Is Sub-Saharan Africa on a Genuinely Sustainable Development Path? Evidence Using Panel Data

Abstract

Although resource rich, sub-Saharan Africa (SSA) has in general been characterised by poor economic performance and widespread poverty. The region is, however, now poised to enjoy high levels of growth and is increasingly attracting foreign direct investment (FDI). However, it is important to determine the sustainability of this development path. Given the lack of research on sustainability in the context of SSA, this study attempts to bridge this gap. A capital approach is adopted using the genuine savings (GS) rate computed by the World Bank, a measure of weak sustainability. GS endeavour to assess the sustainability path of countries, based on how ‘well’ they manage their total capital stock through the difference between consumption in natural capital and counter-balancing investments in other forms of man-made capital, namely physical and human. Since GS is based on the assumption of perfect resource substitutability, it can be taken as a limit value of sustainability whereby a country experiencing a positive value of GS is deemed to be weakly sustainable. This article thus aims to investigate whether SSA is on a sustainable development path and the factors affecting GS for this sample of countries. This study looks at a panel data set of 30 SSA countries over a period of 35 years. The fixed, random as well as dynamic effects are taken into account using System-GMM. In particular, improving institutional quality in the countries considered in the sample could directly and significantly improve their weak sustainability.

JEL Classification: O11, Q01, B22, C23

Keywords

Sustainable Development Genuine Savings Rate Developing Countries sub-Saharan Africa Panel Data System GMM

1. Introduction

The concept of sustainable development was instituted in the development discourse in the 1980s. Perhaps the most frequently cited definition of sustainable development remains to this day, ‘development that satisfies the needs of the present without compromising the ability of the future to meet their own needs’ (Brundtland, 1987, p. 43).

However, sustainable development, although a widely used phrase, evokes different things to different groups of people (Barr, 2008; Dobson, 1996; Lele, 1991). An important area of contention in the literature pertains to two schools of thought based on the concepts of ‘weak’ and ‘strong’ sustainability (Haughton & Hunter, 1994; Neumayer, 2003, 2012). Proponents of weak sustainability contend that natural capital and man-made capital are substitutable while supporters of strong sustainability believe that they are not. Weak sustainability further proposes that growth is essential for environmental protection and relies on technological innovation to resolve environmental problems. On the other hand, strong sustainability proponents criticise this, highlighting that human-made capital cannot replace essential forms of natural capital such as the ozone layer.

Weak and strong sustainability are therefore based on notions of capital (Davies, 2013). It is the ‘type’ of capital that matters; strong sustainability entails passing on the same amount of natural capital and increased man-made capital to the next generation, while weak sustainability entails a diminishing natural capital stock over time but an increase in man-made capital (Dasgupta, 2004, 2007).

Pezzey and Toman (2002, p. 213) argue that ‘theoretical work has vastly outstripped empirical work’ in the area of the economics of sustainability. This article’s main research objectives are to investigate whether sub-Saharan Africa (SSA) is on a sustainable development path using the genuine savings (GS) rate, a measure of weak sustainability, computed by the World Bank and the factors affecting this rate for this sample of countries. Assessing weak sustainability is not merely a theoretical exercise but is critical, as traditional national income accounting indicators, such as GDP growth, which do not completely reveal the trade-offs that may exist between different forms of capital (Dasgupta, 2001) and focus only on economic growth, specially which is generated by using up natural capital, may in fact lead to a decrease in the future well-being of populations.

The choice of SSA for this study has been motivated by several reasons. Many countries in the region have sustained 5–6 per cent growth rates for more than a decade (IMF, 2014), a much higher rate than other parts of the world, and the trend is projected to accelerate in future years. It is thus important to investigate whether the economic performance of the region is at the expense of its natural capital, as a large part of the African success story is based on the fact that it has become one of the world’s fastest growing regions for foreign direct investment (FDI) (fDi Intelligence, 2014) with resource-seeking multinationals (MNEs), for instance, in the oil and gas sectors, representing approximately one-third of the total.

Furthermore, the empirical literature on sustainability in the region is scant and this article contributes in bridging this gap. Hamilton and Clemens (1999) came up with empirical estimates for the GS rate retrospectively for a large number of countries from 1970 and report that GS rates are negative in general for SSA countries, suggesting these countries are progressively being impoverished in terms of their natural and total capital stock. This has important implications for future generations: a deterioration of natural capital by current generations will affect the ability of future generations to meet their needs. This article thus aims to extend the current empiricism related to sustainable development by making use of a panel dataset of developing countries, namely SSA countries, characterised in general by natural resource endowments but also by poor institutions. The study provides insights into the complexities of achieving more sustainable development and enables the formulation of pertinent policies.

