The incredible increase in foreign exchange reserves in India in particular and Asia in general has received a lot of attention. While there has been a surge in research on this topic and intense debate, basic questions still remain tantalisingly difficult to answer clearly.
This article highlights the optimising framework typically used in analysing foreign exchange reserves and tries to compare the actual reserves to optimal reserves, analyses the present management of reserves and emphasises the need for a thorough detailed study to serve as a guide for the future.
Economists are tuned to thinking in terms of costs and benefits; approach the demand for reserves with the same framework. The benefit of holding reserves is, you can avoid the unpleasant situation of not having foreign exchange to pay for imports, interest and foreign loans.
The luxury of having reserves comes with a price. Usually reserves are invested in short-term instruments, so that they are readily available. While India borrows domestically at 5.10 per cent (10-year bond yield) our reserves managed by the Reserve Bank of India (RBI) earned a paltry 3.1 per cent as per the RBI annual report for 2002-2003.
With slightly modified assumptions on the statistical process driving reserves, economists have arrived at different formulas for optimal reserves.
The table is a graphical depiction of the ratio of actual reserves as a proportion of the optimal reserves (reserves based on the optimisation framework) as indicated by the three different formulae based on economic research.
These formulae have been derived based on optimising frameworks by economists Heller (1966), Hamada &Ueda (1977) and Frenkel and Javonovic(1981). The key variables are the variability of balance of payments measured by the variability in changes in reserves (s), propensity to import (m) and the opportunity cost variable
Disclaimer: These are personal views of the writer. They do not necessarily reflect the opinion of www.business-standard.com or the Business Standard newspaper