Savings Rate A Rashomon Analogue

Let the facts speak first. Gross national savings ratio, which was around 23.6 per cent in 1990-91, declined to 21.2 per cent in 1992-93, mid-year of the five-year period beginning 1990-95. This means that a fall in savings rate was not only not precipitous but also temporary and soon recovered to a level perhaps the highest ever. Its composition shows that the decline was mainly accounted for by a sharp fall in public savings which could not be traced to the structural reform per se.

Undoubtedly, the savings rate was marred by fluctuations in the household sector savings. What needs to be determined is whether the gyrations in savings rate and its near stagnancy are a reflection of a statistical mirage or the changes in the Indian economy. The household sectors saving is decomposed into two segments in the form of physical assets and financial assets. A sharp decline in household sector savings is explained by a steady fall in its physical component while the other financial component increa-sed since 1990 except in 1992. There seem to be strong reasons to believe that the fluctuations may have been a consequence of the way savings and investments are estimated.

 

The procedures adopted to measure investment and savings are not independent of each other. Therefore the biases in one are carried into the other, particularly in regard to the estimation of the physical component of household sector savings. Aggregate investment is estimated using a commodity flow methodology. It is then decomposed into government, private corporate and hou-sehold investment, by deducting from it government investment; the residual is household investment. Aggregate saving is arrived at by the summation of government saving, corporate and household saving and foreign saving. Except the physical asset component of household savings all others are independently estimated. Though savings and investment figures should be equal, they are not so with a statistical discrepancy of around 2 per cent of GDP. The Central Statistical Organisation (CSO) which considers the savings estimate more reliable than investment estimate, adjusts the latter for errors and omissions, which is not strictly warranted.

To remove the biases from the household savings rate, two alternative routes can be used. First, the savings estimate is adjusted to equate it to investment by adding errors and omissions to savings rather than to investment. Second, given that physical savings pick up a large part of the errors, they can be re-estimated differently in relation to the other reliable saving component financial saving. Estimates derived by applying these procedures show that the recent gyrations in domestic savings in India may have been exaggerated. The adjusted savings considerably moderates a decline in early 1990s, contrary to what the CSO estimate has revealed.

Lack of fluctuations in the savings ratio and the trends therein highlighted by the alternative measures, help dispose of the argument that the domestic savings rate has fallen mainly due the upsurge in consumption of durable goods production which increased after the reforms. There is empirical evidence to counter this argument. As pointed out by Mr Balkrishnan of CSO, the proportion of final consumption expenditure on durable and semi-durable goods in total private consumption expenditure went down from 14.4. per cent in 1991-92 to 13.7 per cent in 1993-94. This has also been corroborated by the trends in the production of durable and semi-durable goods.

It could be plausibly argued that though the domestic savings rate was not as depressed as was widely believed or that the fluctuations were small, the Indian savings rate did not show any perceptible increase since the reforms. Going by Indias historical experience since the 1950s and the insights provided by inter-country studies, the obvious factor holding back private savings rate from rising has been the measly rate of GDP growth in the aftermath of the 1991 reforms. During the 80s, the average GDP growth rate was around 6.2 per cent. In 1990-91, GDP growth was 5.4 per cent. There was subsequently a dip in this rate i.e. 1991-92 and 1992-93. Mr Balkrishnan tested a hypothesis of the household savings ratio and the rate of growth for India, using data for a period 1950-51 to 1990-91. He found that a deviation from the trend value of the household savings ratio had been a function of the rate of growth, with the Indian economy getting on a higher growth trajectory. After the strengthening of structural reforms, the rate of growth could steadily move from its current level to anywhere between 6 and 7 per cent in the years ahead, bolstering the household savings rate. If a higher rate of growth is accompanied by a rising per capita income as is likely, the household savings rate growth would be further reinforced as shown empirically for a number of developing countries. Can we then say that the savings rate, like beauty, lies in the eyes of the beholder?

(The author has been associated with the RBI and the World Bank)