The article is structured as follows: Second section gives an overview of literature pertaining to the GS rate as a measure of weak sustainability; third section presents the empirical strategy along with the data and model specification; fourth section presents the empirical findings, analysis and discussion; and fifth section concludes.

2. The Weak Sustainability Literature

As far back as 1776, Adam Smith had argued that the pre-requisites for economic growth included natural capital endowments as well as physical and human capital. The roots of the weak sustainability paradigm can be traced back to works of two neoclassical economists namely Solow (1974, 1993) and Hartwick (1977, 1990). The ‘Hartwick rule’ contends that the net investment in physical, human and natural capital should be positive for a satisfactory and enduring growth performance, and that a country’s total capital stock should at least remain constant over time. The discussion on how to evaluate this capital growth beyond the traditional measure of GDP led, among other developments, to the concept of GS (Boos & Holm-Müller, 2012).

Development can in fact be viewed as a process of building and managing a portfolio of assets (World Bank, 2011). The challenge is not only to manage the total volume of assets in terms of how much to save versus how much to consume, but also in terms of the composition of the asset portfolio by choosing the amount of investment in the different types of capital, including institutions and governance that forms part of the social capital.

The GS rate thus endeavours to assess the sustainability path of countries based on how ‘well’ they manage their total capital stock (Hamilton & Clemens, 1999). The weak sustainability paradigm and its measure, the GS, assume substitutability between the different forms of capital. The GS therefore measures the difference between the consumption of natural capital and investments in physical and human capital based on the ‘Hartwick rule’ (Hartwick, 1977), which provides a ‘rule of thumb’ for weak sustainability: Since the country’s total capital stock should be kept at least constant, countries should ‘invest into all forms of capital at least as much as there is depreciation of all forms of capital’ (Neumayer, 2010, p. 127).

The GS rate as a measure of weak sustainability has been developed by several researchers: Pearce and Atkinson (1993), Hamilton (1994), Pearce, Atkinson and Hamilton (1996) and Hamilton and Clemens (1999). It is based on the gross national income (GNI) and incorporates the rate of change of the three forms of capital, namely, man-made or physical (KP), human (KH) and natural (KN). Pearce and Atkinson (1993) started by working out a measure of savings which subtracted the depreciation of physical and natural capital as a measure of sustainability. Afterwards, Hamilton (1994) and Pearce et al. (1996) included investment in human capital and damages from pollution to the depreciation of natural capital. To distinguish between this indicator and the traditional net savings, which refers solely to physical capital, Hamilton (1994) named this indicator as GS. The formula for GS is given below:

GS = Investment in produced capital – Net foreign borrowing + Net official transfers – Depreciation of produced capital + Current education expenditures – Net depreciation of natural capital.

Critics of GS dispute the methodology for the calculation of resource depletion and the assumptions used in the concept and its measurement; however, GS constitutes a step in the right direction (Atkinson & Hamilton, 1996; Hamilton & Atkinson, 2006). The significance of GS stems from its capacity to show whether changes in the total capital stock are beneficial or harmful to future well-being and it has the advantage of being available for a large number of countries across time, which is not the case for most indicators of sustainability. GS have been calculated retrospectively for more than 160 countries from 1970 till date by the World Bank (World Bank, 2014). GS also recognise the trade-offs which occur during economic activity and is a concept that policymakers can respond to. It has been used as a measure of weak sustainability in a number of other studies (de Soysa & Neumayer, 2005; Dietz, Neumayer & de Soysa, 2007).

Since the GS is based on the assumption of perfect resource substitutability, it can be taken as a limit value of sustainability (Costantini & Monni, 2008), where a GS > 0 implies sustainability, a GS value of 0 indicates a minimum level of sustainability and a GS < 0 indicates non-sustainability. Countries with natural resource endowments, such as oil, have higher natural depletion rates, but policies such as spending on education and other investments in capital may more than counterbalance the depletion. Many resource-rich countries, however, seldom make investments in other forms of capital, using up the resource rents, thus endangering their future welfare and eroding their future development potential (Boos & Holm-Müller, 2012). However, no formal theoretical models have been developed for the determinants of GS except for a few empirical studies which test the impact of different sets of variables such as trade openness (de Soysa & Neumayer, 2005) and institutional quality (Boos & Holm-Müller, 2012; Dietz et al., 2007).

