significant impact on real gross fixed capital formation. Moreover, both renewable energy consumption
and the labour force appear to serve as complements to
real gross fixed capital formation. With respect to Eq.
(8d), at the 95% confidence interval, the results indicate a 1% increase in domestic electricity consumption
causes a rise in GDP of 0.032, while a 1% increase in
fixed capital formation causes a rise in GDP of 0.757.
While a 1% increase in labour causes a drop in GDP of
0.04, both labour and real gross fixed capital formation have a positive and statistically significant impact
on the GDP in the short run while electricity consumption is statistically insignificant. The coefficients are
exactly identified. Therefore, this makes the VECM
an ideal test for this study.
5.2 GDP and electricity consumption
5.2.1 Stationarity
Using ADF tests, data was found to be non- stationary
at level, but all exhibited stationarity at first difference
(see Appendices Table 5a and 5b)
5.2.2 Lag length selection
Since it has been shown that ADF tests are sensitive to
lag lengths (Campbell & Perron 1991), we determine
the optimal lag length by using Akaike’s information
criterion (AIC). The formal testing of the lag structure
is based on the maximum likelihood function.
5.2.3 Normality
The data was normally distributed (refer to Figure 3 in
the appendices).
5.2.4 Granger causality
The finding using the Granger causality test showed
a bidirectional relationship between log of GDP and
log of electricity (as shown in Table 4 in the appendices). This confirmed the feedback hypothesis of the
energy–economic growth nexus. However, there was
unidirectional relationship running from log of labour
force to log of GDP and log of electricity Granger
causes log capital formation.
5.2.5 Cointegration test
The Johansen trace and max test statistics suggest
the existence of at least one cointegrating equation
between GDP (LGDP) and electricity consumption
(Lelec), So we reject the null hypothesis at the 5% level
of significance, thus we go ahead to determine the long
run relationship using VECM results that show that
a positive and significant relationship exists between
electricity consumption and GDP. Specifically, a 1%
increase in electricity consumption tends to increase
GDP by 0.06% as shown in Table 3 in the appendices.
5.3 Discussion
The empirical results from the multivariate framework
shows that GDP is explained by 3% electricity consumption, capital 75%, and labour 4%, which is in
agreement with Kummel and Lindenberg (2014) and
Stern (2004), that electricity consumption has a small
cost share yet a very big contribution to the overall
GDP, as shown in Table 6. When the model is regressed
log electricity consumption alone to log GDP is 37%
showing a large contribution of electricity consumption to GDP as shown in Table 7. The results from the
error correction models lend support for the feedback
hypothesis given that both short-run and long-run bidirectional causality exists between renewable energy
consumption and economic growth.
6 CONCLUSIONS AND RECOMMENDATIONS
6.1 Conclusions
The paper investigates the causal relationship between
electricity consumption and economic growth for
Uganda over the period 2008–2018. It is worthwhile
to examine electricity consumption as a clean renewable energy. Uganda has performed well in increasing
hydroelectric power supply in the last decade with the
hope to drive the economy at a much faster rate.
There is a bidirectional causality implying that
increasing electricity consumption increases economic growth. And economic growth in turn increases
electricity consumption. This explains the critical
increase in electricity supply capacity is not followed
by increasing electricity consumption in Uganda due
to a low economic growth rate.
6.2 Recommendations
In this paper, we investigated the causal relationship
between electricity demand and economic growth for
Uganda and based on the findings make the following recommendations for both energy economists and
policy authorities.
Policy makers should encourage a multilateral
effort to promote renewable energy and energy efficiency in the region. Regional cooperation on the
development of renewable energy markets between
public and private sector stakeholders could begin
with sharing information across countries with respect
to ongoing projects, technologies, as well as the
financing and investment strategies. The establishment of partnerships between the public and private
sector would also facilitate the technology transfer process of bring renewable energy projects to
market.
In addition, policy makers should introduce the
appropriate incentive mechanisms for the development and market accessibility of renewable energy.
Such incentives could include tax credits and/or subsidies for the production and consumption of renewable energy. The establishment of markets for tradable renewable energy certificates along with the
implementation of renewable energy portfolio standards may promote the expansion of the electricity
281
and the labour force appear to serve as complements to
real gross fixed capital formation. With respect to Eq.
