HT Parekh finance forum
Indian Currency Regimeand Its Consequences
This article explains India’s currency regime, highlights the implications of a pegged currency and then goes on to discuss the time variation in the flexibility of currency in the Indian context.
ILA PATNAIK
S
A distinction between the de jure currency regime – as claimed by a central bank – and the de facto currency regime
– that is actually in operation – has been established in the literature. Many countries that claim to float have a “fear of floating” [Calvo and Reinhart 2002]. This suggests an investigation into the Indian rupee.
What Is the Indian Currency Regime?
A currency regime is classified as a de facto peg when the volatility of the exchange rate against one currency is very low, owing to trading by the central bank. Reinhart and Rogoff (2003) identify the Indian currency regime over 1979-2001 as a “peg to the US dollar”. Calvo and Reinhart (2002) propose a metric of currency flexibility that combines volatility of the exchange rate, volatility of reserves and interest rate volatility. The intuition of their measure is that when the central bank tries to set the exchange rate, this will require greater reserves volatility and greater interest rate volatility. Their calculations suggest that currency flexibility in India has not changed in the 1979-1999 period. This evidence is extended till 2003 in Patnaik (2003) with the same result. These results suggest that the de facto currency regime in India did not change after 1979.
Table 1 for cross-currency volatilities, adapted from Patnaik (2003), highlights the phenomenon of pegging. The INR/USD has the lowest volatility of a range of cross-currency rates.1 In the extreme case, if the INR/USD were a fixed rate, INR/JPY volatility would be exactly the same as USD/JPY volatility, since every change in the USD/JPY rate would yield an identical change in the INR/JPY rate. The table shows INR/JPY volatility – of
0.738 per cent per day – is close to the USD-JPY volatility of 0.697 per cent. Indeed, INR exchange rates other than the INR/USD have volatilities like the other floating rate currency pairs in the table. These are characteristics of a de facto INR/USD pegged exchange rate. The INR is pegged to the USD, and thereby floats with respect to all other currencies.
One of the best tools for inferring the currency regime in operation based on currency market data is based on a regression.2 An independent currency, such as the Swiss franc (CHF), is chosen as an arbitrary “numeraire”. Exchange rate timeseries are re-expressed as daily returns, i e, log price differences, and the following regression is estimated:
r INR = β1+ β2r USD + β3r JPY CHF CHF CHF
+ β4r EUR + β5r GBP + ε CHF CHF
As an example, r USD is the one-day
CHF return on the USD-CHF exchange rate. Interpretation of the estimates falls into the following categories: Fixed: If India runs a fixed exchange rate to the USD, then every fluctuation in the INR/CHF rate merely reflects a fluctuation in the USD/CHF rate. The regression yields β2 = 1, β3 = β4 = β5 = σε = 0 and R2 = 1. Floating: A floating rate is characterised by high values of σε and low R2. A floating rate does not mean that β = 0. If (say) Japan is important in trade and financial transactions, the JPY will be significant. The coefficients reflect the true trade and financial linkages present in the economy. Pegged: Under a pegged INR/USD rate, we see β2 = 1, and near-zero values for β3, β4, β5. The size of σε, and the regression R2, convey how tight the peg is. Low values of σε, and hence high values of the R2, are associated with a less flexible exchange rate. If a basket peg is in operation, the regression will show the weights of the USD, JPY and EUR in the basket.
Through this, evidence can be obtained about the de facto currency regime in the period of interest. It also offers a mechanism for monitoring the currency regime, so as to know the changes in the tightness of a peg and the identity of the currency or basket of currencies against which pegging is done.
Table 2 applies this model to a few interesting situations. These serve as examples of the regression approach to characterising currency regimes. In addition, they generate international comparisons which help give us a cross-country perspective for thinking about the INR currency regime.
The first example is that of the Chinese yuan in the fixed exchange-rate period. The USD coefficient is 1, with an enormous t-statistic of 2665.7. All other coefficients are 0. The residual standard deviation is just 0.007, and the R2 is 99.99 per cent. This shows us what the regression estimates look like when faced with a fixed exchange rate. China announced that from July 22, 2005 onwards, the Chinese yuan would no longer be a fixed exchange rate but would instead be pegged to a basket of currencies. While this may be the stated de jure situation, de facto
The H T Parekh Finance Forum is coordinated by Errol D’Souza, Shubhashis Gangopadhyay, Subir Gokarn, Ajay Shah and Praveen Mohanty.
Economic and Political Weekly March 17, 2007
there is a USD peg. As the table shows Figure: Time-Varying Volatility of INR/USD Exchange Rate
(one-year moving window)
the coefficient of the USD barely moved
– from 1 to 0.9975, the coefficents of all
other currencies remain insignificant, and the R2 is still at 0.9894. There is no
1.0 –
evidence in favour of a basket peg; it is
0.5 –
a simple USD peg.