3. Methodological Approach and Definition of Variables

Boos and Holm-Müller (2012) highlight that the GS can be disaggregated into its different components namely natural capital, human capital and physical capital. The empirical equation used in this article comprises physical capital, human capital and natural resource capital and other conditioning variables which potentially affect the GS of a nation (Boos & Holm-Müller, 2012; de Soysa & Neumayer, 2005; Dietz et al., 2007). The baseline model comprising different forms of capital is subsequently adjusted by either adding or excluding variables in the light of existing theoretical arguments and empirical evidence on the GS.

The general framework used in this article conforms to the following specification:

\begin{matrix} Y_{i t} = β_{0} + β {}_{1}{(p h y s i c a l c a p i t a l)}_{i t} + β_{2} {(h u m a n c a p i t a l)}_{i t} \\ + β_{3} {(n a t u r a l c a p i t a l)}_{i t} + β_{4} {(o p e n n e s s)}_{i t} \\ + β_{5} {(F D I s t o c k)}_{i t} + β_{6} {(i n s t i t u t i ​ o ​ n s)}_{i t} + β_{7} {(i n c o m e i n e q u a l i t y)}_{i t} \\ + β_{8} {(w a r d u m m y)}_{i t} + β_{9} {(o i l d u m m y)}_{i t} + \\ β_{10} {(g l o b a l f i n a n c i a l c r i s i s d u m m y)}_{i t} + e_{i t} \end{matrix}

where,

Y_it is the GS growth rate.

The figures used for GS in this article have been sourced from the World Bank database (2014).

Definition of Variables

Physical Capital

de Soysa and Neumayer (2005) argue that since certain variables form part of GS, their inclusion would construct a partial identity between the left-hand side and the right-hand side of the equation. The measure of physical capital formation is therefore proxied by the rate of change in physical capital formation (de Soysa & Neumayer, 2005) instead of the gross fixed capital formation (GDFCF) as is customary in the literature.

Human Capital

Human capital is represented by educational attainment and is proxied by the secondary school enrolment rate (Barro, 2000).

Natural Capital

Natural capital is represented by natural resource rents. According to de Soysa and Neumayer (2005), natural resource rents capture the dependence of an economy on natural resources more accurately than the variables used previously in the literature, such as primary commodity exports, oil exports and reserves. Two categories of natural resources can be distinguished in the literature, namely point resources, such as food, agriculture and forestry, and diffuse resources comprising energy and mineral resources. The data on natural rents used in this study comprise mineral and energy rents, as a share of GNI. For lasting and weakly sustainable growth, the reinvestments of natural resource rents in physical and human capital should at least equal the depletion of natural capital, thus holding a country’s capital stock constant (Neumayer, 2010). Thus the expected signs for physical and human capital are positive while that for natural capital is expected to be negative especially for resource-rich regions.

Openness

Economic globalisation is tracked through the degree of trade openness and FDI penetration (Birdsall & Lawrence, 1999; Nye & Donahue, 2000). Trade openness is proxied by the sum of imports and exports divided by GDP, the usual proxy in the empirical literature.

FDI Stock

In addition, FDI stocks are used as a proxy for economic globalisation as they represent the accumulated investment in a host economy over time which captures the structural presence of MNEs better than do FDI flows alone (Bornschier & Chase-Dunn, 1985; de Soysa & Neumayer, 2005; Grimes & Kentor, 2003). FDI stock data are obtained from UNCTAD (2014). In terms of the expected signs, a pioneering study by de Soysa and Neumayer (2005) finds that economic openness proxied by trade openness and FDI stocks have positive effects on GS.

