectors. Both test the null hypothesis that the number of
cointegrating vectors is less than or equal to 0, 1, 2, 3, or 4. For each case, the null hypothesis is tested against the alternative
hypothesis.
Table 1: Data description and source
Variable Definition Data Source
GDP Real GDP growth rate World Bank
GOVE Government consumption expenditure as a percentage of GDP World Economic Outlook (IMF)
GOVD Gross government debt as a percentage of GDP (as a proxy for public debt) World Economic Outlook (IMF)
INV Investment as a percentage of GDP World Economic Outlook (IMF)
INFL Inflation (consumer price) in percentage World Bank
POPG Population growth (%) World Bank
OPEN (Sum of export and import) as a percentage of GDP as a proxy for capital mobility UNCTAD
© 2016 The Authors. African Development Review © 2016 African Development Bank
120 V. Owusu-Nantwi and C. Erickson
The results show that the trace test statistic for the null hypothesis of the number of cointegration (H
0
:r¼0) is 543.2732. This
value is above the critical value of 125.6154 at the 5 percent significance level which indicates the rejection of the null hypothesis
(no cointegration) in favor of the alternative hypothesis. It is important to note that one limitation of this study is the cointegration
test. The Johansen cointegration test is based on asymptotic theory which involves a larger sample size and therefore the use of a
small sample size in this study makes it a concern. However, other studies with a small sample size used this technique and some
of them include Daud et al. (2013), Ogunmuyiwa (2011) and Wijeweera et al. (2005).
The maximum Eigenvalue test shows that the null hypothesis of no cointegration is 200.3767 which is greater than the critical
value of 46.23142 at the significance level of 5 percent. The null hypothesis is rejected and the alternative hypothesis of
cointegration is accepted. For r1, r2 and r3 in the context of the trace test and maximum Eigenvalue test, the null
hypothesis is rejected while the alternative hypothesis of the presence of cointegration is accepted. Contrary to the above
discussion, for r4, the null hypothesis for the trace test statistic and maximum Eigen test is not rejected at the significance level
of 5 percent. Therefore, there are three cointegration relationships between the variables.
The Long-run Relationship
The estimates for the long-run relationship between economic activity and public debt are presented in Table 5. The empirical
estimates of the long-run relationship show that debt-to-GDP ratio, government consumption expenditure, investment and
Table 2: Unit root test
Augmented Dickey Fuller (ADF) Phillips-Perron (PP)
Level First difference Level First difference
Intercept Trend & intercept Intercept Trend & intercept Intercept Trend & intercept Intercept Trend & intercept
GDP –2.5894 –3.7805 –6.6228 –4.0244 –3.6842 –2.3169 –10.0313 –10.8172
(0.1049) (0.0301) (0.000)(0.0187)(0.073) (0.1727) (0.0000)(0.0000)
GOVD –2.5176 –2.4681 –4.5022 –4.42757 –2.2078 –2.2601 –4.3543 –4.2510
(0.1206) (0.3406) (0.0011)(0.0067)(0.4704) (0.1901) (0.0016)(0.0103)
GOVE –1.7935 –4.0349 –8.8808 –8.7792 –4.0349 –1.5770 –8.9130 –8.8400
(0.3773) (0.0168) (0.0000)(0.000)(0.0168) (0.4831) (0.0000)(0.0000)
INFL –4.5538 –6.3788 –13.5314 –13.5721 –6.3688 –4.6060 –17.1735 –27.8886
(0.0009) (0.000) (0.000)(0.000)(0.000) (0.0008) (0.0001)(0.000)
INV –2.030 –4.5829 –6.4503 –6.3366 –4.5051 –1.9040 –17.2996 –17.6749
(0.2732) (0.0044) (0.000)(0.0001)(0.0054) (0.3266) (0.000)(0.0000)
OPEN 0.01394 –3.6973 –4.7252 –4.7618 –3.8325 –0.09193 –4.6277 –4.5661
(0.9533) (0.0367) (0.0006)(0.0030)(0.0272) (0.3266) (0.0008)(0.0000)
POPG –0.2474 –4.4713 –4.4358 –4.3225 –3.6070 –1.6399 –3.9661 –3.6345
(0.9218) (0.0081) (0.002)(0.0094)(0.0446) (0.9423) (0.005)(0.042)
Significance level at 1% level of confidence, significance at 5% level of confidence.
