Is the persistent rise in the rupee's value against the dollar likely to take its toll on Indian exports soon?
It turns out that the answer lies in the choice of model used to analyse this phenomenon.
One set of models show that exports are quite independent of prices and exchange rates; exports are driven up or down by changes in global incomes. The other set of models reveal a completely different picture.
Indian exports pick up during downturns in global spending and slacken during upturns. This somewhat contrarian pattern is driven essentially by fluctuations in the exchange rate, adjusted by relative inflation rates.
The traditional approach to modelling export demand runs in terms of relating levels of exports to the real exchange rate, levels of world demand and lagged values of exports. Exports are measured either in dollars or are deflated by the exchange rate and the price index for exports (the unit value index).
The 'real exchange rate' refers to the rupee's exchange value either against a basket of currencies or simply against the US dollar, adjusted appropriately for relative inflation between India and these economies.
Thus, if the currency depreciates by 5 per cent but Indian inflation in the period exceeds US inflation by 5 per cent, the real value of the rupee versus the dollar remains static.
We fitted quarterly data to this model from 1993-94 to 2001-02 and arrived at some interesting results.
For one, exports seem fairly insensitive to the real exchange rate, measured against the dollar, a basket of five major currencies or a larger basket of 36 currencies.
It is, on the other hand, extremely sensitive both to world income (measured here as the sum of US and Euro-zone incomes) and also to the previous quarter's exports.
The sensitivity to world incomes illustrates the 'rising tide' effect: growth in world incomes leads to a surge in spending on traded goods. Indian goods benefit as a result.
The presence of lagged export term suggests persistence in the export cycle. An up-tick in a quarter is likely to spill over into the following quarter.
Why would Indian exports be insensitive to exchange rates?
Clearly for traditional categories like textiles and leather products, the price factor and hence the exchange rates remains important. However, there is an increasing share of items that are specialised and cannot be easily substituted on price factors alone.
These are categories like pharmaceuticals, engineering goods, speciality chemicals and the like. In 2001-02, these together constituted roughly 35 per cent of total exports.
Of the biggest category in overall exports (a 16 per cent share), gems and jewellery, the relationship with the exchange rate is somewhat ambiguous.
India adds value by cutting and polishing rough diamonds that it imports. Thus, while a depreciation lowers the effective cost of this service, it raises the price of the input.
An appreciation raises the cost of the service but lowers the import cost. The net impact would depend on which of these two effects dominate.
However, there are some technical problems with this model. For one, the levels of exports as well as world incomes increase uni-directionally over time, that is, there is a secular trend in these variables.
A simple model that links up these two variables could be reflecting this underlying trend in the two variables and not the causal link between the two.
One way of getting past this problem is to strip the data of this trend and look at 'cycles' instead. Cycles simply refer to the deviation in a particular period from the underlying trend. A positive deviation signifies a 'boom' and a negative deviation a 'bust.'
Both exports and world incomes have gone through marked cyclical phases that are fairly easy to capture. But here lies the puzzle.
A boom in Indian exports has, over the last decade, coincided with a bust in world incomes whereas busts in exports have come in periods where world incomes have boomed. So the 'rising tide' effect doesn't quite seemed to have worked.
What explains this aberration? A look at the data reveals that cycles in exports have been set off by sharp movements in real exchange rates, particularly versus the dollar.
A large adverse movement (appreciation) in the exchange has triggered a downturn while a favourable movement (depreciation) has set off a rebound.
Though a causal connection is as yet not clear, periods of large depreciation in the rupee in recent times have coincided with the onset of a bust in the world income cycle while appreciation has come at a time when the world incomes have perked up.
The bottomline is that cycles in exports seem to have been driven largely by exchange rate movements.
This has dominated the impact of changes in global spending which has moved in the opposite direction. This is consistent with the current boom in exports.
The 16 per cent growth in exports in 2002-03 came at a time when both US and European incomes where sagging. This export boom incidentally followed a period of sharp real depreciation of the rupee over 2001-02.
The two models seem to point to different policy prescriptions. The first calls for a somewhat passive approach toward exchange rates while the second calls for an active exchange rate strategy.
One way of reconciling the two is the following: over the short term, variations in export growth are insensitive to variations in exchange rate and depend instead on the purchasing power of buyers.
That's the message from the first model. The lesson from the second model is that over the medium term, periods of large overvaluation or undervaluation of the rupee can set off a turn in the export cycle, even if the export basket is relatively insensitive to small changes in the currency.
This is relevant in the current situation. The rupee has persistently appreciated against the dollar over the last few months. This is very likely to arrest the current boom phase and set off a downturn.
Even if there is a recovery in world incomes, the slackness in exports could continue. An active exchange rate management strategy may prove to be the best course of action for now.
The author is a senior economist at Crisil.
More from rediff