Institutions

de Soysa, Bailey and Neumayer (2012) show that democracy has a positive influence on GS while Dietz et al. (2007) show that the level of corruption has a significant negative impact. In this study, the institutional quality variables from the Worldwide Governance Indicators developed by Kaufmann and Kraay (2014) are used. They cover six dimensions of governance namely: voice and accountability measuring political, civil and human rights; political stability measuring the likelihood of violent threats to, or changes in, government, including terrorism; government effectiveness measuring the competence of the bureaucracy and the quality of public service delivery; regulatory quality measuring the incidence of market-unfriendly policies; rule of law measuring the quality of contract enforcement, the police and the courts, as well as the likelihood of crime and violence; and finally, control of corruption measuring the exercise of public power for private gain, including both petty and grand corruption and state capture. These indicators take values ranging from –2.5 to 2.5 inclusive, with an increase consistently implying better quality of institutions. An overall institutional quality indicator is calculated as the arithmetic mean of these six composite indicators for each country for each year. This practice has been used in several studies (Faria & Mauro, 2009; Thomas, 2010; Lagon & Arend, 2014). Better institutional quality is expected to have a positive impact on the GS rate.

Income Inequality

In a theoretical paper, Neumayer (2011) examines the channels that may link different forms of inequality to weak sustainability and he argues that a country characterised by higher levels of inequality is likely to experience lower levels of GS. A measure of income inequality, the Gini coefficient is included in the model. The inequality data have been taken from the UNU-WIDER World Income Inequality Database (WIID).

Dummy Variables

SSA has been plagued with wars and conflicts (Collier & Hoeffler, 2002) throughout its history and various studies have highlighted these facts and the negative impacts generated. A dummy variable (War Dummy) is added to the model for war years. The war years’ data are from the Uppsala/Prio database. Moreover, a dummy variable covering the period of the financial crisis from 2007 is also added to the model to capture any impact of the crisis on international capital flows and investment levels. Finally, a dummy variable is included for countries whose fuel oil exports make up more than 50 per cent of their GDP, as they can be considered to be highly dependent on resource rents (de Soysa & Neumayer, 2005).

This study is based on a panel data set of 30 SSA countries, compiled for a period of 35 years. The time horizon spans the period 1980–2014. The number of countries has been restricted by the availability of data. (The list of countries is given in Appendix A.) All data are taken from World Bank 2014 unless specified otherwise. All the data are logged to reduce skewness.

Panel data have been used as since it pools both cross-sectional and time series units, thus enabling insights from two perspectives, evolutionary trends across countries and across time. Furthermore, given that the group of countries is heterogeneous, panel data techniques enable control for such heterogeneity. Panel data regression techniques such as the fixed effects (FE) and random effects (RE) are first employed, and subsequently the System Generalised Method of Moments (GMM).

4. Empirical Findings, Analysis and Discussion

The RE and FE models were first estimated. The Hausman’s test was carried out to choose between them and the results recommend using the FE instead of the RE model. Four variables were significant at the conventional levels, namely natural resource rents, institutional quality and both of the globalisation variables—trade openness and FDI stock.

The coefficient of natural resource rents is negative and significant depicting a natural ‘resource curse’ for the GS growth rate (Boos & Holm-Müller, 2013) for this sample of countries for the time period considered. Substantial empirical evidence exists confirming that resource-abundant countries are often characterised by slower economic growth, known as the so-called ‘resource curse’ (RC) hypothesis. Very recently, a phenomenon strongly reminiscent of a resource curse has been detected with respect to GS by a few studies (Boos & Holm-Müller, 2013; Hamilton & Clemens, 1999). They find that countries with natural resources tend to have lower GS rates.

Furthermore, column 2 in Table 1 shows that contrary to the findings of de Soysa and Neumayer (2005) who argue that trade openness is positively related to GS, the coefficient of trade openness is significant but negative in this case. Therefore, higher trade openness is in fact decreasing the GS growth rate for the sampled countries under consideration. The reasons that can be put forward to explain this phenomenon include: trade openness could lead to the over-exploitation of the natural resources of developing countries and constrain the degree to which they are able to achieve sustainability (Zammit, 2003); and the revenue from the exploitation of natural resources in these countries is not sufficiently being re-invested into other forms of capital so that the sum of natural and man-made capital remains constant.