Source: Authors’calculations.
Table 3: Lag length selection test
Lag length test
Final prediction
error (FPE)
Akaike information
criterion (AIC)
Schwarz information
criterion (SIC)
Hannan-Quinn information
criterion (HQ)
0 1.32e þ09 40.86646 41.19026 40.97201
1 767306.3 33.33221 35.92264 34.17663
2 50465.67 30.07601 34.93307 31.65929
3 32.8900420.8823228.0060023.20446
indicates lag order selected by the criterion.
Source: Authors’calculations.
© 2016 The Authors. African Development Review © 2016 African Development Bank
Public Debt and Economic Growth in Ghana 121
population growth rate have a positive impact on the real GDP growth rate and are statistically significant. The coefficient for
debt-to-GDP ratio is statistically significant and has a positive long-run impact on the real GDP growth rate, suggesting that the
public debt contributes effectively and significantly to the economic growth of Ghana.
This empirical result is consistent with Baum et al. (2012) and Maghyereh (2003). The coefficient suggests that debt is
one of the key catalysts for Ghana’s economic growth, and this is supported by Egbetunde (2012). He examined the impact
of public debt on economic growth in Nigeria between 1970 and 2012 using the vector autoregression model. His findings
revealed a positive long-run relationship between public debt and economic growth. The coefficient for government
consumption expenditure variable is positive and statistically significant at 0.01 level. It implies that there is a positive
long-run relationship between government consumption expenditures and real GDP growth rates. It also shows that
government consumption expenditure is an important determinant of real GDP growth rate. This is supported by Kumar and
Woo (2010) whose empirical findings also reported a positive long-run relationship between government consumption
expenditure and real GDP growth rate for 12 European countries. The investment coefficient is positive and statistically
significant, indicating that investment has a positive long-run impact on real GDP growth rate. This finding is consistent
with the study by Maghyereh (2003) which reported a positive long-run relationship between investment and economic
growth of the Jordanian economy. The study included population growth rate as a proxy for labor. The coefficient for
population growth rate is positive and statistically significant at the 0.01 level. It shows that in the long run increases in
population growth rates will cause an increase in real GDP growth rate. This empirical finding is supported by Panizza and
Presbitero (2012).
Furthermore, openness and inflation have a negative effect on the real GDP growth rate, though only openness is statistically
significant. The negative effects indicate that increases in inflation and openness would tend to lower economic growth in Ghana.
The no effect relationship between inflation and real GDP growth rate is supported by Reinhart and Rogoff (2010a). The long-
run coefficients presented in Table 5 can be regarded as long-run elasticities because each coefficient measures the percent
change in the response variable following a unit or percent change in a particular explanatory variable.
The long-run Granger causality result is presented in Table 6 and is identified in the error correction term (ECT-1) for each
variable. The result indicates that the error correction term (ECT-1) for the real GDP growth rate variable, which is also called
speed of adjustment, is negative and statistically significant. The speed of adjustment of the error correction is –0.517 and this
implies that the system corrects its previous level of disequilibrium by 51.7 percent within one period. That is, 51.7 percent of the
previous year’s real GDP growth rate disequilibrium will be corrected in the long run.
Table 4: Johansen cointegration test
Model
Null
hypothesis
Trace
statistic
Critical
value (5%)
Maximum
Eigen
Critical
value (5%) Results
Lag length:3
#
r0543.2732 125.6154 200.3767 46.23142 Both Standard trace and Maximum Eigen
values showed three cointegration vectorsr1342.8965 95.75366 161.6589 40.07757
r2181.2377 69.81889 107.3161 33.87687
r373.92155 47.85613 64.46011 27.58434
r4 9.461445 29.79707 7.408386 21.13162
Trace statistics and maximum Eigen tests indicate 3 cointegrating equation(s) at the 0.05 level.
