Friday, April 30, 2010

Is the REER model of any use?

Every day, in the context of Exchange Rate management in India, one comes across proponents of a model called REER. REER has been, and continues to be used to justify a weak Rupee without regard to the fact that it is an inherently flawed model which has been abandoned across the world. However, knowledge of these flaws doesn’t stop influential proponents like Dr. Shankar Acharya  (Indifference about the big rupee rise?, Business Standard, April 22, 2010) from screaming murder whenever the Rupee appreciates.

The Real Effective Exchange Rate model, or the REER model, allows relatively closed economies a means to value their currencies on a relative basis. Not only relative to other currencies, but also relative to the domestic currency’s own value at some point in the past. It works by adjusting the value of the currency by the inflation differential between the domestic economy and the reference economy, supposedly factoring in the incremental cost differences experienced by the domestic economy vis-a-vis the reference economy.

For example, let us assume the USD/INR rate at the beginning of 2009 was Rs. 40/USD. If in the year 2009, inflation in India was 9% and inflation in the US was 4%, the REER model would indicate that the value of the Rupee should reduce by 5% (9%-4%) to adjust for this inflation. Therefore, according to the model, at the end of 2009, the INR/USD rate should be Rs. 42/USD, a depreciation of 5% for the year.

In effect, the REER model is an incremental Purchasing Power Parity (PPP) model. It can only account for changes in the purchasing power of one currency when compared to another on the basis of inflation differentials in the two economies.

The two major flaws of this model are,

1) Assumption of “base rate” accuracy: The REER model can only indicate the required change in the value of the currency relative to another currency (or a group of currencies) over a set time period. It assumes that the “base rate”, or the earlier exchange rate used to calculate this change is accurate. In the above example, the rate of Rs. 42/USD at the end of 2009, as calculated by the model holds true only if one assumes that the rate of Rs. 40/USD at the beginning of 2009 was accurate. If not, REER only compounds the error but cannot rectify it.

In India’s case, the Base Year for REER model used by the RBI is 1993-94. So, in essence, all REER calculations made by the RBI, and the ones Dr. Shankar Acharya refers to, assume that the exchange rate in 1993-94 was an accurate representation of the purchasing power of the Indian Rupee at that time. Nothing could be further from the truth.

All of us who have vivid memories of the 1991 crisis remember that India was on the verge of a foreign currency obligation default which was staved by pledging our holdings of Gold. We also remember that the Rupee was devalued very sharply in the run up to this default and immediately after it. From a value of Rs 12.75/USD in January 1988 to the artificially pegged value of Rs. 31.37/USD in January 1993, the Indian Rupee lost close to 60% of its value in this period. The number doesn’t do justice to the magnitude of this movement because of the manner in which currencies are quoted. To put things in perspective, a move of similar magnitude would result in the Rupee quoting at Rs. 109.5/USD in 5 years time.

This movement was not caused by inflation differentials, but by an acute shortage of foreign exchange, a demand-supply gap. Now that this shortage has turned into a surplus, sometimes of unmanageable proportions, it is only fitting that this devaluation be reversed to restore the purchasing power of the Rupee to its pre 1988 levels. But by taking 1993-94 as the base, the RBI and it’s followers only legitimize the devaluation as one caused by inflation differentials. My suggestion to all proponents of REER would be to calculate it by taking 1987-88 as the Base Year. It would solve at least a part of the problem, but not all of it. The second flaw in the REER model would still be as much of an issue.

2) Circular Causality: The REER model suffers from a fatal flaw called circular causality. This enigmatic term is easier explained by using the chicken & egg example, but not in the traditional manner. In this case the existence of a chicken (if it is what came first) leads to eggs, which hatch into chickens resulting in more eggs. When REER uses inflation differential to cause an adjustment in currency value, this change in currency value results in further inflation differential, requiring a further adjustment in currency value. When domestic inflation is higher, the resulting devaluation of one’s currency causes the price of all imported goods to rise in the domestic market. In the case of a country like India, which imports almost all of its Oil, this price increase translates directly into inflation across the spectrum of goods available in the domestic market due to higher transportation costs, apart from the direct impact of higher fuel prices. Higher inflation then shows up as currency overvaluation in the model, requiring another devaluation and the cycle starts all over again.

While the first flaw of the model is surmountable, as mentioned above, the second can only be overcome by the use of extremely complicated adjustments which remove the impact of imported inflation from total inflation, and take only net domestic inflation as input for REER calculations. This has proved close to impossible so far.

Using REER as a reason for currency devaluation is devoid of all rationality. It can be a guidepost if implemented properly, but can never be the bedrock of a nation’s exchange rate policy. It is time that REER was given just the respect it deserves and nothing more.

After almost 2 decades of misplaced trust the Indian Rupee, and India, deserve better.

This article was featured on on April 26, 2010.

No comments:

Post a Comment