Article Type : Research Article
Authors : Bellepea NY, Ozdeser H and Seraj M
Keywords : Economic growth; Inflation oil price; Net trade; Money supply
Based on a PMG-autoregressive framework from 1980 to 2020, this study
attempts to show the impact of inflation, money supply, net trade oil price,
and economic growth in five selected major oil-exporting African economies
(Algeria, Congo, Egypt, Nigeria, and Gabon). Due to cross-sectional dependence
and structural breakdowns, the researchers used the Karavias and Tzavalis unit
root test despite our variables being at various orders of integration.
Subsequently, the panel used a combination of Fisher Johansen and Kao
co-integration to demonstrate a long-term link between the variables. The
study's outcome reveals a negatively statistically significant short- and
long-term association with economic expansion. Furthermore, our findings
suggest that when inflation is high, currency devaluation results, and import
costs will be higher, potentially diminishing the standard of living. In
addition, the study finds a positive relationship between net trade, money
supply, and oil prices on economic growth. An increment in any of these factors
has a beneficial effect on growth. Consequently, African economies must further
lower trade barriers and boost international commerce by decreasing and
streamlining processes and restrictions. Also, the study suggests that placing a
premium on creating and flexibly deploying monetary policies that encourage
expansion is essential.
The connection between inflation and economic
expansion has drawn much interest in the literature on inflation and economics
during the last few decades. Given the link between the economies of African
oil producers and the oil industry, it stands to reason that the continent's
economy is susceptible to shifts in crude oil prices. As a result, the dramatic
drop in crude oil prices in 2015 may be directly attributed to the lifting of
international sanctions on Iran, which increased Iranian oil shipments. For
example, oil dropped from a high of $105 per barrel in 2014 to a low of $37 per
barrel in 2016. (IMF, 2016). Many African countries that rely on oil exports
are still reeling from the recent drop in oil prices. Consider Nigeria, whose
GDP growth dropped from 1.1% in December 2015 to 1.2% in the first quarter of
2016 and continued at that pace through the third quarter of 2016. (NBS, 2016).
Even though the average annual increase in production for this economic union
was 1.3% in 2015, it was 2.1% in 2014. (IMF, 2016). Most oil-producing nations
in Africa, including Algeria, Congo, and Ogede, are likewise concerned with
this problem [1,2]. In addition, empirical studies, such as those by Malka, and
G. Wang, have provided compelling arguments for examining the relationship
between inflation and economic growth [3,4]. However, broad money and the CPI
are the primary determinants of economic development, affecting employment
opportunities, poverty levels, per capita income, and the quality of life in a
country (Phibian, 2010). In Nigeria, attempts to achieve this macroeconomic
goal have been fruitless for many years. A lack of understanding of the
relationships between the factors may be to blame. Therefore, it is crucial to
comprehend the causal links between the money supply and production and their
relationship. Consider how efforts to boost economic growth in Africa will be
made in response to the desire to promote human welfare. Nevertheless, we know
this depends on several variables, including oil prices [5]. The majority of
economic sectors rely on oil to meet their energy needs. For example, energy is
needed for transportation in the real estate, services, and financial sectors,
industrial processing of goods, and electricity. Therefore, even a modest shift
in crude oil prices impacts every sector of an economy. Consider how the cost
of manufacturing items, transportation, service delivery, and all other
economic sectors increases in response to increased crude oil prices. As a
result of the decreased discretionary income brought on by inflation, people's
well-being will be greatly impacted. Researchers were interested in elucidating
the relationship between the inflation of oil prices and economic growth from a
theoretical and empirical standpoint have been drawn to the effects caused by
variations in crude oil prices. Oil prices impact economic growth through
supply and demand [6]. Oil is a raw commodity used in manufacturing; hence its
price changes positively impact the price of manufactured items. This is a
result of inflation caused by manufacturers that increase the price of a
product to account for their increased manufacturing expenses. Research
indicates that crude oil prices positively and negatively impact economic
development. Many countries benefited from the price fluctuations, including
the United Arab Emirates, Algeria, Iran, Iraq, Kuwait, Libya, Oman, Qatar, and
Syria, as shown by studies by Barument [7,8]. In addition, claims that lowering
crude oil prices has a significant positive impact on the economies of Saudi
Arabia, Ghana, and the Middle Eastern and North African (MENA) states [9].
However, in the United States and Sanchez in the OECD nations found that crude
oil prices hurt economic activity [10]. It was determined by Ilhan Oztuk, that
the variations in some African counties were the reason for this disparity in
effect [11,12]. Some African countries import oil while others export it.
Additionally, the disparities were attributed to the study's setting, timing,
and sample. Therefore, this study aims to investigate how inflation affects GDP
expansion in certain pivotal oil-exporting countries in Africa. Unfortunately,
to our knowledge, this research has yet to be conducted in Africa. The expected
outcomes will deal with the link between crude oil prices and GDP expansion,
the impact of inflation on that link and the use of the autoregressive
distributed lag (ARDL) PMG model to examine both the long- and short-term
connections between those two variables.
