India rebounded strongly from its 1991 balance-of-payments crisis, aided by structural reforms and other policy adjustments. The government has sought to reinvigorate the process of structural and fiscal reform. The paper examines trends in interstate differences in rural poverty; reviews India's postal saving system and possible reform issues; describes and evaluates the current system of pensions and provident funds, and discusses reform options. The paper also briefly reviews the structure of and recent developments in the Indian foreign exchange market.


India rebounded strongly from its 1991 balance-of-payments crisis, aided by structural reforms and other policy adjustments. The government has sought to reinvigorate the process of structural and fiscal reform. The paper examines trends in interstate differences in rural poverty; reviews India's postal saving system and possible reform issues; describes and evaluates the current system of pensions and provident funds, and discusses reform options. The paper also briefly reviews the structure of and recent developments in the Indian foreign exchange market.

X. Foreign Exchange Market Developments and Policies1

1. The exchange rate for the Indian rupee was unified and officially floated in 1993, and in the IMF’s classification of exchange arrangements the rupee is currently classified as a managed float. The Reserve Bank of India plays an active role in the foreign exchange market, and analysts have observed that the rupee/dollar rate exhibits extended periods of stability, before undergoing discrete adjustments.2

2. This chapter briefly reviews the structure and recent developments in the Indian foreign exchange market. Particular emphasis is laid on how the rupee’s volatility compares with that of other floating currencies. In addition, the dynamics of the offshore market and their implications for the adjustment of the onshore spot rate are examined, and the responsiveness of the spot rate to intervention, are also analyzed.

A. Overview of the Indian Foreign Exchange Market

3. Following the 1991 balance of payments crisis, India shifted from a fixed to a flexible exchange rate regime. This process began with the transition from a basket-linked managed float to a dual exchange rate system in March 1992, and culminated in the adoption of a unified and flexible exchange rate system in March 1993. This shift was accompanied by an easing of restrictions on current transactions and the achievement of full current account convertibility in August 1994.

4. Since the shift to a flexible exchange rate system, the rupee/dollar rate has exhibited periods of stability, broken by discrete, sharp movements. For example, the rupee underwent step adjustments against the dollar in the period leading up to its float, but was virtually unchanged at Rs 31.4/$ until August 1995. At that point, concern regarding competitiveness led to foreign exchange market pressures, and the rupee depreciated by about 12 percent against the dollar from this point to end-1995. The exchange rate remained in a narrow range until spillovers from the Asia crisis caused the rupee to depreciate by about 15 percent between September 1997 to July 1998. More modest movements occurred around August 1999, related to concern that the elections would disrupt the reform momentum, and again in May-August 2000, associated with fears that higher oil prices would undermine the balance of payments.

5. The movement in India’s nominal effective exchange rate has closely mirrored the rupee/dollar rate, and the exchange rate has been relatively stable in real effective terms (Chart X.1). In particular, the rupee/$ and nominal effective exchange rate have been almost perfectly correlated (i.e., with a correlation coefficient of 0.96), reflecting the high weight on the dollar and currencies closely linked to the dollar, and the nominal effective exchange rate depreciated by around 20 percent between March 1993 and July 2000. However, the real effective rate has remained stable and fluctuated within a relatively narrow band around its 1993 value.3

Chart X.1.

India: Exchange Rates and Reserves, 1990-2001

Sources: Data provided by the Indian authorities; IMF, Information Notice System; and WEFA.1/ Index based on WPI for India and CPI data for partner countries.

6. The Reserve Bank of India (RBI) is active in the foreign exchange market. In its public statements, the RBI has often noted that the thinness of the foreign exchange market, as well as leads and lags in large transactions (including those related to debt service and oil payments) can result in excess volatility. Thus, on a day-to-day basis, the RBI will intervene with a view to evening out supply and demand and to ensure that markets remain orderly.

