Scholarly article on topic 'Signatures of water resources consumption on sustainable economic growth in Sub-Saharan African countries'

Signatures of water resources consumption on sustainable economic growth in Sub-Saharan African countries Academic research paper on "Economics and business"

CC BY-NC-ND
0
0
Share paper
Keywords
{"Water resources consumption" / "Sustainable economic growth" / Translog / "Sub-Saharan Africa"}

Abstract of research paper on Economics and business, author of scientific article — Suinyuy Derrick Ngoran, Xiong Zhi Xue, Presley K. Wesseh

Abstract We adapt the transcendental logarithmic (translog) production model to examine the role of water resources consumption on economic growth in 38 Sub-Saharan African (SSA) countries. Labor, capital and energy are incorporated into the model to provide for omitted variable bias. Several findings have been documented from the investigation. First, the results suggest that economic growth in SSA is driven mainly by water and labor. Capital and energy were found not to significantly drive economic growth. Second, technical change in SSA is scale-biased and factor augmenting; suggesting that efficiency of water withdrawals and labor use would lead to technological progress in SSA. Third, substitution possibilities between water and labor exist indicating that restrictions on water withdrawals would lead to labor intensiveness and vice versa. Finally, a more general insight from the study is that efficient use of water resources promotes technological innovation and hence, critical for sustainable development.

Academic research paper on topic "Signatures of water resources consumption on sustainable economic growth in Sub-Saharan African countries"

HOSTED BY

Gulf Organisation for Research and Development International Journal of Sustainable Built Environment

ScienceDirect www.sciencedirect.com

CrossMark

Original Article/Research

Signatures of water resources consumption on sustainable economic growth in Sub-Saharan African countries

Suinyuy Derrick Ngoran a, Xiong Zhi Xue b'*, Presley K. Wesseh Jr.c

a Department of Environmental Sciences and Environmental Engineering, College of the Environment & Ecology, Xiamen University, 361102 Xiamen,

Fujian Province, China

b College of the Environment & Ecology, Coastal and Ocean Management Institute (COMI), Xiamen University, 361102 Xiamen, Fujian Province,

c China Center for Energy Economics Research, College of Economics, Xiamen University, Xiamen 361005, China Received 22 October 2015; accepted 4 April 2016

Abstract

We adapt the transcendental logarithmic (translog) production model to examine the role of water resources consumption on economic growth in 38 Sub-Saharan African (SSA) countries. Labor, capital and energy are incorporated into the model to provide for omitted variable bias. Several findings have been documented from the investigation. First, the results suggest that economic growth in SSA is driven mainly by water and labor. Capital and energy were found not to significantly drive economic growth. Second, technical change in SSA is scale-biased and factor augmenting; suggesting that efficiency of water withdrawals and labor use would lead to technological progress in SSA. Third, substitution possibilities between water and labor exist indicating that restrictions on water withdrawals would lead to labor intensiveness and vice versa. Finally, a more general insight from the study is that efficient use of water resources promotes technological innovation and hence, critical for sustainable development.

© 2016 The Gulf Organisation for Research and Development. Production and hosting by Elsevier B.V. All rights reserved.

Keywords: Water resources consumption; Sustainable economic growth; Translog; Sub-Saharan Africa

1. Introduction

Geographically, Sub-Saharan Africa (SSA) is the extent of the African continent that falls south of the Sahara Desert. Political wise, it comprises of all African nations

* Corresponding author. E-mail addresses: derrick_ngoran@yahoo.co.uk (S.D. Ngoran), xzxue@xmu.edu.cn (X.Z. Xue), masterpresley@yahoo.com (P.K. Wesseh Jr.).

Peer review under responsibility of The Gulf Organisation for Research and Development.

that are completely or partly located south of the Sahara (excluding Sudan)1 (Fig. 1). SSA is endowed with abundant natural resources such as minerals, forests, wildlife and rich biological diversity. Yet, these natural resources are largely unexploited and do not reflect the welfare of the inhabitants in the region (Ngoran et al., 2015a). In this segment of the African continent, we find some of the world's biggest tropical rain forests and highest equatorial mountains. Strategic natural resources are unevenly

1 http://unstats.un.org/unsd/methods/m49/m49regin.htm (Retrieved: 20/05/2013).

http://dx.doi.org/10.1016/j.ijsbe.2016.04.002

2212-6090/© 2016 The Gulf Organisation for Research and Development. Production and hosting by Elsevier B.V. All rights reserved.

Figure 1. Location of Sub-Saharan Africa. Source: http://blogs.ft.com/material-world/2014/05/11/part-lll-sub-saharan-africa/ (Retrieved: 20/01/2015).

repartitioned. For instance, more than 20% of the remaining tropical forest is in the Democratic Republic of the Congo, while the river Congo, river Niger, the river Nile, river Zambezi and Lake Victoria and Lake Chad detain more than half of SSA's fresh water resources (Urama and Ozor, 2010; Ngoran et al., 2015b).

