Tuesday, May 01, 2007

Who's afraid of a strong rupee

RBI has been doing less trading on the currency market. Without these artificial forces in play, the INR has appreciated. Shankar Acharya wrote an eloquent article in Business Standard saying this is all wrong, that we should just go back to the happy days of the policy framework of the mid-1990s. He says that the INR is overvalued and that does fundamental damage to India's exports growth and ultimately India's GDP growth.

I disagree. My main point is that there is a lot going on in exports growth, and that the negative impact of INR appreciation and high local inflation has been swamped by other factors which have helped enable high export growth. The empirical evidence supports the claim that there is no simple and sharp negative impact flowing from INR appreciation to exports growth.

TCA Srinivasa Raghavan helped me to sharpen the argument. Let's start with an AR model for monthly yoy exports growth data, which just captures the time-series structure of the series. I use data from March 1992 onwards. This works out to an AR(14) model. With this in hand, suppose you introduce contemporaneous and past yoy changes of the INR/USD. There's absolutely nothing there. Going up to 6 lags (or beyond) gives nothing. Not a single coefficient is significant. This result holds whether you do the full dataset (from March 1992 onwards) or if you focus on the latest 100 points which is 8.33 years.

In other words, after controlling for the time-series structure of the exports growth series, yoy INR/USD fluctuations don't seem to matter in explaining the fluctuations of yoy exports growth.

Once again, my claim is not that prices don't matter. It's just that the explanatory power of currency fluctuations is small at best, and that the main story of exports growth is elsewhere. To say this differently, standard measures of REER are highly ineffective.


  1. "India exported goods worth $112.4 billion during 2005/06 while export of services was $71.6 billion. Services exports are estimated at $91.5 billion this fiscal, and merchandise exports at $132.7 billion." (Reuters)

    I agree with your article when you say many factors impact export growth, and that even an appreciating rupee has a mixed impact on overall export competitiveness. However, when considering just export of services, which in India's case is becoming sizable, it seems that an appreciating rupee will have amostly negative impact on export growth.


  2. Sir,
    Interesting post.
    Interesting article.
    I disagree with what Shankar Acharya wrote in his article a couple of days back.
    Just a couple of points:
    1. How do you think the appreciation affects the competitiveness of different sector?
    2. Policies to ameliorate the adversely affected sectors

  3. Sir,

    > INR, while has appreciated by 18%, on a point-to-point basis, it has mostly remained within a more narrow band of Rs.44-47.6 to the USD, during 2002-07. That brings the INR apprcn to less than 10% over a five year period.

    > Similarly, on the exports front, the big chunk of exports growth has come about due from petroleum products, commodities (such as iron ore),etc. Infact, share of petroleum products has shot up to over 12 per cent, from less than half of this till a few years ago.

    Which also explains, why unit realisations have jumped by around 40% during the period referred to in your article. Will this mean, that a large part of the inflation was being passed on?

    India does not dominate world trade in any of the major items in our merchandise exports basket (barring gems & jewellery). We have simply been price takers, given our lower competitiveness or quality.

    A sharply appreciating rupee can only worsen things for our exports??

    Also, barring an odd car-maker or a POSCO, there haven't been any large foreign investments in India, with an intention to make it an export hub. Most of the investments have been to cash in on the consumer boom within India.

    Ravi Purohit.

  4. Sir,

    What I meant in the earlier comment was: "Is it fair to conclude that INR appreciation has no impact upon exports based on the 2002-07 period?". Since this is the only period during which the INR witnessed a serious appreciation and further was also a period which coincided with a major run-up in the commodity cycle across-the-globe. Prior to 2002, the INR was mostly moving unidirectionally, depreciating against the USD.

  5. sir,

    You more than anyone should be cautious of applying the results of the regression outside the narrow band.

    Yes, INR-USD fluctuations may not matter for export growth while the rupee is within 44 - 47 range. But that does not mean, INR-USD will not matter for export growth when the Re is at 40. Your arguments seems to be analogous to saying that since inflation doesn't hurt growth when it is at 3%, it won't hurt growth when it is at 8%.

    This is especially true in India's case, since most of our major exports (IT, textiles, commodities) do not have significant import components.

  6. It is not asif there is no variation in the data in the historical period. E.g. in March 2004 there was a smart INR appreciation.

    I see no a priori reason to assume there will be threshold effects. Should we assume that whatever fluctuations have been seen in the post-1992 period are not sound evidence?

    If there were even weak coefficients for the post-1992 period or the post-1997 period, then one could argue about the situation. What is fascinating is that the coefficients are all zero. That's it. Not even one is significant.

    Now, I'm a good neoclassical; I'm not trying to argue that price effects are not important. Just that there's a lot else going on, and that the case for concern about INR appreciation is a lot weaker than meets the eye.

  7. sir,

    Your model includes nominal yoy changes of INR/USD. But since India's inflation rate (especially wage inflation) has been much higher than its trade partners, the real value of the rupee has appreciated much more. I wonder if the results would be different using trade weighted REER value of the rupee.

    As Arvind Virmani concluded in 2004, “the ratio of machinery to total investment and the rate of change in the real exchange rate have significant effect on TFPG.” Moreover, “an appreciation of the exchange rate is correlated with lower TFPG.”

    Slowing productivity growth (exacerbated by infrastructure constraints) seems to be a pretty good reason to believe that export growth (and GDP growth) will slow down.


Please note: Comments are moderated; I will delete comments that misbehave. The rules are as follows. Only civilised conversation is permitted on this blog. Criticising me is perfectly okay; uncivilised language is not. I delete any comment which is spam, has personal attacks against anyone, or uses foul language.