The rise in empirical studies results from the
aftermath of the last financial crisis and the turmoil generated globally from
analysis of some sovereign African countries—this sharp rise in Inflation on
Economic productivity in some major oil-exporting countries in Africa. Many
substantive studies have revealed the risk of high and increasing Inflation on
Economic Growth. Numerous contributions have argued that high inflation levels
on economic activities are primarily harmful in the long term. Although few
studies also asserted that inflation positively impacts Economic expansion.
Several studies, like ELIAS, use the neoclassical and endogenous models [13].
They wanted to know how commerce with other countries affected the economy of
Nigeria and how trade with other countries affected the economy of Nigeria.
When conducting their analyses, multiple regression analyses were used to
estimate the various international trade components. The study's data, which
covered 1980 to 2012, was taken from the 2012 edition of the CBN statistics
bulletin. Their analysis demonstrated a considerable influence of export
commerce on the expansion of the Nigerian economy. Additionally, their research
showed that import trade had no discernible influence on the expansion of the Nigerian
economy. Hence, Olawunmi Omitogun looked into the connection between Nigeria's
economic growth, revenue fluctuations, and oil prices [14]. Their analysis
makes use of secondary data that was collected between 1981 and 2016. The link
between the variables over the long and short terms was examined using the ARDL
model. The short-term results show that the consumer price index and the
exchange rate have a depressing effect on economic expansion. On the other
hand, the economy's expansion positively and significantly correlates with oil
prices and revenues. Long-term, income from oil has a detrimental effect on
economic growth. However, the price of oil, the CPI, and the exchange rate all
show positive relationships. Aroyehun Olawale MUSBAU also looked into how the
abundance of oil resources affected the economy of Nigeria [15]. In order to
accomplish their goal, annual data from the World Bank Development Indicators
and the Central Bank of Nigeria (CBN), covering the years 1980 to 2018 gathered
(WDI). Oil production was used as a stand-in for the number of oil resources.
In comparison, inflation and the exchange rate served as the controls. The data
was analyzed using the Autoregressive Distributed Lag (ARDL) Model. The
analysis found that in both the long and medium term, Nigeria's economy
benefited from the country's abundance of oil resources by 6.9% and 2.2%. The
outcome also shows that inflation negatively impacts the Nigerian economy over
the long term (-0.9.2%) but has a positive influence over the short term (1.0%)
on RGDP. However, the exchange rate had a favorable long-term and short-term
influence on the Nigerian economy.
Therefore, evaluated the impact of inflation on
economic growth in a developing country's context by using Nigeria's data [16].
This research looks at the correlation between real GDP growth and inflation
using data from the Central Bank of Nigeria's (CBN) website for 30 years
(1986-2016). In addition, the Augmented Dickey-Fuller test implement to
guarantee that the variables remained static (ADF). The Granger causality test
was then used to determine the direction of the link between Nigeria's
inflation and economic growth. This was done to determine whether or not
inflation caused economic development. There was no connection between
Nigeria's rising cost of living and the country's burgeoning economy. In
addition, no lagging variable exists in the association between the increase in
GDP and inflation in Nigeria. According to the study's findings, this effect
does not lead to a rise in either inflation or economic growth. In 2018,
research was conducted by several people, including Olugbenga Anthony
Adaramola, on the impact that inflation has on the growth of the Nigerian
economy. They employ a method that is referred to as autoregressive distributed
lag, and it is applied to a variety of parameters. The following variables from
1980-2018 are accounted for in this study: natural GDP, inflation, interest
rate, exchange rate, degree of economic openness, money supply, and government
consumption expenditures. Based on the data, it can be claimed that the
increase in the money supply and the interest rate has a significant and
beneficial influence on the development of the economy. However, this is
counterbalanced by the fact that inflation and the actual exchange rate have
significant detrimental consequences. It was discovered that the other factors
in the model did not affect the economy's progress in Nigeria. According to the
study's findings on the relationships between variables, the interest rate, the
exchange rate, government consumption expenditures, and GDP are all
interconnected and affect one another, although in opposing ways. There is no
link between openness to trade and inflation or GDP. Koenker and Xiao have
gathered a substantial amount of practical information throughout their
research on whether or not fluctuations in the price of oil are related to
inflation. Changes in the price of oil may have a unique impact on inflation as
assessed by the Consumer Price Index. However, the published empirical research
has largely neglected this possibility. They employ a quantile regression model
to investigate the relationship between changes in oil prices and inflation in
the African states that are net exporters of oil. They can capture potential
variation and the inflation distribution's progress toward its long-run
equilibrium by analyzing the behavior of inflation over a broad range of
quantiles. This allows them to study the distribution of inflation. According
to the findings of their study, changes in oil prices have a deflationary
impact on prices across most quantiles. In addition, the investigation's
finding demonstrates a significant disparity between the magnitude estimations
and sign estimates of the parameters.