Given the background that the SSA boasts some of the largest natural resources in the world, including water, it lags behind the wheels of economic growth. From a global perspective, sustainable economic growth in SSA remains a daunting challenge. Within a span of 20 years (1980-2000), economic growth in SSA countries stood low due to macro-economic instability. According to the World Bank data (2012), the average growth rate in Gross Domestic Product (GDP) per capita in SSA was —0.6% from 1980 to 2000. Also, Banya and Zajda (2015) portrayed that economic growth in SSA averaged 4% over the last ten years, but, poverty and inequality have remained persisting problems challenging sustainable development. Hence, growth pathways that provoke strong changes in SSA's production and consumption patterns are necessary. Nowadays, many SSA countries are experiencing a lot challenges linked to water resource consumption. Following the growing importance of mining activities and the overdependence of SSA on agriculture (agriculture accounts for over 75% employment, contributes about

30% GDP and with an estimated 40% of exports) (Reij and Smaling, 2008)), water resources consumption is bound to increase. Again, with SSA burgeoning population, urbanization, adverse effects of climate change and the strive by various SSA governments to eradicate poverty through economic growth, the problem of dwindling water resources will further be compounded. For instance, the Lake Chad, once noted as one of Africa's largest freshwater lakes, has shrunk significantly owing to related anthropogenic actions and the effect of climate change. The surface area of Lake Chad reduced from 45,000 square kilometers to 10,000 square kilometers in 1960 and 1998, respectively (Ngoran et al., 2015a,b). Lake Victoria, the world's largest tropical lake (68,790 square kilometers) and the second-largest freshwater lake in the world, is fast losing its waters for similar reasons (Urama and Ozor, 2010). Based on the fact that more than half of the population in SSA lives in Urban areas coupled with the disproportionate allocation of primary industries, the anthropogenic built-up of urban milieu is affirmatively a game-changer in shaping water resources (Ngoran and Xue, 2015). However, studies tallying only on data from the interchanging factors of the built-up environment would represent a big error margin since data estimates are collected at national levels by international institutes (Wesseh and Lin, 2016a). Therefore, it is imperative to

Table 1

Definition of variables.

Variable

Component

Annual freshwater

withdrawals Labor input

Gross fixed capital

formation Total population

Technical change

Million dollars

Billion cubic meters Number of

people

Millions

The sum of gross value added by all resident manufacturers in the economy plus any product taxes and minus any subsidies not involved in the value of the products

Total water withdrawals, not including evaporation losses from reservoirs. Withdrawals for agriculture and industry are total withdrawals for irrigation and livestock production and for direct industrial use Employment to population ratio multiplied by the active population

Average annual growth of gross fixed capital formation based on constant local currency

All inhabitants irrespective of legal status or citizenship, excluding refugees that are not permanently established in the country of asylum, who are normally considered part of the inhabitants of their country of origin

Time trend representation

study the holistic link between water resource consumption and economic growth, especially given the fact that limited works exist in this domain.

On the one hand, previous econometric studies have tallied mostly on the link between water and energy produc-tion2, water resources demand and conflict3, water footprint and water resources consumption4, water resources consumption and climate change5, water demand and agriculture6. Despite the fact that latter studies have not investigated the link between water and economic development; they have all portrayed the important role that water resources contribute to all the sectors of the economy.

On the other hand, the limited investigations on the relationship between water resources consumption and economic growth have proceeded with structural decomposition analysis (Cazcarro et al., 2013), panel data set (Tir et al., 2014), Environmental Kuznets Curve (EKC) (Katz, 2014), linear programing model and sector error-correction model (Salami et al., 2009). These studies all posit that water resource consumption plays a determinant role in either increasing or decreasing economic growth. Though water availability can potentially propagate economic growth, long term economic growth can only be ensured if an increase in water provision is defined on sustainable bases. Therefore, there is need for more robust approaches as it can be noticed that the translog production model has never been applied to the investigation of water resource consumption and impacts on sustainable economic growth. The contributions of this study tend to be enormous and therefore would offer a fresh perspective into the water-economic growth debate,

2 See e.g., Okadera et al. (2014), Varbanov (2014) and Bindra et al. (2014).

3 See e.g., Madulu (2003), Mbaiwa (2004) and Bohmelt et al. (2014).

4 See e.g., Chapagain and Orr (2009), Wang et al. (2013a,b), Sallam (2014) and Yu and Qingshan (2014).

5 See e.g., Akhtar et al. (2008); Piao et al. (2010), Dawadi and Ahmad (2013) and Palazzoli et al. (2015).

6 See e.g., Howell (2001) and Fereres and Soriano (2007).

both in terms of methodological issues and policy planning in SSA.

