Time was when every increase in our foreign exchange reserves received unalloyed welcome. Not so, in recent times ?? all good things can turn blas?. The umbrage of people regarding our burgeoning forex reserves takes many forms ?? from the naive (seeing a parallel with our embarrassingly large foodgrain stocks) to the discriminatory (a perceived gross misallocation of capital resources). We?ll skip this particular argument, and focus on the facts of the case. How much have our forex reserves grown ?? in reality? And this is not about another accounting misadventure. It?s about the difficulties that arise when the unit of measure is itself changeable, about as fundamental a problem as one has in economics.
The Reserve Bank of India reports its forex reserves in US dollars. Ignoring the gold and Special Drawing Rights, the RBI invests its forex reserves in various types of marketable instruments ?? mostly gilts issued by foreign governments. In these days of super-currencies, there are only a few to choose from ?? US dollar ($), euro, sterling and the yen. Very wisely the RBI keeps to itself the foreign currency composition of these assets, and the maturity structure of these investments. But we can guess that given the currency composition of our foreign trade and other current account obligations, the majority of it must be in US dollars, just as we also know that a significant part of it must be in euro, sterling and yen.
Now, if the $ appreciates against the euro, the assets denominated in the latter will be worth less when measured in dollars; similarly so with yen and sterling assets. This is roughly what happened through much of 2000. You might recollect that in May-June 2000, RBI?s forex assets declined for weeks together and it was not till August that things settled down. To offset the negative media attention of this continuous decline in reserves the RBI, in its Weekly Statistical Supplement, put a big asterisk against the entry for changes in forex assets and began to print the explanatory footnote (saying that this included valuation changes) in bold lettering.
Since the beginning of 2002, it has been the turn of the US dollar to lose ground. The appreciation of the RBI?s euro, sterling and yen assets vis-a-vis the dollar are reflected in the increase in the reported value of its foreign currency assets. Now, we know that between 1 April and 28 June 2002, these assets rose by $3.65 billion, and by another $2.68 bn up to 16 August 2002 — a total increase of $6.33 bn in the first four and a half months of this financial year.
So, how much of it was due to valuation changes, and how much was due to real flow of financial resources? It is an impossible question to answer with any degree of precision, for we do not know the currency composition of the forex asset portfolio. But we can estimate the likely range of values for the real and valuation components of the increase, using different starting currency weights. Generally, lower the dollar proportion at the beginning of 2002-03, bigger would be the valuation impact, and lower the real increase.
The most interesting thing is that the outcome is very different in the first quarter (April-June), and in the six week period of July to mid-August. For Q1, we estimate that of the total increase in forex assets of $3.65 bn, as much as $1.5 to $2 bn was due to pure valuation changes. Thus, something between $1.6 and $2.1 bn constituted real increase in the reserve position ?? roughly half-and-half between valuation and real. Between 1 July and 16 August however, the greenback recovered some ground. As a result, the valuation component of the total increase of $2.68 bn in forex assets during this period was a negative $0.2 to $0.3 bn. Therefore, the real increase in reserves in this period was a whopping $3 bn ?? 50 per cent higher than the entire first quarter. What was the source of this big surge? We?ll have to leave that for later.
The author is economic advisor to ICRA (Investment Information and Credit Rating Agency)