7. The RBI’s shorter-run intervention also can be placed in the context of its view regarding the consistency of the real effective exchange rate (REER) with longer-term fundamentals. Although the authorities have not defined a specific target for the REER, the RBI has stated that it “would not hesitate to take actions to quell persistent volatility or misalignment. The broad objective of the exchange rate policy will be to ensure a reasonably stable real effective exchange rate.”4

8. The RBI utilizes a range of instruments to influence conditions in the foreign exchange market. Besides policies to affect domestic money market and credit conditions, as well as intervention in the spot market, the RBI undertakes both forward and swap transactions in support of its exchange rate objectives. These swap transactions, for example, involve the simultaneous loan of rupees and borrowing of foreign exchange, which would be unwound at some preset period in the future.

9. Administrative measures to influence exchange market conditions also are actively used. During the period of the Asian crisis, measures taken to reduce downward pressure on the rupee included restricting the scope for market participants to rebook forward transactions and the imposition of a 15 percentage point interest rate surcharge on import finance in December 1997, and a further increase in the surcharge and a halving of the export refinance limits in January 1998. As the rupee came under pressure in mid-2000, the RBI again resorted to a variety of administrative measures including imposing temporary interest surcharges on import finance in May, and tightening repatriation requirements on exporters in July.

B. The Structure of the Indian Foreign Exchange Market

10. Foreign exchange transactions in India remain highly regulated. All transactions are required to be effected through authorized dealers (ADs)—commercial banks specifically authorized by the RBI to engage in foreign exchange market transactions. The market is dominated by the State Bank of India, which is majority owned by the RBI and is responsible for conducting foreign exchange transactions related to oil imports and debt service on behalf of the government and government-owned enterprises. Transactions in the interbank market and between ADs and their customers may be for foreign exchange on a spot or forward basis.

11. Significant steps toward liberalizing current account transactions were taken with the unification of the exchange rate. The foreign exchange budget was eliminated and exchange controls on trade-related transactions were abolished in 1993, and most other restrictions on current account transactions were lifted in 1994, culminating in India’s formal acceptance of the IMF’s Article VIII. Exchange regulations were further liberalized with the Foreign Exchange Management Act of 2000, but foreign exchange regulations are still significant. For example, Indian residents are still required to surrender foreign exchange to authorized dealers within pre-specified time limits (usually 90 days).

12. There has been less progress toward easing restrictions on capital account transactions. In early 1997, a high-level committee issued a report—the Tarapore Report—calling for the establishment of capital account convertibility within a three-year period. However, with the onset of the Asia crisis, as well as difficulty in achieving some of the macroeconomic and structural preconditions that were highlighted by the Report as necessary for successful liberalization, progress in this area has not been rapid. Although regulations have been eased on longer-term capital inflows—including foreign ownership limits on foreign direct investment and limits on borrowing abroad by domestic corporations—and portfolio inflows by foreign institutional investors are relatively uninhibited, restrictions on short-term capital flows are significant.


Daily Foreign Exchange Turnover in 1998

Citation: IMF Staff Country Reports 2001, 181; 10.5089/9781451818550.002.A010

13. The rapid growth of external trade and an easing of restrictions on transactions have facilitated an increase in spot market volumes in recent years. Average monthly turnover increased to $109 billion in 1998/99, from $50 billion in 1993/94, but declined to $95 billion in 1999/2000, owing to restrictions on rebooting forward contracts (RBI, 2001). BIS estimates for April 1998, which exclude double counting by local dealers, put daily turnover at around $2½ billion (BIS, 1999). Comparing the BIS figures against a group of East Asian countries, India’s foreign exchange turnover as a share of GDP appears strikingly low, although as a percentage of trade turnover, the ratio is quite large. These two ratios underline the fact that trade, as a proportion GDP, is still relatively small in India.

14. Although restrictions on participation in the forward market have eased in recent years they remain significant. For example, in 1998 the authorities began allowing foreign portfolio investors to fully participate in India’s forward market, but only for incremental investments. Foreign institutional investors were allowed to take forward cover in the domestic market for up to 15 percent of the market value of outstanding investment at end-March 1999 plus the increase in market value/inflows after end-March 1999—FIIs were allowed to exceed these limits on a case-by-case basis. As discussed above, forward contracts may only be written in the case of a matching current account transaction, and in order to restrict the use of the forward market as a speculative vehicle, participants are restricted from canceling and rebooting contracts.