This article, therefore, investigates the contribution of water resources to economic growth in SSA and analyzes the sustainability of this growth. Furthermore, the possibilities of substituting between water and labor, while propelling economic growth, have been estimated. The applied methodology is still novel in this literature. We use the translog production model in this work. The work, from the authors' viewpoint is the first of its kind to evaluate the impact of water resources consumption, and substitution effects on sustainable economic development for such a large group of African countries. Hence, the present study brings new insights into the literature and offers opportunities for further research questions.

The rest of the paper is set out as follows. Section 2 describes the model and estimation technique. Section 3 presents the empirical results and discussion while Section 4 ends up with the conclusion.

2. Data and methodology

2.1. Data

In most SSA countries, data are seldom collected at the national level due to the near absence in data accounts. Most of these data are therefore estimated by international institutions such as the World Bank and the International Monetary Fund. For the purpose of this study, several datasets are analyzed. The country-level panel dataset includes observations on GDP, labor, annual freshwater withdrawals, energy and gross fixed capital formation. Only Sub-Saharan countries with at least two data points for each considered variable are included in the study. GDP and population data points are available for all the countries considered, whereas there are some irregularities on other variables. To ensure robustness in the data sets, transformations of the datasets are performed whereas GDP and capital are considered at constant prices (2005 = 100) in order to exclude the influence of inflation. The data collected are from 38 SSA countries spanning the

period 1980-2011 giving us an imbalanced panel of 31 observations. While data on energy consumption are obtained from the US Energy Information Administration (EIA), GDP, annual freshwater withdrawals, and gross fixed capital formation are made available on the World Development Indicators database. Labor inputs here are not observed directly, but obtained by multiplying the employment to population ratio with the active population (Table 1).

2.2. Model specification

By definition, the translog production model is a representation of a second-order differential approximation (Taylor approximation) at any given point. It represents a rather simple and locally flexible functional form that places little or no restrictions on the underlying production technology (Wesseh and Lin, 2016a,b). In other words, this model imposes no requirements on the value of the function, whether on its first or second derivatives, at the approximation point. It describes the relationship between output and input services from several different productive factors.

As a general consensus in most literature, the simplest forms of production model is the functional form estimated as a linear relationship. This can be expressed as;

ln(Y) = ao + £ a ln(/,)

periods; j, k = 1, 2, ..., J are the applied inputs; ln Yit is the logarithm of the output associated with the ith country in time t; ln xjit is the logarithm of the jth input associated with the ith country in time t; t* is the time-trend representation of technical change; b, c and a are parameters to be estimated. In specific terms, btj 's are the factors related with inputs that provide information about input substi-tutability; y1 and y2 are the autonomous constituents of technical change which show a neutral-swing effect on the production function that is not attributed to any specific input; and aij 's are the biased technical change which is a scale expansion effect that influences efficiency when various inputs are employed.

In order to estimate expression (3), a number of conditions are supposed to be met. Firstly, symmetry is imperative. That is to say, Young's theorem must be fulfilled and this entails that bjk = bkj for all j, k. This denotes that the production structure in expression (3) has 1 neutral-scale factor (b0), j + 2 first-order factors (by, C1, C2) and (j + 1)(j/2) + j second-order factors (bjk, aj). Secondly, the marginal products of inputs should have a monotonicity or should be positive, i.e. MPjtt = 0yit/0 Xjit > 0. According to the translog production function, we compute the marginal product of input j by multiplying the average product of input j by the logarithmic marginal product. Hence, the equation below has to be positive to ensure that the monotonicity condition of the translog specification is met.

As noted, the general Cobb-Douglas expression employed in macroeconomic computation is;

Y = K aLb (2)

where K = capital and L = labor.

When the model exponents sum to one, the production function is first-order homogeneous, this denotes constant returns to scale. That is, when input doubles, output has to double.