Table 1

Regression Estimates (Dependent Variable: GS Growth Rate)

	RE	FE	Sys GMM
Physical capital	0.0006	0.0005	0.0009
	(0.0004)	(0.0004)	(0.0003)**
Human capital	0.011	0.007	0.004
	(0.0148)	(0.0150)	(0.044)
Natural capital	–0.204	–0.080	–0.182
	(0.0421)***	(0.0431)*	(0.073)***
Openness	–11.056	–12.688	–2.845
	(3.309)***	(3.292)***	(4.437)
FDI stock	0.00003	0.00004	–3.43e–06
	(0.00002)	(0.00002)**	(0.0004)
Income inequality	0.061	–0.012	–0.109
	(0.609)	(0.064)	(0.421)***
Institutions	1.295	6.981	4.439
	(0.759)*	(0.745)***	(1.327)***
War dummy	–4.568	–0.243	–1.033
	(2.790)*	(2.797)	(1.408)*
Oil dummy	omitted	omitted	–2.236
	–	–	(0.726)**
GFC dummy	0.017	–0.384	0.177
	(0.702)	(0.688)	(1.459)
Lagged GS rate	–	–	0.588
	–	–	(0.257)***
Constant	2.972	8.807	10.931
	(3.609)	(3.394)***	(17.626)***
Overall R²	0.3361	0.4040	–
Wald test	57.99	–	1,631.82
F-statistic		5.57	–
p-value	(0.0000)	(0.0000)	(0.0000)
No. of countries	30	30	30
No. of observations	1,049	1,049	1,020

Source: Author’s estimates.

Notes: Standard errors are given below coefficients in parentheses. *, **, and *** denote significance at 10, 5 and 1 per cent, respectively. All regressions have a constant term. FE is the fixed effects model and RE refers to random effects model.

(1) The Hausman test was carried out to choose between the FE and RE models. The Hausman test: Accept H1 rejecting RE as the preferred model (χ² = 82.86 and p-value of 0.00 is significant).

(2) Hansen’s J test: Accept H₀ Over-identifying restrictions are valid (χ² = 0.168378 and p-value of 0.6816 is insignificant); Arellano-Bond test for AR (2) in first differences: accept H₀ of no autocorrelation (z = 1.41 and p-value of 0.1585 is insignificant).

The coefficient of the other economic globalisation variable, namely FDI stocks, which represents the presence of MNEs is also significant but conversely positive, thus higher levels of FDI stocks in the host countries are resulting, ceteris paribus, in higher levels of GS growth rates in line with de Soysa and Neumayer (2005). On one side, MNEs’ presence is likely to increase physical and human capital formation in the country and, on the other side, through the creation of jobs, the dependence on natural resources for survival by the poor is reduced.

The coefficient of the institutional quality variable turns out to be positive and highly significant. Thus, countries with better institutional quality are likely to experience higher GS growth rates.

However, the estimates from the FE model do not cater for potential endogeneity between the variables and should be interpreted with caution. Acemoglu & Johnson (2005) argue that political and economic development paths are interlinked and are jointly affected by various factors. Endogeneity arises if a regressor in a model has a cause-and-effect relationship, that is, the regressor influences the dependent variable and the latter also influences it simultaneously. As such, there is a two-way causality between the two variables (for example, better institutions may lead to a higher GS growth rate, but if institutions are to be included as a determinant of GS growth, it is important to control for potential endogeneity arising because GS growth may itself lead to better quality institutions or there might be a third variable such as natural resources that affects both institutions and GS growth). Furthermore, neither the FE nor the RE models tackle the correlation of the dependent variable with lagged dependent variables (for instance, growth is usually believed to be a function of its past levels and this could also be the case for GS growth), which makes the estimators inconsistent. This scenario therefore necessitates the use of a more efficient model.

Arellano and Bond (1991) came up with an estimator which provides consistent estimates for such models. This estimator is known as the ‘difference’ GMM estimator. The ‘difference’ estimator takes the first difference of the data and uses lagged values of the endogenous variables as instruments for the (differenced) variables. Such a procedure allows the elimination of country-specific effects as well as any endogeneity due to the correlation of country-specific effects and of the explanatory variables. However, Arellano and Bover (1995) and Blundell and Bond (1998) further argue that when the explanatory variables are persistent over time, the lagged levels become weak instruments for first differences. Blundell and Bond (1998) hence propose a more efficient estimator, known as the ‘system’ GMM estimator, which alleviates the problem of poor instruments by using additional moment conditions. This estimation method also corrects for any potential unobserved country heterogeneity, omitted variable bias and measurement errors, in addition to the endogeneity concerns. Therefore, in this study, the system GMM approach is used which generally produces more efficient, precise and reliable estimates compared to the difference GMM by improving precision and reducing the finite sample bias (Baltagi, 2008).