(8d), at the 95% confidence interval, the results indicate a 1% increase in domestic electricity consumption
causes a rise in GDP of 0.032, while a 1% increase in
fixed capital formation causes a rise in GDP of 0.757.
While a 1% increase in labour causes a drop in GDP of
0.04, both labour and real gross fixed capital formation have a positive and statistically significant impact
on the GDP in the short run while electricity consumption is statistically insignificant. The coefficients are
exactly identified. Therefore, this makes the VECM
an ideal test for this study.
5.2 GDP and electricity consumption
5.2.1 Stationarity
Using ADF tests, data was found to be non- stationary
at level, but all exhibited stationarity at first difference
(see Appendices Table 5a and 5b)
5.2.2 Lag length selection
Since it has been shown that ADF tests are sensitive to
lag lengths (Campbell & Perron 1991), we determine
the optimal lag length by using Akaike’s information
criterion (AIC). The formal testing of the lag structure
is based on the maximum likelihood function.
5.2.3 Normality
The data was normally distributed (refer to Figure 3 in
the appendices).
5.2.4 Granger causality
The finding using the Granger causality test showed
a bidirectional relationship between log of GDP and
log of electricity (as shown in Table 4 in the appendices). This confirmed the feedback hypothesis of the
energy–economic growth nexus. However, there was
unidirectional relationship running from log of labour
force to log of GDP and log of electricity Granger
causes log capital formation.
5.2.5 Cointegration test
The Johansen trace and max test statistics suggest
the existence of at least one cointegrating equation
between GDP (LGDP) and electricity consumption
(Lelec), So we reject the null hypothesis at the 5% level
of significance, thus we go ahead to determine the long
run relationship using VECM results that show that
a positive and significant relationship exists between
electricity consumption and GDP. Specifically, a 1%
increase in electricity consumption tends to increase
GDP by 0.06% as shown in Table 3 in the appendices.
5.3 Discussion
The empirical results from the multivariate framework
shows that GDP is explained by 3% electricity consumption, capital 75%, and labour 4%, which is in
agreement with Kummel and Lindenberg (2014) and
Stern (2004), that electricity consumption has a small
cost share yet a very big contribution to the overall
GDP, as shown in Table 6. When the model is regressed
log electricity consumption alone to log GDP is 37%
showing a large contribution of electricity consumption to GDP as shown in Table 7. The results from the
error correction models lend support for the feedback
hypothesis given that both short-run and long-run bidirectional causality exists between renewable energy
consumption and economic growth.
6 CONCLUSIONS AND RECOMMENDATIONS
6.1 Conclusions
The paper investigates the causal relationship between
electricity consumption and economic growth for
Uganda over the period 2008–2018. It is worthwhile
to examine electricity consumption as a clean renewable energy. Uganda has performed well in increasing
hydroelectric power supply in the last decade with the
hope to drive the economy at a much faster rate.
There is a bidirectional causality implying that
increasing electricity consumption increases economic growth. And economic growth in turn increases
electricity consumption. This explains the critical
increase in electricity supply capacity is not followed
by increasing electricity consumption in Uganda due
to a low economic growth rate.
6.2 Recommendations
In this paper, we investigated the causal relationship
between electricity demand and economic growth for
Uganda and based on the findings make the following recommendations for both energy economists and
policy authorities.
Policy makers should encourage a multilateral
effort to promote renewable energy and energy efficiency in the region. Regional cooperation on the
development of renewable energy markets between
public and private sector stakeholders could begin
with sharing information across countries with respect
to ongoing projects, technologies, as well as the
financing and investment strategies. The establishment of partnerships between the public and private
sector would also facilitate the technology transfer process of bring renewable energy projects to
market.
In addition, policy makers should introduce the
appropriate incentive mechanisms for the development and market accessibility of renewable energy.
Such incentives could include tax credits and/or subsidies for the production and consumption of renewable energy. The establishment of markets for tradable renewable energy certificates along with the
implementation of renewable energy portfolio standards may promote the expansion of the electricity
281