#
indicates lag length.
Source: Authors’calculations.
Table 5: Estimates of long-run cointegration model
Independent variables
Dependent variable (GDP) GOVD GOVE INF INV OPEN POPG Constant
Coefficient 0.035151 0.267281 –0.0019 0.0148 –0.06789 7.11540 –24.593
t-value [65.819] [25.809] [–0.967] [2.283] [–41.648] [45.606]
Source: Authors’calculations.
© 2016 The Authors. African Development Review © 2016 African Development Bank
122 V. Owusu-Nantwi and C. Erickson
Short-run Granger Causality
The short-run Granger causality test result is also presented in Table 6. The Wald test is estimated to investigate the short-run
causal relationship. The null hypothesis states that there is no Granger causality; thus no linear relationship between the real GDP
growth rate, public debt, government consumption expenditure, inflation, investment, openness and population growth rate is
observed in Ghana from 1970 to 2013. However, the alternative hypothesis suggests otherwise, indicating the existence of linear
Granger causality.
The results in Table 6 show a significant causal relationship between debt to GDP ratio and the real GDP growth rate. That is,
in the short run, the debt-to-GDP ratio Granger causes the real GDP growth rate. Next, the results show a statistically significant
causal link between the real GDP growth rate and government consumption expenditure; investment, openness; and population
growth rate respectively. These variables have a Granger causal effect on real GDP growth rate in the short run. However,
changes in government consumption expenditure and inflation are statistically insignificant indicating that these variables are not
important for explaining the variation in real GDP growth rate in the short run.
Even more, the results also indicate a statistically significant causal relationship between the real GDP growth rate and debt-
to-GDP ratio; government consumption expenditure; investment; and openness respectively. The positive short-run relationship
between real GDP growth rate and debt-to-GDP ratio is consistent with the study by Baum et al. (2012). Inflation and population
growth rate are statistically insignificant indicating that there is no causal links between real GDP growth rates and these
variables. This indicates that variations in these variables do not cause changes in real GDP growth rates. In summary, the causal
link between real GDP growth rate and public debt in the short run is bidirectional, indicating that the causation run from the real
GDP growth rate to public debt and also from public debt to the real GDP growth rate. This is because the p-values in both cases
are less than the assumed critical values. This bidirectional link between debt-to-GDP ratio and real GDP growth rate reported is
consistent with the study by Egbetunde (2012).
Equally important, the results exhibit statistically significant causal links between government consumption expenditure and
debt-to-GDP ratio; and population growth rates respectively. That is, in the short run, variation in government consumption
expenditure is determined by variations in the debt-to-GDP ratio and population growth rates. However, real GDP growth rates,
inflation, investment and openness in the short run are statistically insignificant indicating no causality. Therefore, the causal link
between government consumption expenditure and debt-to-GDP ratio runs in both directions (bidirectional) indicating that the
Granger causality runs from debt-to-GDP ratio to government consumption expenditure and vice versa. In addition, there is a
Table 6: Vector error correction model (VECM)
Dependent Independent variables –Chi-square value (Wald test) t-statistic
variable GDP GOVD GOVE INFL INV OPEN POPG Error Correction Term
GDP 15.782
(0.0004)
22.2058
(0.0000)
0.440
(0.8025)
33.5483
(0.0000)
48.590
(0.000)
65.538
(0.000)
–0.51707
[–6.46112]
GOVD 2.76168
(0.0251)
0.62884
(0.0730)
0.78879
(0.6741)
14.7929
(0.0006)
6.01481
(0.0494)
3.3583
(0.1865)
6.43897
[0.68193]
GOVE 2.019220
(0.364)
0.24109
(0.0864)
2.66026
(0.2644)
2.36221
(0.3069)
3.5146
(0.1725)
4.8867
(0.0869)
1.16579
[1.41382]
INFL 8.779442
(0.0124)
4.22142
(0.1212)
13.8110
(0.0010)
4.15908
(0.1250)
3.42616
(0.1803)
14.8776
(0.0006)
5.15023
[2.26239]
INV 0.452152
(0.0797)
2.57585
(0.0276)
1.05355
(0.5905)
1.07822
(0.5833)
1.51826
(0.4681)
3.130349
(0.2091)
2.25655
[–0.11844]
OPEN 14.44229
(0.0007)
1.36478
(0.5054)
0.45031
(0.7984)
4.90285
(0.0862)
5.18964
(0.0747)
0.88866
(0.6413)
0.00747
[1.78367]
POPG 3.971395
(0.1373)
8.2177
(0.0164)
4.99483
(0.0823)
2.75076
(0.2527)
3.19685
(0.2022)
4.57092
(0.1017)
2.17527
[–0.51754]
Notes:p-values shown in parentheses.