The findings of the quantile regression, for example,
indicate that there is hardly any correlation between inflation and oil prices
when evaluated at the 5% significance threshold. However, circumstances have
changed since we first predicted a drop in the price of oil. Between the
quantiles of 0.50 and 0.90, the coefficient has a large and significant
negative value. Umar Bala investigated the relationship between shifts in oil
prices and inflation in Algeria and Libya and came up with contradictory
findings. The study team averaged the prices of Brent, WTI, and Dubai oil. They
also considered the natural spot pricing of oil in several other countries.
Using dynamic panels with autoregressive distributed latency, we could
immediately and later (ARDL) ascertain the outcomes. They concluded that
changes in oil prices, whether an increase or a decrease, had a positive
influence on inflation. On the other hand, research revealed that the impact
was more significant when oil prices were lower. After it was shown that lower
oil prices led to higher inflation, researchers looked into other factors. They
found that increased food production had the opposite impact. On the other
hand, increases in money supply, changes in the exchange rate, and overall
economic growth are all positively connected with inflation. The authors of
this study, Dahmani use a Structural Vector Autoregressive model to investigate
the influence of fluctuations in oil prices on Algeria's economic growth,
unemployment rate, level of government revenue, and level of government
expenditure from 1970 to 2017 [17]. They adopt a unique approach and a
restriction based on the present time to detect structural shocks in the oil
price in Algeria. This is done in order to identify the structural shocks.
According to the calculations, the favorable change in oil prices has a
relatively small but positive influence on GDP growth. However, because of
Dutch illness, adverse oil shocks do not have the sound effects they usually
would.
The focus of monetarism is on the economy's supply
side rather than the market's short-run dynamics. This is one of the defining
characteristics of monetarism. On the other hand, monetarism focuses on the
long-term supply-side features of the economy rather than the short-term
dynamics that are often discussed in economics. This is because monetarists
believe long-term supply-side characteristics are more important than
short-term dynamics For example, Milton Friedman widely acknowledged the
originator of the term; they focused on various essential features of the
economy that continue to be relevant in today's society. Two prominent examples
of these characteristics are the Quantity Theory of Money and the concept of
Monetary Neutrality. The Quantity Theory of Money established a link between
rising prices and expanding economies by providing a straightforward
relationship between the amount of spending in an economy and the total
quantity of money currently in circulation. This enabled the theory to explain
the correlation between the two successfully. Nevertheless, according to
Friedman's theory, inflation occurs when the expansion of the money supply or
the velocity of money increases faster than the expansion of the economy. Friedman's work has cast significant doubt on
the validity of the Phillips Curve hypothesis. His argument was based on the
hypothetical economic premise that prices would double. It was successful
because of this assumption. As a result of having twice as much disposable
income, consumers are unfazed by the fact that the prices of goods and services
are double what they were before. As a result, people think about how quickly
prices will rise in the future and incorporate it into their calculations. As a
direct consequence of this, production and employment levels are unchanged. The
"neutrality of money" is a fundamental principle that underpins
economics. Suppose actual variables, such as GDP, have equilibrium values
unaffected by the amount of money in circulation. In that case, monetary neutrality
may be achievable. We have achieved a condition of super neutrality where
actual variables, such as GDP growth, are unrelated to the pace of expansion in
the money supply over the long term. If inflation operated in this manner,
there would be no adverse consequences. On the other hand, inflation has a
considerable impact on most of the variables that are considered to be
macroeconomic. One of the things that may hold down a nation's economic growth
is inflation, which does this by putting a damper on capital investment,
exports, and savings. According to the monetarist view, inflation has little to
no effect on long-term prices. It is primarily determined by the rate at which
new money is created. Inflation is likely to occur if the pace of economic expansion
is lower than the rate at which the money supply is increasing. The
conventional monetarist view is predicated on the idea that there should be an
increase in the total amount of money in circulation. On the other hand, we see
the reverse of the predicted link between the two: production costs tend to
grow in tandem with price rises. Suppose this increase is more significant than
production. In that case, the monetarists' contention that there is an inverse
connection between the two is proven.
This section of the research discusses the different
methods, techniques, and strategies adopted to collect the necessary data for
the study. In addition, this section reviews and explains the numerous
statistical methods used to assess the study's data collection.
Types of data and
sources
In order to arrive at their findings, most research
projects use two distinct kinds of information, namely, theoretical knowledge
and statistical-econometric analysis. The author of this study used a
methodological strategy comparable to the one they had taken in their earlier
work. In order to get quantitative information on a wide variety of
characteristics, the database of the World Bank was searched. In addition, the
Statistical Review of World Energy was consulted for data about oil prices. The
research will continue to gather information annually for the next four
decades, from 1980 until 2020. This study focuses on the economies of many of
Africa's leading oil-exporting nations to compile its findings.