The pre-crisis Korean won was tightly pegged to the USD. After the crisis, the t-statistic of the USD dropped from 286.0 to 32.1, reflecting an important reform in
the currency regime. The Japanese yen has
1994 1996 1998 2000 2002 2004 2006
Vol of weekly returns in 52-wk window
–
–
–
–
–
–
–
0.2 –
0.1 –
become very important in fluctuations of the won under the new regime. This reflects trade and financial linkages between Korea and Japan. The σε rose from 0.123 to 0.4317 and the R2 dropped from 97.57 per cent to 67.89 per cent, but the won is not yet a floating rate.
The Indian rupee appears to be pegged to the USD. At the same time, the JPY and the EUR do seem to affect the rupee. The R2 of the regression is 84.11 per cent, thus leaving little space for any other factors in the determination of the nominal exchange rate.
Finally, the table has New Zealand, Brazil and South Africa, over a common sevenyear period. The symptoms of floating seen in the regression are: statistical significance of multiple currencies, a high σε and a low R2. The coefficients seen on the
Table 1: Cross-Currency Volatility
(Daily returns, 4/1993 – 1/2007)
| USD | GBP | EUR | JPY | |
|---|---|---|---|---|
| INR USD GBP EUR | 0.291 | 0.575 0.508 | 0.663 0.606 0.465 | 0.738 0.697 0.721 0.705 |
Source: Adapted from Patnaik (2003).
regression, with a floating exchange rate, reflect the country composition of trade and finance – as an example, the USD has an important coefficient for the Brazilian real. For the South African rand, which is a floating rate, we see that the coefficients of all currencies are significant and the R2 is extremely low at 0.12. The coefficient of
0.58 for the euro shows the importance of Europe in South Africa’s trade and finance.
Implications of Pegged Currency
A key insight of open economy macroeconomics, which has come to prominence in recent decades, is the idea of the “impossible trinity” [Mundell 1961]. This consists of the assertion that no country can simultaneously have an open capital account, a fixed exchange rate, and monetary policy independence. Specifically, once the capital account is open, and the exchange rate is fixed, monetary policy is solely driven by the need to uphold the fixed exchange rate.
As an example, suppose a central bank embarks on tight monetary policy when
Table 2: Currency Regime Seen Through Rregression Estimates
| Country and Period | USD | JPY | EUR | GBP | σε | R2 |
|---|---|---|---|---|---|---|
| Evolution of the Chinese currency regime | ||||||
| September 28, 1998 | 1.000 | 0.0001 | 0.0004 | 0.0001 | 0.007 | 0.9999 |
| to July 21, 2005 | (2665.7) | (0.4) | (0.4) | (0.3) | ||
| July 22, 2005 | 0.9975 | 0.0076 | -0.0144 | -0.0080 | 0.0590 | 0.9894 |
| to December 31, 2006 | (132.0) | (1.0) | (-0.6) | (-0.7) | ||
| Pre-crisis Korea | ||||||
| January 1, 1981 | 0.9981 | 0.0004 | -0.0127 | 0.0081 | 0.1232 | 0.9757 |
| to January 1, 1996 | (286.0) | (0.1) | (-1.5) | (1.7) | ||
| Post-reforms India | ||||||
| April 1, 1993 | 0.9560 | 0.0286 | 0.0522 | 0.0213 | 0.2904 | 0.8411 |
| to December 31, 2006 | (88.2) | (3.6) | (2.2) | (1.6) | ||
| Post-reforms Korea | ||||||
| January 1, 2000 | 0.7323 | 0.2530 | -0.0295 | 0.0771 | 0.4317 | 0.6789 |
| to December 31, 2006 | (32.1) | (13.1) | (-0.6) | (2.7) | ||
| Floating rates, all January 1, 2000 – December 31, 2006 | ||||||
| Brazilian real | 0.8254 | 0.1063 | 0.4504 | 0.1126 | 0.9808 | 0.3208 |
| (15.9) | (2.4) | (4.0) | (1.7) | |||
| New Zealand dollar | 0.2076 | 0.1471 | 0.5294 | 0.2639 | 0.6345 | 0.2516 |
| (6.2) | (5.2) | (7.2) | (6.2) | |||
| South African rand | 0.2060 | 0.1714 | 0.5814 | 0.2740 | 1.0240 | 0.1268 |
| (3.8) | (3.7) | (4.9) | (4.0) |
there is a fixed exchange rate and no capital controls. Tight monetary policy gives higher interest rates, which attract capital inflows. The central bank has to buy foreign currency in order to prevent appreciation. This gives higher money supply, which frustrates the attempt at having tight monetary policy. Few countries today have fixed exchange rates. However, trying to “manage” an exchange rate yields similar conflicts. When an exchange rate target is sufficiently important, the attempt to force a desired price on the currency market can lead to a loss of monetary policy.
As Fischer (2006) points out, during the 1990s many economies were struggling to control or stabilise inflation. A pegged exchange rate was a tool for controlling inflation. The process of liberalisation of trade and capital flows then generated an impossible combination, and a series of financial crises erupted around the world.