indicates significance at an alpha of 1% level, indicates significance at an alpha of 5% level, indicates significance at an alpha of 10% level.
Source: Authors’calculations.
© 2016 The Authors. African Development Review © 2016 African Development Bank
Public Debt and Economic Growth in Ghana 123
unidirectional link between government consumption expenditure and real GDP growth rates. This shows that causation runs
from only government consumption expenditure to real GDP growth rates and not vice versa.
The results show that the coefficients for real GDP growth rates, debt-to-GDP and government consumption expenditure
are statistically significant. This indicates that there are short-run causal relationships between inflation, and real GDP
growth rates; debt-to-GDP ratio; and government consumption expenditure respectively. The coefficients for investment,
openness and population growth rates are statistically insignificant and these exhibit no causal relationship with respect to
inflation. In summary, the causal link between inflation and real GDP growth rate is unidirectional indicating that the
causation runs from inflation to real GDP growth rate and not otherwise. There are unidirectional causal relationships
between inflation and government public debt as well as government consumption expenditure. These mean that causation
runs from both public debt and government consumption expenditure to inflation and not vice versa.
The results show the causal relationships between investment, and real GDP growth rates; and debt-to-GDP ratio. Also there
are bidirectional links between investment and real GDP growth rates; and debt-to-GDP ratio. The direction of causality runs in
both directions. The other variables are statistically insignificant showing no Granger causality.
The results also indicate Granger causality between openness and the real GDP growth rate, inflation and investment. That is,
there is a bidirectional link between openness and real GDP growth rates and a unidirectional relationship between openness and
inflation and between openness and investment. This means short-run causality runs from inflation and investment to openness
and not vice versa.
Table 6 indicates there are causal links between population growth rates, and debt-to-GDP ratio, as well as government
consumption expenditure, and openness in the short run. Beyond this, there is a bidirectional relationship between population
growth and government debt as well as between population growth and government consumption expenditure. Furthermore,
there is a unidirectional relationship between population growth (POPG) and openness (OPEN). This shows that openness
(OPEN) Granger causes population growth, but population growth does not Granger cause openness. All other variables are not
statistically significant showing no Granger causality relationship with population growth.
6. Conclusion and Recommendation
This study empirically investigates the long-run and the short-run relationship between public debt and economic growth in
Ghana. Vector error correction and Johansen cointegration analysis were employed to test for causal relationships between the
variables for the period 1970 to 2012. The empirical results reveal a positive and significant long-run relationship between real
GDP growth rate and public debt, indicating that public debt contributed to economic growth in Ghana. In the short run, there is a
bidirectional relationship between public debt and economic growth, meaning that public debt Granger causes economic growth
and vice versa.