Variables
and the measurement of variables
The database of development indicators maintained by
the World Bank and the British petroleum statistics repository was used as
secondary data sources for this investigation. Each item of data collected for
the research was sorted into two categories: factors that were used to explain
or explain the results of the inquiry and elements that were not used to
describe the findings (dependent and independent variables). We decided to use
GDP as a stand-in for economic growth and use it as our dependent variable in this
study. Our regressors include inflation, the M1 money supply, net traded value,
and oil price.
The economic relationship among our variables can be
specified in the equation as follow:
GDP = f(INF, MS, NT,
OP)--------------------------------------------------------------------EQ1
According to the above relationship, GDP denotes per
capita income; INF represents inflation; trade represents net trade, and OP
indicates oil price. The fundamental assumption here is that economic growth is
proxied by per capita income and regressed on (INF), (MS), (NT), and (OP).
Hence, the econometric model and equation can be specified as follows:
In this equation, ln GDP represents the
natural log of GDP per capita, ln INF represents the logarithm of inflation, ln
MS represents the natural log of money supply, ln NT represents the logarithm
of net trade, and ln OP represents the logarithm of oil price. We consulted the
World Development Indicators and the Statistical Review of World Energy to
write this piece. The data that we were able to acquire was for the period from
1980 to 2020, and it was collected annually.
Cross-sectional
dependency test
The choice of additional econometric tests
employed in empirical research, such as co-integration and unit root tests, in
panel analysis is heavily influenced by cross-sectional dependence across
variables. As a result, it is predicted that cross-sectional dependency will
significantly affect the statistical characteristics of panel unit root tests.
However, when applied to data series with cross-sectional dependence,
first-generation panel unit-root tests exhibit size distortions and
insufficient power (O'Connell, 1998). Therefore, this feature must be
considered in our panel analysis because of the influence of cross-sectional
dependency on test results. In addition, a cross- sectional dependence test may
be used to assess or validate the use of conventional ADF and PP unit root, or
employs second-generation panel unit root testing should be used. In addition,
Pesaran (2004) proposes an alternative CSD test that does not need a prior
model and may be applied to several model parameters. Under the null hypothesis
of no cross-sectional dependency, the Pesaran CD test statistic has the
following qualities:
The LM test for CD by Pesaran (2015) can
handle slope heterogeneity and cross-sectional issues in a small sample.
However, it usually assumes that the observed test statistics for the studied
residuals (u) are asymptomatically distributed, thus CD N. (0, 1). The result
discussion shows Pesaran's (2015) and Pesaran's (2007) CD testing results.
Thus, a unit root test must account for CD limitations to confirm a long-term
relationship between variables. This study uses Pesaran's CADF and CIPS panel
unit root test. In addition, the Im IPS unit root test was enhanced (2003).
Panel unit root
We cannot continue with the traditional panel unit
root tests since our panel has a cross-sectional reliance on the examined
variables (also known as first-generation panel unit root tests). As a result,
we depend on unit root tests that might potentially take into consideration the
use of cross-sectional data, which are of the second generation. The next
generation of diagnostic tools will be created using heterogeneity as its primary
support structure. We also designed a new version of the original IPS test
using its statistical framework, which is referred to as the Cross-sectionally
Augmented Dickey-Fuller, or CADF, for short. Even though oil prices have
structural fractures, the researcher nonetheless used the unit root test, which
was validated using cross-sectional dependence by Karavias and Tzavalis [18].
Because the tests are invariant under the null to the initial condition, there
is no need to make any assumptions about the nature of the data prior to
carrying out the test, which is not the case with other fixed-T tests. This is
in contrast to the situation where assumptions must be made to carry out other
fixed-T tests. In addition to being resistant to linear trends, the tests do
not depend on the coefficients of the deterministic components. Within the
scope of this thesis, we provide xtbunitroot, a brand-new software that
executes the panel-data unit-root tests developed by Karavias and Tzavalis. In
this study, we presented xtbunitroot, a recently built community-contributed
software that employs structural breakdowns in panel data to apply the
unit-root tests devised by Karavias and Tzavalis. This program was made
possible by the community's contributions (2014). With the use of this first
command, panel unit-root testing that includes structural fractures is a
possibility. Furthermore, it is possible to test for either one or two
structural fractures, depending on the environment. It also allows for errors
that are not normal, dependency, and nonlinear trends, as well as
heteroscedasticity in cross-sectional analyses. The xtbunitroot command was
used for four variables taken from a bank's balance sheet. The results showed
that the bank's noninterest income, assets, and equity returns are stable over
time. Total assets, on the other hand, do not stay the same over time. In the
meanwhile, the idea that Karavias and Tzavalis are discussing may be
characterized as follows:
In the above equation, the drift under null
hypothesis is ?i, while the trend coefficient are ?i and
?2,i.
