The impact of institutional factors on the performance of genuine savings

ABSTRACT

This study investigates institutional factors affecting the performance of genuine savings (GS), which are often used in assessing sustainable development, and adopts a model of autoregressive conditional heteroscedasticity in mean. We pay particular attention to the contribution of institutions to decrease the volatility level of the GS path. The estimation results show that there are two ways in which institutions affect GS performance. First, high quality of institutions enhances the GS level directly. Second, high quality of institutions enhances the GS level via stabilizing the volatility of the GS path. Considering both effects in their totality, institutional improvement plays an important role in realizing a sustainable development path.

KEYWORDS:

• Sustainable development
• genuine savings
• volatility
• institution

Acknowledgments

We thank Catia Cialani, Shunsuke Managi, Ayumi Onuma, and Kazuhiro Ueta for their helpful comments.

Disclosure statement

No potential conflict of interest was reported by the authors.​​​​

Notes

1. This concept is classified as ‘weak sustainability’, in which it is conceded that there is substitutability among the different types of capital. For details on strong and weak sustainability, see Neumayer (2000).

2. Research related to GS has addressed the change in inclusive stock via different terminologies, such as ‘inclusive investment’ (Dasgupta 2007) and ‘adjusted net savings’ (World Development Indicators provided by the World Bank). A recent research project developed the concept of the ‘inclusive wealth index’ and stressed the difference between this concept and GS in terms of theoretical assumptions and empirical techniques (UNU-IHDP 2012, 2014). In this study, we focus on the path of GS and utilize the data from the World Bank owing to its richness.

3. Dasgupta (2004) included knowledge capital in Equation (1). However, as mentioned in Section 3, our data do not include knowledge capital because the available database for GS is based on only three types of capital (manufactured capital, human capital, and natural capital) in Equation (1).

4. Regional specifications are based on the World Bank’s classification.

5. Available from http://databank.worldbank.org/

6. Only the results for these four institutional variables are reported in Table 2 to save space. We obtain similar results for the other eight institutional variables. The results are available upon request.

7. This is calculated from Equations (7) and (9) as

where X denotes the institutional variables, that is, Corruption; β₅ is a direct effect (i.e. the coefficient of institutions in the mean regression); γ is a coefficient vector, (γ₁, γ₂, γ₃, γ₄, γ₅, γ₆, γ₇, γ₈); and Z is the variable vector (ggdp, ggdp², inflation, trade, gov_size, region_dummy1, region_dummy2). A similar procedure can be adopted for the other institutional variables.

8. Although there are 12 institutional variables, we select 1 institutional variable when calculating the total effect of the control variables. In addition, we calculate the total effect of trade openness for the other 11 institutional variables and find that the total effect is smaller than its direct effect.

Additional information

Funding

We greatly appreciate financial support from the JSPS Research Fund KAKEN (26000001).​​​​