Based on these findings, the following recommendations are made. First, Ghana should acquire public debt for very high
priority projects and programs that are well appraised and self-sustained that could contribute positively to the economic growth
of Ghana. Although public debt contributes positively to economic growth, Ghana should be mindful of its public debt
acquisition. This is because higher public debt could in the long term contribute negatively to economic growth as suggested by
Reinhart and Rogoff (2010a, 2010b), Checherita and Rother (2010) and Kumar and Woo (2010). Ghana should continuously
pursue sound fiscal and monetary policies as it creates an enabling environment for economic growth. Creating such an
environment serves as a precursor or necessary prerequisite for the effective use of public debt.
References
Aghion, P. and E. Kharroubi (2007), ‘Cyclical Macro Policy and Industry Growth: The Effect of Countercyclical Fiscal Policy’,
Working Paper, Harvard University.
Amin, A. A. and I. Audu (2006), ‘External Debt, Investment and Economic Growth: Evidence from Nigeria’,CBN Economic
and Financial Review, Vol. 44, No. 1, pp. 88–113.
Anyanwu, J. C. and A. E. O. Erhijakpor (2004), ‘Domestic Debt and Economic Growth: The Nigerian Case’,West African
Financial and Economic Review, Vol. 1, pp. 98–128.
© 2016 The Authors. African Development Review © 2016 African Development Bank
124 V. Owusu-Nantwi and C. Erickson
Bagahwa, M. D. S. and A. Naho (1995), ‘Estimating the Second Economy in Tanzania’,World Development, Vol. 23, No. 8,
pp. 1387–99.
Baldacci, E. and M. Kumar (2010), ‘Fiscal Deficits, Public Debt, and Sovereign Bond Yields’, IMF Working Paper 10/184.
Barro, R. (1979), ‘On the Determinants of the Public Debt’,Journal of Political Economy, Vol. 85, No. 5, pp. 940–71.
Barro, R. (1995), ‘Inflation and Economic Growth’, NBER Working Paper No. 5326.
Baum, A., C. Checherita-Westphal and P. Rother (2012), ‘Debt and Growth: New Evidence for the Euro Area’, European
Central Bank Working Paper Series No. 1450.
Buchanan, J. M. (1958), ‘Public Principles of the Public Debt’,The Independent Review, Vol. 11, No. 3, pp. 435–49.
Caribbean Development Bank (2013), Public Sector Debt in the Caribbean: An Agenda for Reduction and Sustainability,
Caribbean Development Bank.
Checherita, C. and P. Rother (2010), ‘The Impact of High and Growing Government Debt on Economic Growth: An Empirical
Investigation for the Euro Area’, IMF Working Paper No. 10/1237.
Clements, B., R. Bhattacharya and T. Q. Nguyen (2003), ‘External Debt, Public Investment, and Growth in Low-Income
Countries’, IMF Working Paper No. 03/249.
Cochrane, J. (2010), ‘Understanding Policy in the Great Recession: Some Unpleasant Fiscal Arithmetic’, Working Paper,
University of Chicago.
Daud, S. N. M., A. H. Ahmad and W. N. W. Azman-Saini (2013), ‘Does External Debt Contribute to Malaysia Economic
Growth?’,EkonomskaIstrazivanja-Economic Research, Vol. 26, No. 2, pp. 51–68.
Diamond, P. (1965), ‘National Debt in a Neoclassical Growth Model’,American Economic Review, Vol. 55, No. 5, pp. 1126–50.
Dickey, D. A. and W. A. Fuller (1979), ‘Distribution of the Estimators for Autoregressive Time Series with a Unit Root’,Journal
of American Statistical Association, Vol. 74, pp. 427–31.
Dotsey, M. (1994), ‘Some Unpleasant Supply Side Arithmetic’,Journal of Monetary Economics, Vol. 33, pp. 507–24.
Egbetunde, T. (2012), ‘Public Debt and Economic Growth in Nigeria: Evidence from Granger Causality’,American Journal of
Economics, Vol. 2, No. 6, pp. 101–106.