In the real world production scenario, it is the case that a number of other variables including energy, water resources, etc also play a significant role on sustainable development. It is therefore attractive to apply models that yield close approximations of real world issues. For this purpose, we used the transcendental logarithmic production model which is considered to be the most flexible, functional form in applied production analysis (Wesseh et al., 2013; Wesseh and Lin, 2014).

j 1 J K

ln Yit = b0 + bj ln Xju + J2bjk ln xjit ln xkit

2 j=1 k=1

+ c/ +1C21*2 + aj ln Xji,t* 2 j=1

In the equation above, i = 1, 2, ..., 38 represents the cross-sectional units; t =1,2, ..., 21 indicates the time

@yjL_ylL @lny,t

■jit

Xjit' d ln Xjit

— • bj + £ bjk ln Xkit + a/ > 0 jit k 1

From Eq. (4), it is evident that monotonicity would be determined based on the sign of the term in parenthesis given that both y and xj are positive numbers. Lastly, the marginal products of inputs as well should be decreasing in inputs,thus, satisfying the law of diminishing marginal productivity. Therefore, the following expression should be met:

(b + XX bjk ln Xku + a/j • ^ bj - 1 + XX bjk ln xku + a/j < 0

It is essential for investigators to verify the quantitative nature of the terms in Eqs. (4) and (5). Summarily, positive marginal products for each input calls for bj > 0 for all j whereas diminishing marginal productivities requires bj(bj - 1) < 0 for all j, if and only if 0 < bj <1 for all j. In general, if these conditions are satisfied for a sufficient number of observations, then the production function is said to be well-behaved. The specific form of the model we estimate is given by

ln 7, = bo + bx ln Kit + fc ln L t, + bE ln E u + bw ln W „

+ bxL ln Kit ln Lu + bxE ln Kit ln Eu + bxw ln K ln Wit + bxE ln Li, ln E + bxw ln Li, ln Wit + bEW ln E ln Wit + bKK (ln Kit )2 + bLL(ln Lit)2 + bEE(ln Eit)2 + bWW (ln Wft)2 1 2

+ y1 f + 2 y2i*2 + aK ln Kitf + aL ln Liti* + aE ln Eiti* + aW ln Witi*

2.3. Production structure

Given to the flexibility of the translog production function, it has gained increasing attention, especially in the field of applied production. The translog estimation is considered the most superior functional form due to its ability to approximate a number of popular models. For our studies, the translog estimation is principally advantageous since output and substitution elasticities are allowed to fluctuate with the levels of inputs, therefore homotheticity is not applied. By definition, a production function is homothetic where the marginal rate of technical substitution is homogenous of degree zero in inputs. The translog production function is homogenous of degree k in inputs if:

Yb, = k,^bjk = 0 and ^a, = 0

If k = 1 in the above expression, then the production technology is said to demonstrate constant returns to scale or linear homogeneity. That is, when input doubles, output has to double. Similarly, the translog production model is strongly considered separable if j = 0 for all j, k and this corresponds to a Cobb-Douglas function with input-biased technical change. Conversely, if j = 0, a = 0 and y2 = 0 for all j, k, then expression (3) would follow a Cobb-Douglas technology with Hicks-neutral technical change.

The expressions below indicate three key parameter restrictions on (3) which shall be investigated in the empirical analysis:

= 0 (homotheticity)

y^b/ = 1 (linear homogeneity)

bjk = 0 (separability)

(9) (10)

According to Tzouvelekas (2000), all the above hypothesis can be verified using any of the available conventional statistical tests. To this point, it is noteworthy that due to the flexibility of the translog estimation the value of output elasticities, returns to scale, elasticities of substitution and technical change do not require priori restrictions In the first place, the output elasticity of the jth input from Eq. (3) is given by:

Jgjit J @@ ]nx j=1 j=1 °

J2\bj + bjk ln Xk„ + j

j=1 V k=1

where gjit are the output elasticities that change with state of technology and input levels From the aforementioned, these are basically the logarithmic marginal product of the translog function and measure how output responds to a % age change in the jth input. Economics of scale estimates from the translog function is given as the sum of the output elasticities. Subsequently, the substitution elasticities that change with the quantity of inputs used are derived from the marginal elasticities, where MP is the marginal product, the relevant symmetric substitution elasticity between inputs j and k for the ith country in time t is given as:

rjk(«)

bjk(it) + (j/gkU)bkkl

"gjit + gte

Inputs j and k are substitutes if ojk > 0; independent if °>(«) = 0 or compliments if rjk(.() < 0. To this end, output elasticity with regard to the rate of technical change can be computed as:

dln yh

= C1 + C2f + }_Ja ln;

dXj=0 j=1

Technical change in the expression above is both time and country specific and also differs with the level of input. Whereas this measure is generally nonnegative, negativity could arise in the situation where countries encounter stiff or new regulations. Following Hicksian view, technical change is input j using if a > 0; neutral if a = 0 or saving if a <0. Tzouvelekas (2000) stresses that factor-increasing technical change with equal rates of augmentation is equivalent to Hicks-neutral technical change for constant returns to scale. Nonetheless, if there are no constant returns to scale, then a = 0 for all j is an essential requirement for Hicks-neutral technical change.

