Improved estimates of household saving are both possible and necessary.
For the last seventy-five years measuring and understanding the saving behaviour of households has been a central concern of economists.
In what we now call the developed countries (some of whom then were not much richer in the 1930s than the middle-income countries of Asia today), interest in saving was largely stimulated by the work of John Maynard Keynes and his attempt to understand the sources of the Great Depression. As he famously observed, in an economy closed to international trade and capital movements, overall economic activity adjusts to equate desired saving with desired investment.
This insight, together with the implication that government must intervene when desired investment falls, desired saving increases and credit markets fail (all of which are true of the United States today) lies at the heart of today's call for global fiscal stimulus.
Keynes' insight was the origin of what we now call macroeconomics, and stimulated an explosion in research in the post-war period into what drives saving behaviour at the individual and household level. Keynes' primitive consumption function assumed that personal saving rose as income rose. This reflected the common-sense observation that, at any moment in time, the rich save more than the poor. The corresponding implication seemed to be that as countries became richer, they would be condemned to stagnation owing to saving rates that would structurally exceed desired investment.
Reconciling this contradiction between cross-section observation and the behaviour of a growing economy was the source of a rich and nuanced debate in the post-war period, leading to the life-cycle theory of household saving. Despite this large body of work, undertaken over many decades, there is much that remains idiosyncratic and mysterious about differences in household saving behaviour across nations, leading to the need for country-level research into these issues. As the above discussion indicates, such analysis needs to undertake the difficult task of reconciling microeconomic insights with macroeconomic behaviour, and also to link deeper structural determinants with business cycles in the economy. As the present global crisis has demonstrated, large inter-country differences between saving and investment (and the gap between the two, which is the current account deficit) have been powerful drivers of the global imbalances that many, myself included, believe were responsible for the present financial crisis.
These reflections are prompted in part by the recent publication of the ‘Report of the High Level Committee (HLC) on Estimation of Saving and Investment’ chaired by Dr C. Rangarajan and commissioned by the Ministry of Statistics and Plan Implementation (MOSPI). As a former Governor of the Reserve Bank of India and the Chairman of the National Statistical Commission which reported in 2001, Dr Rangarajan is ideally suited to lead this effort; indeed the secretariat for the exercise was provided by the Reserve Bank of India.
More From This Section
As the transmittal letter notes, "the issues relating to the estimation of gross domestic savings, have assumed immense significance in the recent times, when the Indian economy has begun its journey on a high growth trajectory and structural changes are taking place at rapid pace with respect to savings behaviour of the economic agents". The report is fundamentally concerned with improving the reliability and robustness of the national accounts estimates of saving and investment in GDP (what are called the final demand components).
While the report addresses all the three major sectors (or "institutions") within which national accounts saving and investment estimates are currently made (household, private corporate sector and public sector), the heart of the report is really about how to improve the estimates of financial savings of the household sector which, in the words of the report, “continues to pose a challenge in view of the heterogeneity and residual character of this sector in the National Accounts Statistics”.
The household sector, as defined in the National Accounts Statistics (NAS) is by far the largest source of gross domestic savings (somewhere between 70% and 80% of total savings, or around 24% of GDP in recent years), and the most important net supplier of resources to the other two sectors. The household sector's saving is further divided into financial and physical savings. As it happens, over the bulk of this decade net household financial savings have more or less equaled the saving-investment gap for the sector, implying that terms all the net financial savings of the sector have been made available to the other two sectors.
For purposes of the NAS the household sector is defined very widely; apart from individuals, the sector encompasses "all non-Government, non-corporate enterprises like sole proprietorships and partnerships owned and/or controlled by individuals and non-profit institutions which furnish educational, health, cultural, recreational and other social and community services to households". At present, both financial and physical saving by this large and amorphous sector are estimated by what is called the "residual method": material and financial balances are adjusted for amounts that are allocable to the public sector and the private corporate sector, and the residual is applied to the household sector. This approach is contrasted with practice in the US and the UK, where direct household income-expenditure surveys are the principal basis for such estimates.
At this point I must declare my institutional interest. As noted at several places in the report, NCAER has been one of the few non-governmental institutions that has attempted income-expenditure household surveys of the kind that the HLC believes are essential to improve the NAS estimates of household financial saving. We have been doing so for almost forty years. Our purpose in doing so has not been to provide a basis for the national accounts estimates of household financial saving, but rather to collect a series of datasets rich enough in associated variables to permit systematic testing of various hypotheses. There is a considerable wealth of experience that has been gained through this process, and the HLC ought perhaps to have assessed that experience more carefully and critically than it seems to have done, particularly before accepting the view of the National Sample Survey Organisation (NSSO) that undertaking such surveys is instrinsically difficult. The report's flat judgement, that "there is no evidence of such privately-run surveys contributing to an improvement in the estimation procedure for savings and capital formation" betrays an unexpected, and in my view unjustified, narrowness of perspective in a document of this importance and range.
The author is Director-General, NCAER. Views expressed are personal.