Academic journal article The Journal of Developing Areas

Savings and Growth in Bangladesh

Academic journal article The Journal of Developing Areas

Savings and Growth in Bangladesh

Article excerpt

ABSTRACT

Savings and growth are often strongly correlated. Thus, it is important to study the determinants of savings and the direction of causality between savings and growth as these have important implications for development policy. Yet, careful empirical studies on savings behavior are missing for Bangladesh. Thus, in this paper, we estimate the long-run total and private savings functions for Bangladesh using recent time series techniques. We find that the total savings rate is mainly determined by the GDP growth rate, dependency ratio, interest rates and bank density. The private savings rate is also affected by the public savings rate. Further, using the Granger Causality tests, we find that in Bangladesh, there is a bi-directional causality between savings and growth. We also carry out the Forecast Error Variance Decomposition (FEVD) analysis using the VAR framework. The FEVD results confirm the causality results obtained using the Granger causality tests as well as the estimated savings functions.

JEL Classifications: E21, O16

Keywords: Savings, Growth, Bangladesh, Causality, Co-integration

(ProQuest: ... denotes formulae omitted.)

INTRODUCTION

Savings and growth are often strongly correlated. Thus, it is important to study the determinants of savings and the direction of causality between savings and growth as these have important implications for development policy. There are some recent studies (see Loayza et al. 2000; Lopez et al. 2000; Schmidt-Hebbel et al. (1996) on the determinants of savings behavior in pooled time series cross-section data on a large number of countries. However, savings behavior shows considerable variation across countries depending on their socio-economic structure and one cannot be sure whether the results of such pooled studies, which are applicable to the average country in the sample, apply to the particular country in question. Thus, cross country regression analysis based on the assumption of homogeneity is not appropriate for each sample country. Given the vast differences among countries with respect to structural features, institutional aspects and also the quality of data, cross-country comparison and estimation is fraught with danger (Deaton, 1989).

It is evident therefore that country specific studies are also important. Taking the example of Bangladesh, a proper empirical study using modern time series techniques is missing (with the exception of Chowdhury, 2001). While the emphasis in Chowdhury's paper is on the financial sector and the impact of financial liberalization on private savings in Bangladesh, the emphasis in our paper is on analyzing the direction of causality between savings and growth, in addition to the overall determinants of savings. Although we do briefly examine the impact of financial development (measured through the number of bank branches per capita, and financial policies, such as the interest rates on bank deposits and broad money) on the savings rate, our treatment of the financial sector and of the impact of financial liberalization on the savings rates is not as detailed as that of Chowdhury. Thus, our study complements Choudhury's and the two studies together provide a fuller analysis of the various aspects of the savings-growth relation.

The previous literature on the determinants of both domestic savings and private savings is very rich. Many previous authors have analyzed the specific factors affecting the savings behavior in different countries1. Trying to examine the determinants of private and public savings in 36 Latin American countries, Edward (1996) finds that per capita income is the most important determinant of private savings along with the demographic structure, social security expenditure and the depth of the financial sector. The strong positive relationship between savings and income have been found by Lahiri (1989) and Dayal and Thimann (1997). But public savings is less affected by these factors. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.