Panel
co-integration test
The
validity of this research is examined via the use of three standard panel
co-integration tests. These tests are designed to ascertain whether or not a
panel analysis constitutes a connection that is sustained throughout time.
First, the Pedroni, Kao, and Johansen Fisher co-integration test was used to
investigate the degree to which the correlation was stable over time. Second,
Pedroni was the first to construct a battery of tests in 1997, 1999, 2000, and
2004 that consider heterogeneity in co-integration analysis. Pedroni created
these tests [19]. In Pedroni's test, it is permissible for cointegrated vectors
to experience both short-term and long-term variations. The co-integration test
developed by Kao considers the variable nature of the co-integration vectors
[20]. Nevertheless, hyperbolic equality causes a violation of the criterion
that independent variables must be endogenous. This requirement must be met for
the model to be valid. In addition, the co-integration test developed by Kao
and based on the Engle-Granger framework was used in this research. The Kao
test for the existence of a constant may be calculated by determining the
long-term variance using the Schwarz criteria and then using the Newey-West
estimators to analyze the data. Table 4 displays the findings that may be
obtained by applying the test to the panel data set. The relevance of the
probability value was highlighted in the findings of the Kao co-integration
test, which suggested that the null hypothesis of no co-integration was incorrect
and should be rejected. On the other hand, the hypothesis of no co-integration,
known as the null hypothesis, was shown to be true. In order to accomplish
this, we used the Johansen co-integration test to investigate the
co-integrating link between GDP growth and inflation in the leading
oil-exporting countries all around Africa. The Sren Johansen co-integration
test has the potential to confirm the co-integration time series. The Johansen
method of multivariate co-integration is based on using an error correction
formulation of a p-order Vector Autoregressive model with Gaussian error. This
formulation is the foundation of the Johansen technique.
Where ? is most first difference operator, ri
= -(1-A1…….Ai) is the coefficient matrix indicating
short-run changes, and II denotes by II = -(1-A1………-Ai)
is an n&n matrix, where I is an identity of the matrix whose rank
determines amount of co-integrating vectors. However, two likelihood ratio
tests were developed by Johansen for testing the number of co-integration
vectors ®. Mathematically, the trace test can be expressed as follows:
And maximum eigenvalue test statistics given
by:
Trace statistics check the null hypothesis of
no co-integration H0: r = 0 against the alternative of more than 0
co-integration vector H1: r > 0, whereas maximal Eigenvalue statistics test
the null hypothesis of r against the alternative of r + 1 co-integrating
vectors.
The
ARDL PMG approach introduced by Pesaran, Shin, and Smith is utilized in this
research to investigate both the short-term and the long-term linkages between
increases in carbon emissions and increases in GDP [21-30]. The observations
obtained from the sample are collated and then averaged using this technique.
The PMG estimator is derived from the co-integration version of an ARDL model,
which makes it possible for there to be cross-sectional variation in the slope
and short-run coefficients as well as the co-integration components. This
model's error variances and short-run coefficients are variable across
categories, which is a great advantage (heterogeneous). On the other hand, in
an ideal scenario, the coefficients would be comparable or even the same over a
lengthy period. According to Pesaran the economic policies of different
economies differ. The PMG is recommended compared to other panel data models
because it enables flexible and limitless short-run responses across groups.
Other panel data models do not have these capabilities. Although some
short-term benefits may be associated with group consolidation, the long-term
implications are restricting. Therefore, it is essential for the functionality
of the likelihood-based PMG estimator that the long-run elasticity be the same
across all panels. This will ensure that the estimator functions reliably and
consistently. In addition to the explanations that have been shown thus far,
adopting the PMG-ARDL is beneficial because it is prone to producing outliers
even when there is an issue with observing a small sample size in a panel
investigation. In the process of doing so, it considers serial correlation and
endogenous predators by changing the lag structure of both the dependent and
the independent variables. The ARDL (P,q) equation, on the other hand, may be
represented in its general form as follows:
The variables in X t can be a co-integrated
mixture of I(0) and I(1), which is reflected in the vector Y t. Slope coefficients
I and _1; constant y; optimum lag orders i=1; error term vector _it, means
zero. It's also a symbol for the vector process of white noise (serially
uncorrelated or independent). Based on the general form of the model, we can
specify the short-run model as follows:
Meanwhile, we can specify the ECM and
long-run equation as follows:
Where ?it
it is the long-run adjustment speed, represented by the error correction
model's coefficient. This estimator allows for different short-term estimates,
error variance, and intercepts among groups of nations while maintaining the
same long-term parameters.
