National Accounts estimates for Sri Lanka are made both by the Census and Statistics Department and the Central Bank of Sri Lanka . These estimates have been revised at regular intervals but the revisions have not been made retrospectively. This study presents for the first time a comparable National Accounts series for Sri Lanka at both current and constant prices covering the years 1950-2002. The perceived reliability of estimates of National Accounts depends on knowledge of the sources and methods used. For this reason this study sets out in detail the sources and methods used in each sector, evaluates these sources and methods and suggests areas where further improvements can be made. Separate Chapters analyse the effects of the ethnic conflict on the National Accounts estimates and the trends in GNP growth, structural change, investment and savings during the period 1950-2002. The study includes a Statistical Appendix of 47 tables which provides a comprehensive and comparable set of National Accounts data for the years 1950-2002.
Table of Contents:
Some Technical Aspects relating to Estimating A Long term Series of Gross National Product and Expenditure
An Evaluation of Sources and Methods used to Estimate GNP and GNE
The Influence of Methods used on GDP Growth
Effects of the Ethnic Conflict on Estimates of GNP and GNE
Economic Trends, 1950 – 2002
Sources and Methods
In the period of over 50 years since independence the Sri Lanka economy has gradually transformed itself from a dualistic economy dependent mainly on a few plantation export crops and subsistence agriculture to one where industry and services have become more important.
These changes in the structure of the economy can be realistically studied only with a comparable National Accounts series of Gross National Product (GNP) and Gross National Expenditure (GNE). Such a series is not available for Sri Lanka because revisions of National Accounts have not been made retrospectively.
This study makes an original contribution to estimates of National Accounts in Sri Lanka for the following reasons:
The GNP series at current and constant (1999) prices has been estimated retrospectively to cover the period 1950-2002.
A GNE series in real terms i.e., at constant (1999) prices has been estimated for the first time.
Estimates of real GNE presented in this study also include separate estimates of Private Consumption Expenditure and Investment Expenditure (Gross Domestic Capital Formation) at current market prices and at constant (1999) market prices.
For the first time a long term series of PCE estimated retrospectively has become available. The two series of Private Consumption Expenditure (PCE) i.e., “Private Consumption Expenditure by Object” and “Composition of Private Consumption Expenditure” conform as nearly as possible to the recommendations in A System of National Accounts (1993).
In this study the author has reviewed the sources and methods used in estimating GNP and GNE and has reworked the estimates. The tables presented here can therefore be considered as a new and comparable series.
The tables estimated for this study are presented in a separate Statistical Appendix. It is intended that the data in these tables will serve as a useful quantitative reference work.
The rest of this study is divided into 8 Chapters as follows:
Chapter 2 comments on the technical aspects of estimating a comparable series of National Product and Expenditure at both current and constant prices. In order to put the series presented in this study in perspective, a brief description of the early development of Sri Lanka’s National Accounts estimates is also given. However the focus is mainly on the Central Bank’s estimates.
Chapter 3 takes a critical look at the sources and methods used in estimating GNP and GNE in Sri Lanka and makes some suggestions for improvement of the estimates.
In Chapter 4 the study has separated estimates of GDP that have been made using direct methods from estimates using indirect methods. It has summed up the estimates using direct methods under the three main sectors – Agriculture, Industry and Services. The study compares growth rates of that part of GDP where only direct methods have been used with the growth rates of total GDP. This provides a useful insight into the influence of direct and indirect methods on GDP growth rates.
Chapter 5 looks at the likely distortions in the estimates resulting from the ethnic conflict that has affected mainly the Northern and Eastern Provinces of Sri Lanka for two decades after about 1983.
The availability of a comparable series going back 50 years enables one to examine the pace and pattern of economic growth. Hence, Chapter 6 looks at economic trends in Sri Lanka over this period. However, the focus in this section is on changes in the statistical relationships rather than on an economic analysis of these trends.
Chapter 7 sets out the sources and methods used in making the estimates presented in this study. This is done in the form of separate tables which set out for each item the source of data and the method adopted to arrive at value added at both current and constant prices.
The main references used in this study are set out in Chapter 8.
The Statistical Appendix is at Chapter 9.
by Terrence Savundranayagam