Column 3 of Table 1 reports the results generated by using the system GMM, where openness, FDI, institutions and Gini are treated as potential endogenous variables and the lagged dependent variable (lagged GS growth) is automatically included as an additional explanatory variable. The Hansen test statistic of over-identifying restrictions and the second-order test of no serial correlation in the difference residuals indicate that the instruments are valid and the model is correctly specified.

The coefficient of the lagged GS growth rate is positive and highly significant implying that past GS growth performance in these countries does play an important role in determining future GS growth rate.

The natural resource curse for GS growth is still apparent. For lasting and weakly sustainable growth, the reinvestments of natural resource rents in physical and human capital should at least equal the depletion of natural capital, thus holding a country’s capital stock constant (Neumayer, 2010). Resource-abundant countries deplete more natural capital and thus have to reinvest more in physical and human capital to be able to achieve positive GS growth rates. However, most resource-abundant countries consume from resource exports (Gelb, 2003; Auty, 1993) rather than re-invest the rents in physical capital, such as infrastructure, and human capital, such as education. Therefore, these countries are depleting their total capital stock which is likely to lead to declining future welfare.

The coefficient of the change in physical capital which is positive turns significant. This change is in accordance with theoretical expectations as the GS is based on the concept of substitutability between different forms of capital and therefore in order to increase the GS rate, more investments have to be done in man-made capital, namely physical and human capital. However, the coefficient of human capital is positive but not significant at conventional levels showing that the level of human capital in these countries is not contributing significantly towards improving GS growth rate and, therefore, these governments need to devote more resources towards human capital development in terms of provision of education and other public goods such as health facilities.

In terms of the coefficients of the globalisation variables, the degree of openness retains its negative sign while the FDI stock variable is negative, but both of them lose their level of significance.

The coefficient of the overall institutional quality variable remains positive and significant. Therefore, countries with better institutional quality are likely to experience higher GS growth rates. de Soysa et al. (2012) show that democracy has a positive influence on GS as democratic governments tend to exhaust natural capital less and invest more in human capital. However, democracy has to be backed by good quality institutions as well. In fact, Dietz et al. (2007) show that the level of corruption has a significant negative impact on the level of investments as well as on GS. Corrupt agents are more likely to use rents for their consumption instead of investments and political patronage to certain interest groups is also similarly likely to adversely influence GS.

The level of income inequality turns negative as well and is highly significant (Neumayer, 2011). The war dummy is significant and negative; unsurprisingly war lowers the GS growth rate (de Soysa & Neumayer, 2005) as resource rents might be diverted for military purposes. The oil dummy is negative and significant. Other studies (de Soysa & Neumayer, 2005; Hamilton, 2001;) also report that a high dependence on oil extraction has a strong negative impact on the GS growth rate illustrating once again a natural resource curse.

5. Conclusion and Policy Remarks

The GMM regression estimates show that natural resource exploitation, the level of income inequality, wars, over-dependence on exports of natural resources such as oil and poor institutional quality have a significant negative impact on the GS growth rate in SSA countries. Past GS growth performance in these countries also plays an important role in determining future GS rate growth. Therefore it would seem that even after almost two decades, SSA countries are still not on a weakly sustainable development path in line with Hamilton and Clemens’ (1999, p. 344) findings:

The savings analysis highlights the fact that the situation with regard to future well-being is worse than might otherwise be thought: not only has sub-Saharan Africa performed badly by conventional measures, it is clear that the wealth inherent in the resource stocks of these countries is being liquidated and dissipated (emphasis added in text).

Persistently negative GS rates are an indication of a country that is pursuing an unsustainable path, as revenues from the depletion of natural capital are not being re-invested in other forms of capital, and this has a negative impact on welfare and development in the long run. The natural and total capital stock is declining leading to an impoverishment of these countries.

The GS rate thus allows an analysis of national wealth, which can help a country adopt policies to ensure that this wealth is transformed into productive forms of capital. These policies could include macroeconomic policies that promote GS such as fiscal policies that enable the capture of resource rents and public investment programmes to re-invest resource revenues in other forms of capital, such as education, as well as resource policies conducive to efficient extraction rates. Improving institutional quality could directly and significantly improve weak sustainability.

Failure to implement sound policies could lead to the emergence of the ‘resource curse’. Atkinson and Hamilton (2003) find that countries that have escaped the resource curse utilised resource rents for investments rather than for current public expenditure. These countries have been transforming natural capital into manufactured capital, thus maintaining total capital intact in line with the Hartwick rule.