The table that can be seen above presents and
demonstrates the properties of the data that was used in this investigation. In
addition, it reveals that our data follows the features of a normal
distribution, which is a very significant finding. Using the standard
deviation, one could get assessments of the variance for each variable that is
more precise and thorough than those obtained using the mean. According to the
study's findings, which analyzed 205 different samples, GDP can always be at
least zero. On the other hand, inflation has the distribution's minimum value,
which is the lowest possible value. The degree to which a series is asymmetric
is directly proportional to the degree of skewness shown by one of the
variables. All the variables in the table that came before it has a positive
skewness, except INF, trade, and oil prices. Two indices, namely the gross
domestic product and the money supply, lean more to the right than they should,
which is a favourable direction. However, whether a distribution is flat or
peaky determines how its kurtosis is measured and characterized. Flat
distributions are more accessible to analyze than peaky ones. As seen in the
table, leptokurtic behaviour is shown by both inflation and trade. This points
to the peak being located higher up. The table, on the other hand, reveals that
the other variables have a playkurtic distribution, which suggests that their
distributions are flatter. Using the jaque-bera statistics and probability, we
may determine whether or not our variables are distributed jointly. The study
does not have a normal distribution, however, since the probability value of
the jaque-bera statistics is less than 5%. This indicates that the research
does not have a normal distribution. The results of the CSD analysis are shown
in the table that can be seen above. The numbers provide evidence against the
concept that countries in Africa that are significant oil producers do not
experience CSD. As a result, the CSD is favourable for the African nations
involved in oil exports. There is considerable evidence of cross-sectional
dependence within the series, even if the probability values are highly
significant and fall below 5% at all conventional significance levels. This is
because the probabilities fall below 5% in all cases. The absence of a
cross-sectional dependence is consistent with the conclusion that the null
hypothesis ought to be rejected. Due to economic cooperation, cultural
engagement, political integration, and globalization, a shock in one nation may
have implications in other countries. Because they relied on a cross-sectional
sample, the unit root estimates that are obtained from the more conventional
ADF and PP panels are, by the econometric methodology, erroneous. The
above table presents the result of the stationarity test measured in our
analysis. The unit root tests that allow for structural fractures are carried
out using the approach devised by Karavias and Tzavalis, as seen in the table
that is shown above. These tests permit as many as two structural breakdowns in
the deterministic parts of the series. Moreover, they are applicable in both
small-T and large-T scenarios, where T is the number of time series
observations. Additionally, these tests consider that the number of time series
observations in the sample can vary. The assumption that every series in the
panel is a unit root process is the null hypothesis. The other possibility is
that all the series are stationary, despite breakdowns in the deterministic
definition (intercepts and trends). The break dates are the same for all units
in the panel. However, the severity of the breaks varies depending on the unit.
Moreover, in this case, the decision rules state that a p-value of less than 5%
means the alternative to the unit root hypothesis should be preferred. Such
conditions imply that our variables are stationary. The result shows that
logarithm inflation and oil price are stationary at levels while the remaining
variables are stationary. Thus, the findings of our work revealed that we have
I(0) and I(1) variables that account for a mixed order of integration.
Once it is known whether or not the factors
under consideration are stationary, the standard method is to check for a
long-run link between the gauges of the variables. The findings of the Johansen
Fisher mixed panel co-integration test are shown and explained in the table
that can be seen above. When the probability of the trace statistics is
compared to the max-eigen test, the study's findings indicate that
co-integration does exist. Despite this criticism, the findings of the analysis
indicate that GDP, inflation, the amount of money available for trade, and the
price of oil all have a connection over the long term. In addition, the
information gives us the ability to estimate the relationships between our
regressors using several techniques, which will be beneficial in the long run.
For example, we used the Kao co-integration analysis to do more research on
stability. The results also provide evidence for long-term correlations between
the parameters that are measured in this study. This result indicates that the
competing hypothesis—namely, no co-integration—was deemed acceptable. Using the
PMG-ARDL model, we have calculated all of the long-term correlations that exist
between the regressand and the regressors in the table that is located above.
According to the findings, there is a statistically significant link between
inflation and the slowing of economic growth. For example, the GDP inflation
elasticity is 0.045597% when all other components are held constant. Because of
this, it can be concluded that there is a negative impact of inflation on GDP
of 0.045597% for every percentage point of inflation. When customers'
purchasing power falls as a result of inflation, devaluation takes place in the
economy. The table demonstrates that there is a positive relationship between
the growth of the money supply and the expansion of the economy, with a 1% rise
in the money supply resulting to a 1% expansion in GDP. The chart also
demonstrates a statistically significant link between commerce and GDP. A table
analysis shows that an increase of 1% in business activity is associated with a
0.081713% increase in GDP. There is a strong positive association between the
price of oil and GDP, which brings us to our last point. If nothing else were
to change, a one percent increase in oil price would result in a 0.111949%
increase in GDP. This assumes that everything else would stay the same. This is
because a rise in the price of oil results in increased cash flow for the
primary African nations that export oil, resulting in increased economic growth
in those countries.