3. Results and discussion

To set the stage for a comprehensive empirical analysis, it is necessary to begin with an investigation of correlation among the major explanatory variables, especially given the fact that standard regression models are based on assumptions of no such correlations (i.e. multicollinearity). To do this, the approach in Kmenta (1986) is employed. According to Kmenta, a simple measure of the degree of correlation is obtained by regressing each of the independent variables on the remaining ones and using the resultant coefficient of determination (R2) as the measure. For instance, we regress capital on labor, energy and water; labor on capital, energy and water; energy on labor, capital and water; and water on labor, capital and energy. The calculated R2 for each regression suggested that multi-collinearity is not a serious issue in the applied data. Hence, parameter estimates from the translog model in this study have some degree of reliability.

These coefficients for SSA are reported in Table 1. It may be observed that all major parameters of the model are statistically significant at the conventional level. While labor and water utilization are having the expected signs, energy and capital have the opposite signs. In particular, the results show that for a unit increase in labor employment and water withdrawals economic growth in SSA countries is stimulated by 122% and 181%, respectively. Finding that water withdrawals boost economic activities and growth in SSA countries repost the result of primary interest and is reflective of the significant role that water plays in the various sectors of the economy as previously discussed. On the other hand, results pointing to the negative contribution of capital and energy to economic growth in African countries contradict economic theory, but seem to highlight important characteristics of most SSA countries. In other words, a large portion of society's resources is allocated for capital formation and generating energy services provision but converted into personal uses by corrupt politicians, thereby causing a string on the entire economy. Additionally, the raw material oriented nature of most African economies (rubber, cocoa, coffee, timber, oil, diamond, gold,...), low scale of manufacturing and heavy reliance on foreign aid all substantiate the meager role that capital contributes to these economies.

All major parameters of time are statistically significant, suggesting that the state of technological progress in SSA countries is skilled bias and factor augmenting. This means that, rather than actual technological progress in Africa countries, the state of technological change is driven solely by the efficiency with which various inputs such as energy, capital, labor and water are used.

So far, the results discussed provide parameter estimates of SSA countries in general (Table 2). Hence, for policy purposes and to effectively gauge the role of water resources utilization with individual countries and across time, growth estimates are reported for water utilization and labor employed. Capital and energy are excluded because of evidence that these variables are not drivers of growth in SSA countries over our sample period.

Results of the analysis of output elasticity are shown in Graph 1. The vast majority of these elasticities are positive suggesting the positive impact that continued labor and water resources may have on African economies. This means that in other to enhance sustainable growth in these countries, concrete policy decisions in support of more efficient resource consumption especially water and labor are necessary.

Attempts have also been made to calculate the output elasticity gap between water resource utilization and labor input as a means of facilitating policy decisions necessary for the efficient allocation of these resources. The expression we use to calculate the elasticity gap is given by;

OEGu = OELu ln K + OEWu

where OEGit is the output elasticity gap of the zth region i =1......4 at time t. Positivity of OEGit implies that a unit

Table 2

Parameter estimates of the translog model.

Variable Coefficient Std. error t-statistic Prob.

bo 16.54371*** 4.637953 3.567029 0.0004

bK -1.487905* ** 0.541432 -2.748091 0.0061

bL 1.224290** 0.561834 2.179097 0.0295

bE -0.599124* ** 0.213070 -2.811869 0.0050

bw 1.812721*** 0.151612 11.95635 0.0000

bKL -0.224628 0.779301 -0.288242 0.7732

bKE 0.208800*** 0.026231 7.959999 0.0000

bKL -0.007627 0.018045 -0.422694 0.6726

bKw -0.007627 0.018045 -0.422694 0.6726

bLE -0.137507* ** 0.025801 -5.329408 0.0000

bEW 0.124031*** 0.389137 0.515422 0.6064

bLW -0.078824* ** 0.018285 -4.310770 0.0000

bKK 0.137313 0.389224 0.352787 0.7243

bLL 0.102584 0.390283 0.262844 0.7927

bEE -0.026491* ** 0.009960 -2.659760 0.0079

bWW 0.024683*** 0.004495 5.491558 0.0000

Ci -0.166729* ** 0.018984 -8.782668 0.0000

C2 0.000114** 8.55E-05 1.332813 0.1829

aK -0.007383* ** 0.001735 -4.254943 0.0000

&L 0.016056*** 0.001770 9.072914 0.0000

aE -0.005789* ** 0.001094 -5.293507 0.0000

aw -0.003314* ** 0.000656 -5.054829 0.0000

* Indicates significance at the 1%, 5% and 10% levels, respectively.

increase in labor contributes much higher to economic development than a unit increase in water withdrawal. On the other hand, negativity of OEGit implies a unit increase in water withdrawals contributes much higher to economic growth than a unit increase in labor. Better still, if OEGu = 0, then labor and water withdrawals contributed especially to economic growth in SSA countries.