Footnotes

Appendix

Country
1. Angola
2. Benin
3. BOTSWANA
4. BURKINA FASO
5. BURUNDI
6. CAMEROON
7. CÔTE D’IVOIRE
8. ETHIOPIA
9. GABON
10. GAMBIA, THE
11. GHANA
12. GUINEA
13. KENYA
14. MADAGASCAR
15. MALAWI
16. MALI
17. MAURITANIA
18. MAURITIUS
19. MOZAMBIQUE
20. NAMIBIA
21. NIGER
22. NIGERIA
23. RWANDA
24. SENEGAL
25. SOUTH AFRICA
26. SUDAN
27. TANZANIA
28. TOGO
29. UGANDA
30. ZAMBIA

References

Acemoglu

& Johnson

(2005). Unbundling institutions. Journal of Political Economy, 113(5), 949–95.

Arellano

& Bond

(1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. The Review of Economic Studies, 58(2), 277–97.

Arellano

& Bover

(1995). Another look at the instrumental variable estimation of error-components models. Journal of Econometrics, 68(1), 29–51.

Atkinson

& Hamilton

(1996). Accounting for progress: Indicators for sustainable development. Environment: Science and Policy for Sustainable Development, 38(7), 16–44.

Auty

R. M.

(1993). Determinants of state mining enterprise resilience in Latin America. Natural Resources Forum, 17(1), 3–14.

Barr

(2008). Environment and society: Sustainability, policy and the citizen. Aldershot: Ashgate.

Barro

R. J.

(2000). Inequality and growth in a panel of countries. Journal of Economic Growth, 5(1), 5–32.

Baltagi

(2008). Econometric analysis of panel data. US: John Wiley & Sons.

Birdsall

& Lawrence

R. Z.

(1999). Deep integration and trade agreements. In Kaul

Grunberg

& Stern

M. A.

(Eds), Global public goods. International cooperation in the 21st century (pp. 128–151). New York, NY: Oxford.

10.

Blundell

& Bond

(1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics, 87(1), 115–43.

11.

Boos

& Holm-Müller

(2012). A theoretical overview of the relationship between the resource curse and genuine savings as an indicator for ‘weak’ sustainability. Natural Resources Forum, 36(3), 145–59.

12.

Boos

& Holm-Müller

(2013). The relationship between the resource curse and genuine savings: Empirical evidence. Journal of Sustainable Development, 6(6), 59.

13.

Bornschier

& Chase-Dunn

(1985). Transnational corporations and underdevelopment. New York: Praeger.

14.

Brundtland

(1987). Report of the world commission on environment and development: Our common future. Transmitted to the General Assembly as an annex to document A/42/427-Development and International Cooperation: Environment. Retrieved from http://www.un-documents.net/wced-ocf.htm

15.

Collier

& Hoeffler

(2002). Aid, policy and peace: Reducing the risks of civil conflict. Defence and Peace Economics, 13(6), 435–50.

16.

Costantini

& Monni

(2008). Environment, human development and economic growth. Ecological Economics, 64(4), 867–80.

17.

Dasgupta

(2001). Social capital and economic performance: Analytics. Beijer International Institute of Ecological Economics. Washington, D.C.: World Bank.

18.

Dasgupta

(2004). Human well-being and the natural environment. New York: Oxford University Press.

19.

Dasgupta

(2007). Economics: A very short introduction. New York: Oxford University Press.

20.

Davies

G. R.

(2013). Appraising weak and strong sustainability: Searching for a middle ground. Consilience: The Journal of Sustainable Development, 10(1), 111–24.

21.

de Soysa

& Neumayer

(2005). False prophet, or genuine savior? Assessing the effects of economic openness on sustainable development, 1980–99. International Organization, 59(3), 731–72.

22.

de Soysa

Bailey

& Neumayer

(2012). Free to squander? Democracy and sustainable development, 1975–2000. Cambridge, MA: MIT Press.

23.

Dietz

Neumayer

& de Soysa

(2007). Corruption, the resource curse and genuine saving. Environment and Development Economics, 12(1), 33–53.

24.

Dobson

(1996). Environment sustainabilities: An analysis and a typology. Environmental Politics, 5(3), 401–28.

25.