In the Indian case, a system of strong capital controls backed by the Foreign Exchange Regulation Act (FERA) was present at the outset. This made it possible to have a pegged exchange rate and monetary policy autonomy over the 1973-93 period. In the decade of the 1990s, restrictions on the current and capital accounts were substantially, though not completely, eased. India does not have a completely open capital account, nor does it have a fixed exchange rate. Patnaik (2005) shows that there was a loss of monetary policy autonomy in the 1990s.
As an example, in 1998, when local business cycle conditions were weak, the pegged exchange rate led to a tight monetary policy. Conversely, in the period after 2002, when local business cycle conditions were buoyant, the pegged exchange rate led to monetary expansion and accelerating inflation. The currency regime is now a central issue in understanding monetary policy and capital flows [Shah and Patnaik forthcoming].
Economic and Political Weekly March 17, 2007
While the INR currency regime has overall been de facto pegged to the USD, the extent of pegging has varied significantly through this period.
The Figure shows a “moving window” estimate of INR/USD volatility. The scale of this graph runs from values like 0.1 per cent per week to values like 1 per cent per week – a range of 10:1. There have been multi-month periods where the INR/ USD exchange rate was fixed but there have also been periods where the volatility of the INR/USD was closer to that of the INR/EUR or the INR/JPY.
These changes in currency flexibility have not been preceded by announcements from the RBI. Currency flexibility has risen and dropped without this information being transparently shared with economic agents. This information matters directly to economic agents with currency exposure, who need forecasts of currency volatility for making decisions on currency hedging. In addition, complexities of implementation of the currency regime are now of central importance in understanding monetary policy and capital flows. Given this lack of transparency at the RBI, monitoring and understanding Indian macroeconomics requires an ongoing process where the ideas and methods of this paper are applied in inferring the de facto currency regime from the data.
EPWEmail: ipatnaik@nipfp.org.in
Notes
1 The series was extended backwards in the period prior to the introduction of the euro using thedeutsche mark.
2 This approach was made prominent by Frankel and Wei (1994) but has been invented several times before, going back at least to Haldane and Hall (1991). Also see Benassy-Quere and Coeure (2003).
References
Benassy-Quere, A and B Coeure (2003): ‘On the Identification of De Facto Currency Pegs’, Technical Report, University of Paris; French Ministry of Economy and Finance.
Calvo, G A and C M Reinhart (2002): ‘Fear of Floating’, Quarterly Journal of Economics, Vol 117, No 2, pp 379-408.
Fischer, S (2006): ‘Dollarisation: Consequences and Policy Options’, Technical Report, keynote address, Central Bank of the Republic of Turkey, URL http://www.bis.org/review/r061221e.pdf.
Frankel, J and S J Wei (1994): ‘Yen Bloc or Dollar Bloc? Exchange Rate Policies of the East Asian Countries’ in T Ito, A Krueger (eds), Macroeconomic Linkage: Savings, Exchange Rates and Capital Flows, University of Chicago Press.
Haldane, A G and S G Hall (1991): ‘Sterling’s Relationship with the Dollar and the Deutschemark: 1976-89’, The Economic Journal, Vol 101, No 406, pp 436-43.
Mundell, R (1961): ‘The International Disequilibrium System’, Kyklos, Vol 14, pp 154-72.
Patnaik, I (2003): ‘India’s Policy Stance on Reserves and the Currency’, Technical Report, ICRIER Working Paper No 108, URL http://www.icrier.org/pdf/wp108.pdf.
– (2005): ‘India’s Experience with a Pegged Exchange Rate’ in S Bery, B Bosworth, A Panagariya (eds), The India Policy Forum 2004, Brookings Institution Press and NCAER, pp 189-226, URL http://openlib.org/home/ila/ PDFDOCS/Patnaik2004_implementati on.pdf.
Reinhart, C M and K S Rogoff (2003): ‘The Modern History of Exchange Rate Arrangements: A Reinterpretation’, Technical Report 8963, NBER.
Shah, A and I Patnaik (forthcoming): ‘India’s Experience with Capital Flows: The Elusive Quest for a Sustainable Current Account Deficit’ in S Edwards (ed), Capital Controls and Capital Flows in Emerging Economies: Policies, Practices and Consequences, NBER and University of Chicago Press, URL http:// www.nber.org/papers/w11387.
CENTRE FOR GOOD GOVERNANCE
Dr. MCR HRD IAP Campus, Road No. 25, Jubilee Hills, Hyderabad 500 033, ANDHRA PRADESH
Tel 040-2354 1952 email: recoup@cgg.gov.in
Call for Papers
Researchers Colloquium on Urban Poverty (RECoUP)
RECoUP, a colloquium of researchers set up by the Ministry of Housing and Urban Poverty Alleviation (MoHUPA), GoI in association with UNDP, India serves as a forum to develop the capacity of young researchers engaged in urban poverty and livelihoods to establish the most successful elements of policies, practices and innovative approaches in the following priority areas:
Centre for Good Governance (CGG), the technical secretariat of RECoUP invites original synopsis/ papers from researchers in the above themes.
For details on the colloquium and financial support visit: http://www.cgg.gov.in/RECOUP.html
Economic and Political Weekly March 17, 2007