Results from the analysis as documented in Graph 2 mirrors Table 2. The bars symbolize the elasticity gap and they are wide and positive, indicating that a unit increase in labor use will contribute much higher to economic development than the same increase in water withdrawals. This is clearly demonstrated in the fact that the output elasticities of labor (green line) lie far above the elasticity of water (blue line). The findings here provide that central Africa has the highest dependence on labor.

Turning attention to the state of technological progress, the results in Table 3 certainly demonstrate that innovation in technology is driven mainly by the various factors of production especially labor as opposed to a neutral shift in economic structure. We have profiled in Table 2 that this innovation was done using labor (positivity of coefficient values) and savings in water withdrawals, energy and capital (negativity of coefficient value). This suggests the need for more innovation in the water withdrawing procedures and processes so as to maximize the potential impact that these resource could have on various African economies.

Finally, the study investigates and estimates the possibility of substituting water resources for labor or vice versa. This analysis comes in well as many SSA countries not only suffer limited supply of freshwater but the relevant

Water withdrawals

Lesotho Botswana Namibia Tanzania Swaziland South Africa

Ethiopia Rwanda Burundi Malawi Madagascar Kenya Comoros Mauritius Uganda Tanzania Mozambique

Central African Cameroon Gabon Equatorial Guinea Congo, Rep. Congo, Dem. Rep.

Liberia Benin Togo Sierra Leone Senegal Nigeria Niger Mauritania Mali Liberia Guinea Ghana Gambia

Lesotho Botswana Namibia Tanzania Swaziland South Africa

Ethiopia Rwanda Burundi Malawi Madagascar Kenya Comoros Mauritius Uganda Tanzania Mozambique

Chad Central African .. Cameroon Gabon Equatorial Guinea Congo, Rep. Congo, Dem. Rep.

Liberia Benin Togo Sierra Leone Senegal Nigeria Niger Mauritania Mali Liberia Guinea Ghana Gambia

Southern Africa | Eastern Africa Central Africa Graph 1. Output elasticity.

Western Africa

technology to ensure optimum and efficient use is also lacking. This creates opportunities for analysis that investigates the potential of substituting crucial but limited resources for relatively available ones like labor in this case.

Judging from Graph 3, the study unravels evidence that SSA countries have a great potential of substituting a portion of their limited water resources for labor and thus, maintaining the ability to stimulate economic growth while also mitigating environmental pollution coming at the hands of water withdrawal processes. Opportunities to substitute labor for water can be done indirectly. For instance, where the amount of throughput is reduced especially for energy production, more labor will be required to otherwise perform jobs that powered machines would have done. This implies that, putting a price ceiling on water will reduce labor intensity.

4. Conclusions

Output elasticity gap - Water withdrawals - Labor |

In this paper, a trans-log production function model for Graph 2. Output labor elasticity. SSA economies is established and input factors such as

Table 3

Technological change.