The
following table displays the short-term predictions made using the PMG-ARDL
model. According to the findings of the study, inflation harms economic growth.
This result is in line with the conclusion drawn from the long-run projection.
When inflation reduces consumer spending, the elasticity of economic growth
concerning inflation lowers by 0.013223 percentage points, as shown by the statistics.
This happens because inflation reduces people's discretionary spending. An
additional 0.112219% is added to GDP for every 1% rise in the money supply
showing a positive relationship between the two. The table also demonstrates a
high positive connection between net trade and GDP, with a 1% rise in net trade
leading to a 1% gain in the short term. This is shown by the table containing
both of these statistics. If, for example, there is a one percent rise in net
trade, then there will be a 0.13136 percent increase in GDP, all other things
being equal. Finally, a country's level of commercial activity might provide
insight into its prospective economic output. When we look at the data in the
table, we can see that oil price has a significant inverse relationship with
GDP. This is something that we already knew. The result illustrates a percent
rise in oil price declines economic growth by -0.046104%. Finally, the table
reveals that the error correction term is negative and statistically
significant at all conventional significance levels. The researcher separates
the short and long run because of the negative coefficient. However, the table
reports that any deviation or distortion from the short-run equilibrium, the
velocity at which the economy converges or restores long-run equilibrium, is
37.9099%. The figure above depicts or shows whether our residuals were normally
distributed. The null hypothesis argues for normal distribution against the
alternative. The decision criteria are such that if the probability value is
above 5%, we fail to reject the null hypothesis of normal distribution among
residuals. However, given that the probability value of the jarque-bera
statistics is less than 0.05, we conclude that our residuals are not normally
distributed (Figure 1).
Figure
1: Residual
diagnostics.
This study aimed to determine the relationships among GDP growth, inflation, money supply, trade, and oil price among the top ten oil exporting countries in Africa from 1980 to 2020. Because of the presence of cross-sectional dependence, any change in any variable that is utilized inside a nation may have implications on the economies of neighbouring places. Furthermore, this may happen whenever there is a shift in any variable that is used within the country. This suggests that the economic success of one nation is inextricably linked to the success of other nations. Consequently, ensuring that the variables being utilized were constant was vital for preventing the estimator from arriving at incorrect results. As a consequence, it is essential to provide evidence, in advance of the estimation process, that the series used were stationary, denoted by the symbol I (1). This may be accomplished by using the traditional second-generation unit root tests (CADF and CIPS). Because of the structural holes in our data, however, we decided to utilize the unit root test developed by Karavias and Tzavalis. This test is valid even when the cross-sectional dependency is taken into account. (Table 1-4) contains the solution to the non-stationarity problem that we encountered. As such, our variables were integrated in a different order. After performing the stationarity test, the researchers indicate in table 5 co-integration analysis. Moreover, this table shows whether the variables under scrutiny have a long-run connection. Finally, the researchers employed the combined fisher Johansen panel co-integration test in (Table 5). The result reveals evidence of a long-run connection since the statistics' p-values are highly significant. This test's robustness was verified using the Kao co-integration, the results of which were in line with those of the panel-combined fisher co-integration test. The researchers then used the PMG-ARDL to estimate the relationship, as it allows consistency in long-run coefficients. In addition, the estimation results show our model's long and short-run effects. For example, the coefficient for ECM, which measures the adjustment speed back to equilibrium, was negative, with the range as predicted from 0 – 1 and significant probability values showing 0.379, that is, 37.9% adjustment speed back to equilibrium in the short run. Meanwhile, (Table 6) reports the overall long-term effects of the factors undertaken by the study. It presents a statistically significant association among the variables studied. For example, table 6 reveals a significant inverse correlation to GDP growth in these major leading oil-exporting African nations. As inflation rises in these nations, their economy experiences a decline in productivity. This is because inflation erodes the purchasing power of a dollar.
High
inflation implies a persistent increase in the general price level, which
subsequently increases the cost of production in the long term. The PMG
estimates that inflation has a temporary but detrimental impact on long-term
growth rates, leading to a permanently lower per capita income than would
otherwise be achieved. An extra one percent in inflation is predicted to reduce
yearly growth by around 0.045597% over very long periods, reducing steady-state
per capita income. Investment levels fall, and productive elements are utilized
less effectively due to inflation. Our findings suggest that the marginal cost
of inflation is unrelated to the inflation rate. Nevertheless, inflation has
significant long-term consequences, and attempts to rein it in will pay off
with improved long-term performance and higher per capita income (Table 7).