El-Mahdy, M. A. and M. N. Torayeh (2009), ‘Debt Sustainability and Economic Growth in Egypt’.International Journal of
Applied Econometrics and Quantitative Studies, Vol. 6, No. 1, pp. 21–55.
Gale, W. and P. Orszag (2003), ‘The Economic Effects of Long-term Fiscal Discipline’, Urban-Brookings Tax Policy Center
Discussion Paper No. 8.
IMF Economic Outlook (2013), World Economic Outlook: Transitions and Tensions, IMF, Washington, DC.
Kumar, M. and J. Woo (2010), ‘Public Debt and Growth’, IMF Working Paper 10/174.
Maghyereh, A. I. (2003), ‘External Debt and Economic Growth in Jordan: The Threshold Effect’,International Economics,
Vol. 56, No. 3, pp. 337–55.
Meade, J. E. (1958), ‘Is the National Debt a Burden?’Oxford Economic Papers, Vol. 10, No. 2, pp. 163–83.
Modigliani, F. (1961), ‘Long-Run Implications of Alternative Fiscal Policies and the Burden of the National Debt’,Economic
Journal, Vol. 71, No. 284, pp. 730–55.
Ogunmuyiwa, M. S. (2011), ‘Does External Debt Promote Economic Growth?’Journal of Economic Theory, Vol. 3, No. 1,
pp. 29–35.
Owusu-Nantwi, V. and J. K. M. Kuwornu (2011), ‘Analyzing the Effect of Macroeconomic Variables on Stock Market Returns:
Evidence from Ghana’,Journal of Economic and International Finance, Vol. 3, No. 11, pp. 605–15.
© 2016 The Authors. African Development Review © 2016 African Development Bank
Public Debt and Economic Growth in Ghana 125
Panizza, U. and A. F. Presbitero (2012), ‘Public Debt and Economic Growth: Is There a Causal Effect?’MoFiR Working Paper
No. 65.
Pattilo, C., H. Poirson and L. Ricci (2002), ‘External Debt and Growth’, IMF Working Paper No. 02/69.
Phillips, P. C. B. and P. Perron (1988), ‘Testing for Unit Roots in Time Series Regression’,Biometrika, Vol. 75, No. 2,
pp. 335–46.
Reinhart, C. and K. Rogoff (2010a), ‘Growth in a Time of Debt’,American Economic Review, Vol. 100, No. 2, pp. 573–78.
Reinhart, C. and K. Rogoff (2010b), ‘From Financial Crash to Debt Crisis’, NBER Working Paper, No. 15795.
Rosen, S. G. and T. Gayer (2008), Public Finance, 8th edn, McGraw-Hill/Irwin, New York.
Sargent, T. and N. Wallace (1981), ‘Some Unpleasant Monetarist Arithmetic’,Quarterly Review, (Fall), Federal Reserve Bank
of Minneapolis.
Schclarek, A. (2004), ‘Debt and Economic Growth in Developing Industrial Countries’, mimeo. Available at: http://www.nek.
lu.se/publications/workpap/Papers/WP05_34.pdf
Shah, M. M. H. and P. Shahida (2012), ‘External Public Debt and Economic Growth: Empirical Evidence from Bangladesh’,
Academic Research International, Vol. 3, No. 2, pp. 508–15.
Solow, R. M. (1957), ‘Technical Change and the Aggregate Production Function’,The Review of Economics and Statistics,
Vol. 39, No. 3, pp. 312–20.
Tchereni, B. H. M., T. J. Sekhampu and R. F. Ndovi (2013), ‘The Impact of Foreign Debt on Economic Growth in Malawi’,
African Development Review, Vol. 25, No. 1, pp. 85–90.
Wijeweera, A., B. Dollery and P. Pathberiya (2005), ‘Economic Growth and External Debt Servicing: A Cointegration Analysis
of Sri Lanka, 1952 to 2002’, Working Paper Series in Economics No. 2005-8, University of New England.
© 2016 The Authors. African Development Review © 2016 African Development Bank
126 V. Owusu-Nantwi and C. Erickson