N Country

Autonomous Biased

1 Benin -0.16485 2.331112 2.166264

2 Botswana -0.16485 2.601292 2.436444

3 Burkina Faso -0.16485 2.607878 2.44303

4 Burundi -0.16485 2.596436 2.431588

5 Cameroon -0.16485 2.753018 2.58817

6 Central African Republic -0.16485 2.877727 2.712879

7 Chad -0.16485 3.079611 2.914763

8 Comoros -0.16485 2.795837 2.630989

9 Congo, Dem. Rep. -0.16485 2.964998 2.80015

10 Congo, Rep. -0.16485 2.949719 2.784871

11 Cote d'Ivoire -0.16485 2.86526 2.700412

12 Equatorial Guinea -0.16485 3.299026 3.134178

13 Eritrea -0.16485 2.804221 2.639373

14 Ethiopia -0.16485 3.026181 2.861333

15 Gabon -0.16485 3.167507 2.905676

16 Gambia -0.16485 3.073809 3.002906

17 Ghana -0.16485 3.035621 2.870773

18 Guinea -0.16485 3.000683 2.835835

19 Kenya -0.16485 2.999008 2.83416

20 Lesotho -0.16485 3.254567 3.089719

21 Liberia -0.16485 3.024242 2.859394

22 Madagascar -0.16485 2.919554 2.754706

23 Malawi -0.16485 3.031324 2.866476

24 Mali -0.16485 3.154547 2.989699

25 Mauritania -0.16485 3.093662 2.928814

26 Mauritius -0.16485 3.137234 2.972386

27 Mozambique -0.16485 3.061785 2.896937

28 Namibia -0.16485 4.191406 4.027128

29 Niger -0.16485 3.199238 3.03439

30 Nigeria -0.16485 3.043147 2.878299

31 Rwanda -0.16485 3.206646 3.041798

32 Senegal -0.16485 3.151139 2.986291

33 Sierra Leone -0.16485 3.058539 2.893686

34 South Africa -0.16485 3.083192 2.918344

35 Swaziland -0.16485 3.194114 3.029266

36 Tanzania -0.16485 3.186219 3.021371

37 Togo -0.16485 3.208009 3.043161

38 Uganda -0.16485 3.137647 2.972799

capital, energy and labor are included. The translog production model has been extensively used in the applied production economics literature due to its flexibility and superiority over other functional forms. This study presents a comprehensive outline on the estimation properties of the translog production function in the context of country-level panel data for 38 African countries. Also, output elasticity of each factor and substitution elasticity between each variable are analyzed. Several results have been documented from the investigation. First, the results suggest that economic growth in SSA is driven mainly by water and labor. Capital and energy were found not to significantly drive economic growth (this substantiated the assertion of Wolde-Rufael (2009), that energy plays a meager role in the economic development of African countries). Second, technical change in SSA is scale-biased and factor augmenting; suggesting that efficiency of water withdrawals and labor use would lead to technological progress in SSA. Third, substitution possibilities between water and

Water - Labor

Uganda Togo Tanzania Swaziland South Africa Sierra Leone Senegal Rwanda Nigeria Niger Namibia Mozambique Mauritius Mauritania Mali Malawi Madagascar Liberia Lesotho Kenya Guinea Ghana Gambia Gabon Ethiopia Eritrea Equatorial Guinea Cote d'Ivoire Congo, Rep. Congo, Dem. Rep. Comoros Chad

Central African Republic Cameroon Burundi Burkina Faso Botswana Benin

Graph 3. Estimate elasticities of substitution.

labor exist indicating that restrictions on water withdrawals would lead to labor intensiveness and vice versa. Finally, a more general insight from the study is that efficient use of water resources promotes technological innovation and hence, critical for sustainable development.

This piece of word in no measure looks at the deal influence of foreign direct investment (FDI) on economic growth and water resources consumption. It is therefore recommended that further studies should shed light in this domain, given the fact most SSA countries rely heavily on donor assistance (aid Moreover, a comprehensive cost analysis between water resource consumption, sustainable economic growth and the impacts on the biophysical environment would form a valuable avenue for future research and model development.

Acknowledgments

The authors would like to thank Kingsley Etornam Dogah, Osire Tolbert, Bongajum Simplice Ngoran, Ako

Rajour Tanyi, Ehizuelen Michael Mitchell Omoruyi (Dr.) and Yangfan Li (Dr.) for their valuable comments and suggestions on the earlier version of the paper. We're also grateful to the role played by two anonymous reviewers.

References

Akhtar, M., Ahmad, N., Booij, M.J., 2008. The impact of climate change on the water resources of Hindukush-Karakorum-Himalaya region under different glacier coverage scenarios. J. Hydrol. 355 (1), 148-163.

Banya, K., Zajda, J., 2015. Globalization, the brain drain, and poverty reduction in Sub-Saharan Africa. In: Second International Handbook on Globalisation, Education and Policy Research. Springer, Netherlands, pp. 263-285.

Bindra, S.P., Hamid, A., Salem, H., Hamuda, K., Abulifa, S., 2014. Sustainable integrated water resources management for energy production and food security in Libya. Procedia Technol. 12, 747-752.

Bohmelt, T., Bernauer, T., Buhaug, H., Gleditsch, N.P., Tribaldos, T., Wischnath, G., 2014. Demand, supply, and restraint: determinants of domestic water conflict and cooperation. Global Environ. Change 29, 337-348.

Cazcarro, I., Duarte, R., Sanchez-Choliz, J., 2013. Economic growth and the evolution of water consumption in Spain: a structural decomposition analysis. Ecol. Econ. 96, 51-61.

Chapagain, A.K., Orr, S., 2009. An improved water footprint methodology linking global consumption to local water resources: a case of Spanish tomatoes. J. Environ. Manage. 90 (2), 1219-1228.

Dawadi, S., Ahmad, S., 2013. Evaluating the impact of demand-side management on water resources under changing climatic conditions and increasing population. J. Environ. Manage. 114, 261-275.

Fereres, E., Soriano, M.A., 2007. Deficit irrigation for reducing agricultural water use. J. Exp. Bot. 58 (2), 147-159.

Howell, T.A., 2001. Enhancing water use efficiency in irrigated agriculture. Agron. J. 93 (2), 281-289.

Katz, D., 2014. Water use and economic growth: reconsidering the Environmental Kuznets Curve (EKC) relationship. J. Cleaner Prod.

Kmenta, J., 1986. Elements of Econometrics. Macmillan Press, New York.