Meanwhile,
the result of this study's inverse connection between inflation and economic
growth is consistent with an investigation done by Adaramola. Consistent with
table 6, our result finds a significant positive correlation between money
supply and economic growth. This relationship is in agreement with the
monetarist quantity theory of money. However, the literature of Adaramola also
provides a positive association between money supply and growth, which is
consistent with our findings. In addition, the study found that nations with
high levels of net trade saw higher rates of economic expansion. This research
indicates that international commerce and accelerated economic expansion
mutually benefit and considerably contribute to one another. This conclusion is
significant for analyzing the development made by other African countries that
are comparable to the continents big oil producers since it enables comparisons
to be made between the two types of economies. The findings of this research
demonstrate the significance of international commerce in developing the
economies of several African countries that are major oil exporters. The
results, which are both short-term and long-term in nature, provide evidence in
support of the idea that trade led to growth. Meanwhile, the findings of this
research are consistent with those of a study conducted by Zarra-Nezhad and a
study conducted by Brueckner and Lederman. In conclusion, but certainly not
least, the PMG result shows a positive long-run correlation between the price
of oil and economic growth. Because of this correlation, it comes to reason
that as the price of oil increases, the income of the countries that are the
world's leading oil exporters will also go up. A rise in profits indicates more
money being invested in those countries, which suggests more production.
Because of the increase in the number of industrial operations, there may be an
acceleration in economic development in the long term. Because these nations
rely on money from oil exports, every increase in oil exports leads to an
increase in revenues denominated in foreign currency. On the other hand, since
these economies are developing, they will import goods and services from other
countries, which will cause a drop in the value of their currencies.
Consequently, the cost of importing local products will increase while
exporting foreign goods will decrease. Because it is more difficult and costly
to get them from elsewhere, the prices of imported items will be higher in the
context of the nations under consideration. Because a currency devaluation does
not affect the price of oil exported, it cannot make exports more appealing,
and these nations will not gain from an increase in exports even if those
exports were to grow.
This
study's objective is to investigate how inflation influences the rate of GDP
expansion in important African oil-exporting nations. For this reason, the
research relied on a panel of yearly data spanning the five most significant
leading African nations of 1980-2020. These countries were Algeria, Congo,
Gabon, Egypt, and Nigeria. First, we used the KT unit root that Karavias and
Tzavalis provided because our variables were at varied degrees of integration,
and there was a cross-sectional dependency. There were structural fractures
(2014). After that, dynamic panel ARDL (PMG) models were utilized to
investigate the short- and long-term effects of changes in inflation on
economic growth. Additionally, the panel combination Fisher Johansen and Kao
co-integration was utilized to demonstrate a long-run link among the variables.
This helped to demonstrate that there is a connection between the variables.
The findings of this study provide credence to past research suggesting that
economic expansion is inversely correlated with unemployment rates. We found
that excessive inflation harms the value of the national currency and the price
of imported items, which might reduce the overall standard of life for the
people as a whole. According to this evidence, higher inflation levels over the
long run are detrimental to economic development and discourage investment.
Research conducted by Adaramola and colleagues lends credence to the notion
that inflation and economic expansion go hand in hand (2018). An association exists
between the expansion of the money supply and the development of the economy.
An increase in the money supply influences economic growth that is both
immediate and long-lasting. This effect is known as a multiplier effect.
Consequently,
it is of the utmost importance to make the formulation and flexible execution
of expansionary monetary policies a top priority. Despite this, the outcomes of
this study add validity to the argument that is put forward by the Quantity
Theory of Money. This theory maintains that both the income velocity (V) and
the price level (P) continue to stay constant. If V and P do not change, then
any gain in nominal GDP must be attributed to an increase in the money supply.
This is the case even if no other change exists (M). We also found that growth
in both the amount of net trade and the money supply positively correlates with
economic expansion. As a consequence, the findings of this research lend
credibility to the idea that trade is the primary driver of economic expansion
for Africa's oil exporters. These countries' relationships with other nations
have been a substantial factor in their respective economies' development during
the last several decades. Consequently, African economies must further lower
trade barriers and boost international commerce by decreasing and streamlining
processes and restrictions. These African countries export mostly primary
goods, whose prices fluctuate and are influenced by the international market.
For an outward-oriented strategy to have a much more significant influence on
economic development, the nation should shift from exporting raw materials and
semi-manufactured commodities to exporting high-value-added items. In addition,
trade policy should encourage investments in capital-intensive industries and
build human capital capable of absorbing technology from developed nations. Some
limitations of the research cannot be avoided, notwithstanding the efforts that
were made. First, this study can only generalize about five of the essential
oil-exporting economies in Africa, owing to a need for more data from the rest
of the continent. A second issue is that certain regressors may be endogenous,
and the omitted variable bias in the estimate procedure may be an issue.
Finally, adding additional essential variables to a system of equations where
commerce and capital are also affected by other economic factors would be a
worthwhile expansion of this study. As a result, we may better understand the
many mechanisms through which inflation influences economic expansion.