Faria

& Mauro

(2009). Institutions and the external capital structure of countries. Journal of International Money and Finance, 28(3), 367–91.

26.

fDiIntelligence. (2014). fDi Report. Retrieved from http://www.fDiintelligence.com

27.

Gelb

(2003). Inequality in South Africa: Nature, causes and responses. Johannesburg: Edge Institute.

28.

Grimes

& Kentor

(2003). Exporting the greenhouse: Foreign capital penetration and CO? Emissions 1980–1996. Journal of World-Systems Research, 9(2), 261–75.

29.

Hamilton

(1994). Green adjustments to GDP. Resources Policy, 20(3), 155–68.

30.

Hamilton

(2001). The sustainability of extractive economies. In Auty

(Ed.), Resource abundance and economic development (Chapter 3). WIDER Studies in Development Economics. UK: Oxford University Press.

31.

Hamilton

& Atkinson

(2006). Wealth, welfare and sustainability: Advances in measuring sustainable development. Cheltenham, UK: Edward Elgar.

32.

Hamilton

& Clemens

(1999). Genuine savings rates in developing countries. The World Bank Economic Review, 13(2), 333–56.

33.

Hartwick

J. M.

(1977). Intergenerational equity and the investing of rents from exhaustible resources. The American Economic Review, 67(5), 972–74.

34.

Hartwick

J. M.

(1990). Natural resources, national accounting and economic depreciation. Journal of Public Economics, 43(3), 291–304.

35.

Haughton

& Hunter

(1994). Sustainable cities, regional policy and development series 7. Regional Studies Association. London and Bristol, Pennsylvania.

36.

IMF. (2014). Regional economic outlook for sub-Saharan Africa—Staying the course. Retrieved from https://www.imf.org

37.

Kaufmann

& Kraay

(2014). World governance indicators. Washington, D.C.: World Bank.

38.

Lagon

M. P.

& Arend

A. C.

(2014). Human dignity and the future of global institutions. Washington, D.C.: Georgetown University Press.

39.

Lele

S. M.

(1991). Sustainable development: A critical review. World Development, 19(6), 607–21.

40.

Neumayer

(2003). Weak versus strong sustainability: Exploring the limits of two opposing paradigms. Cheltenham, UK: Edward Elgar Publishing.

41.

Neumayer

(2010). Weak versus strong sustainability: Exploring the limits of two opposing paradigms (3rd ed.). Cheltenham, UK: Edward Elgar Publishing.

42.

Neumayer

(2011). Sustainability and inequality in human development. Human Development Research Paper No. 4. New York: United Nations Development Programme, Human Development Reports.

43.

Neumayer

(2012). Human development and sustainability. Journal of Human Development and Capabilities, 13(4), 561–79.

44.

Nye

J. S.

& Donahue

J. D.

(2000). Governance in a globalizing world. Washington, D.C.: Brookings Institution Press.

45.

Pearce

D. W.

& Atkinson

G. D.

(1993). Capital theory and the measurement of sustainable development: An indicator of ‘weak’ sustainability. Ecological Economics, 8(2), 103–08.

46.

Pearce

Hamilton

& Atkinson

(1996). Measuring sustainable development: Progress on indicators. Environment and Development Economics, 1(1), 85–101.

47.

Pezzey

& Toman

M. A.

(2002). The economics of sustainability: A review of journal articles. Discussion Paper 02-03, Resources for the Future, Washington, D.C.

48.

Solow

R. M.

(1974). The economics of resources or the resources of economics. Classic Papers in Natural Resource Economics (pp. 257–76). UK: Palgrave Macmillan.

49.

Solow

R. M.

(1993). An almost practical step toward sustainability. Resources Policy, 19(3), 162–72.

50.

Thomas

(2010). High levels of corruption in South African business and the role of culture. In 10th annual BEN-Africa conference in Nairobi, Kenya.

51.

UNCTAD. (2014). World Investment Forum 2014: Investing in Sustainable Development, Geneva, Switzerland. United Nations Conference on Trade and Development Report. Retrieved from http://unctad-worldinvestmentforum.org/

52.

World Bank. (2011). World development report 2011: Conflict, security, and development . Washington DC.

53.

World Bank. (2014). Sub-Saharan Africa dataset. Retrieved from http://data.worldbank/org/region/SSA

54.

Zammit

(2003). Development at risk: Rethinking UN-business partnerships (pp. 1–328). Geneva: South Centre.