Madulu, N.F., 2003. Linking poverty levels to water resource use and conflicts in rural Tanzania. Phy. Chem. Earth Parts A/B/C 28 (20), 911-917.

Mbaiwa, J.E., 2004. Causes and possible solutions to water resource conflicts in the Okavango River Basin: the case of Angola, Namibia and Botswana. Phy. Chem. Earth Parts A/B/C 29 (15), 1319-1326.

Ngoran, S.D., Xue, X., 2015. Addressing urban sprawl in Douala, Cameroon: lessons from Xiamen integrated coastal management. J. Urban Manage. 4 (1), 53-57.

Ngoran, S.D., Dogah, K.E., Xue, X., 2015a. Assessing the impacts of climate change on water resources: the Sub-Saharan Africa Perspective. J. Econ. Sustainable Dev. 6 (1), 185-193.

Ngoran, S.D., Yvette, B., Ngoran, B.S., Xue, X., 2015b. Circumscribing the nexus between natural disaster, water resources and poverty in Sub-Saharan Africa. J. Res. Dev. Manage. 7, 31-40.

Okadera, T., Okamoto, N., Watanabe, M., Chontanawat, J., 2014. Regional water footprints of the Yangtze River: an interregional input-output approach. Econ. Syst. Res. 26 (4).

Palazzoli, I., Maskey, S., Uhlenbrook, S., Nana, E., Bocchiola, D., 2015. Impact of prospective climate change on water resources and crop yields in the Indrawati basin, Nepal. Agric. Syst. 133, 143-157.

Piao, S., Ciais, P., Huang, Y., Shen, Z., Peng, S., Li, J., Fang, J., 2010. The impacts of climate change on water resources and agriculture in China. Nature 467 (7311), 43-51.

Reij, C P., Smaling, E.M.A., 2008. Analyzing successes in agriculture and land management in Sub-Saharan Africa: is macro-level gloom obscuring positive micro-level change? Land Use Policy 25 (3), 410420.

Salami, H., Shahnooshi, N., Thomson, K.J., 2009. The economic impacts of drought on the economy of Iran: An integration of linear programming and macroeconometric modelling approaches. Ecol. Econ. 68 (4), 1032-1039.

Sallam, O.M., 2014. Water footprints as an indicator for the equitable utilization of shared water resources: (Case study: Egypt and Ethiopia shared water resources in Nile Basin). J. Afr. Earth Sci. 100, 645-655.

Tir, N.A., Momeni, F., Boboevich, G.T., 2014. Consideration the effects of water sector investment in economic development in Iran. J. Econ. Sustainable Dev. 5 (5), 151-159.

Tzouvelekas, E., 2000. Approximation properties and estimation of the translog production function with panel data. Agric. Econ. Rev. 1, 3347.

Urama, K.C., Ozor, N., 2010. Impacts of climate change on water resources in Africa: the role of adaptation. Afr. Technol. Policy Stud. Netw. 29.

Varbanov, P.S., 2014. Energy and water interactions: implications for industry. Curr. Opin. Chem. Eng. 5, 15-21.

Wang, S., Yang, F.L., Xu, L., Du, J., 2013a. Multi-scale analysis of the water resources carrying capacity of the Liaohe Basin based on ecological footprints. J. Cleaner Prod. 53, 158-166.

Wang, Z., Huang, K., Yang, S., Yu, Y., 2013b. An input-output approach to evaluate the water footprint and virtual water trade of Beijing, China. J. Cleaner Prod. 42, 172-179.

Wesseh, P.K., Lin, B., 2014. Renewable energy technologies as beacon of cleaner production: a real options valuation analysis for Liberia. J. Cleaner Prod.

Wesseh, P.K., Lin, B., Appiah, M.O., 2013. Delving into Liberia's energy economy: technical change, inter-factor and inter-fuel substitution. Renewable Sustainable Energy Rev. 24, 122-130.

Wesseh, P.K., Lin, B., 2016a. Factor demand, technical change and interfuel substitution in Africa. Renewable Sustainable Energy Rev. 59, 979-991.

Wesseh, P.K., Lin, B., 2016b. Output and substitution elasticities of energy and implications for renewable energy expansion in the ECOWAS region. Energy Policy 89, 125-137.

Wolde-Rufael, Y., 2009. Energy consumption and economic growth: the experience of African countries revisited. Energy Econ. 31 (2), 217224.

World Bank Group, 2012. World Development Indicators 2012. World Bank Publications.

Yu, Z., Qingshan, Y., 2014. Decoupling agricultural water consumption and environmental impact from crop production based on the water footprint method: a case study for the Heilongjiang land reclamation area, China. Ecol. Indic. 43, 29-35.