Following a structured programme that carries low risk—less than, say, 1.5%—and captures at least the average gain, would be the best solution
It is well known that the rupee, at the best of times, is an odd bird. Even though the domestic USD/INR market is now reasonably liquid, albeit less so than about 10 years ago, the pattern of rupee movements is a far cry from the near-normal distribution that most liquid currencies exhibit.
The accompanying graphic shows the distribution of 12-month returns of the spot rupee since 2003. On the left-hand side of the accompanying graphic is the percentage of times (counted on the y-axis) that the rupee appreciated; the x-axis shows the degree of appreciation (in percentage terms); the right-hand side shows the periods of rupee depreciation.
The rupee depreciated 61% of the time—in a closer-to-normal scenario, it should depreciate only a little more (or less) than 50%. What is particularly odd, as the accompanying graphic shows, is that the rupee’s most likely performance was a 3-4% appreciation over 12 months, which event occurred more than 20% of the time over the past 17 years. In more liquid currencies, the most likely performance is within a range of -0.5% to 0.5%. In EUR/USD, for instance, the 12-month returns were in that range 56.5% of the time; in the case of the rupee, it was just 17%.
There could be several reasons for this departure from normalcy—first off, the NDF market has a strong influence on the price (which RBI is trying to address); secondly, and quite obviously, there are far too many constraints on market access (no convertibility); and, finally, of course, RBI is largely in, or threatening to be in, the market all of the time. These forces limit the markets ability to develop its “normal” tendencies.
The most important result of this departure from normalcy is that it becomes much more difficult to manage USD-INR risk as compared to, say, EUR-USD risk, even though EUR-USD volatility is higher. For this reason, it is critical in the key parameter in selecting a USD-INR hedge strategy—and one which is overlooked by the vast majority of companies—is to determine the amount of risk the strategy carries, and to assess whether that risk is acceptable.
It is clear from experience that risk (see graphic)—which is measured as the total area under the curve—is much higher for a short USD-INR position than a long one. And, despite the fact that the rupee fell more than today’s forward premium only 41% of the time, it is too dangerous for an importer to stay unhedged. The worst decline on a 12-month horizon was 29.5%, which, given the distribution of returns, results in a risk (VaR) of 21%. This means there is a 5% chance that the rupee will fall by more than 21% over 12 months. [The equivalent numbers for 6-month and 3-month exposures are 13.4% and 9.9%, respectively]. On the other hand, and acknowledging that this risk is blindingly high, the fact that the rupee appreciated between 3 and 4% more than 20% of the time over the past 17 years makes it hard to pass up using some sort of structured programme with a disciplined stop loss to try and capture some of this opportunity.
For a long USD/INR exposure, the opportunity is huge as we have seen. However, the risk is also quite high—the worst rupee 12-month appreciation was 13.8% (with a Value at Risk of 10.8%); for 6-month and 3-month exposures, the VaR is 6.0% and 4.8%, respectively, which are also quite expensive. Clearly, here, too, staying unhedged for 12 months is not advisable; in addition to the spot risk, you would lose the forward premium which would be otherwise available. Even a 50% hedge strategy carries a lot of risk (including the foregone premium on the open part). For a 12-month export, the accompanying graphic shows the risk and opportunity of different approaches.
While all of these (with the exception of the 5% range) can generate substantial value if the rupee does actually crash, it is, in my view, more prudent to judge these against the average return. Here, the 5% range actually produces the best risk to return ratio. On a pure risk minimisation approach, which I feel should be the starting point, both the ATM option and the 5% range win out; however, the option costs money upfront, which is generally—and more so today—a dampener. Here again, following a structured programme that carries low risk—less than, say, 1.5%—and captures at least the average gain, would be the best solution.
The author is CEO, Mecklai Financial
Views are personal