15. The forward market has grown in importance in recent years, although it remains susceptible to occasional large mismatches and turbulence. Volatility has been compounded by the tendency of Indian corporates and importers to refrain from covering their exposure during periods of exchange rate stability and to rush to hedge when spot market pressures arise (RBI, 2001). Moreover, the requirement that forward contracts cannot be entered into without an underlying current account transaction means that the market remains shallow at shorter maturities (seven days or less) and the market for longer-term contracts (i.e., greater than one year) has only just begun to develop.5 However, as the RBI has recognized in recent reports, the lack of regular two-way movement of the exchange rate appears to reduce the incentive for participants to use the forward market in a consistent and orderly manner.

C. Gauging the Flexibility of the Rupee

16. The rupee has been remarkably stable against the U.S. dollar, especially compared to other floating exchange rates. The lack of movement in the rupee/dollar rate is illustrated in Chart X.2, which shows that the volatility of the rupee/dollar rate has been relatively modest and that the number of days when the rupee depreciated significantly exceed the number of days it appreciated.6 Moreover, since 1996, the standard deviation of the daily and monthly percentage changes in the rupee/dollar rate have been consistently lower than that of other floating currencies (Table X.1). In contrast to the movement toward increased exchange rate flexibility by other countries since the Asian crisis in 1997, the rupee appears to have displayed a trend toward greater stability. Indeed, the present volatility of the rupee resembles that of currencies that previously had very tight linkages to the dollar (e.g., the Korean won and the Thai baht before the Asian crisis).

Chart X.2.

Daily Exchange Rate Movements (log difference)

Table X.1.

India: Exchange Rate Volatility, 1996-2000 1/

article image

Standard deviation of daily/monthly movements (percentage change) against the U.S. dollar.

17. Consistent with greater exchange rate stability, India’s interest rate and reserves volatility was correspondingly greater (Tables X.2 and X.3). Despite relatively significant capital account restrictions, exchange rate stability seemingly required significantly greater volatility (on a monthly basis) in India’s interest rates than in other comparator countries.7 Although less clear cut, it also appears that monthly reserves volatility has been greater than for other countries.

Table X.2.

India: Interest Rate Volatility, 1996-2000 1/

article image

Standard deviation of monthly changes in interest rates.

18. Summary indices of exchange rate flexibility illustrate the relative stability of the rupee. Following Glick and Wihlborg (1997) and Bayoumi and Eichengreen (1998), an index of exchange market flexibility was constructed by dividing the standard deviation of exchange rate movements by a measure of exchange market pressure (Table X.4).8 Although cross-country comparisons are difficult given definitional differences with the data, the index shows a trend toward less flexibility for the rupee, consistent with the previous results and in contrast to the other countries in the comparator group.

Table X.3.

India: Reserves Volatility, 1996-2000 1/

article image

Standard deviation of percentage monthly changes in reserves.

Table X.4.

India: Exchange Rate Flexibility, 1996-2000 1/

article image

Calculated as SDEX/(SDEX+SDREV), where SDEX is the standard deviation of log differences of the exchange rate against the U.S. dollar, and SDREV is the standard deviation of the changes in the central bank’s reserves divided by the lagged stock of base money.

Source: International Financial Statistics, International Monetary Fund, May 2001.

Reclassified from independent float to managed float: December 2000.

Moved from crawling band to independent float: September 1999.

Moved from managed float to independent float: December 1997.

Moved from crawling band to independent float: April 2000.

Moved from fixed peg to independent float: July 1997.

19. Regression analysis confirms the rupee’s tendency to closely track the U.S. dollar (Table X.5). Following Frankel and Wei (1994) and McKinnon (2000), the Swiss franc was used as a numeraire for measuring exchange rate variation, and for a representative sample of exchange rates, the following regression equation is estimated using daily data: dlog(local currency/SF) = β1 + β2 dlog(USD/SF) + β3 dlog(JPY/SF) + β4 dlog(DEM/SF) + ε

Table X.5.

India: Exchange Rate Regressions—Co-movement with the U.S. Dollar 1/

article image

Regression Model: dlog(local currency/SF) = bl + b2 dlog(USD/SF) + b3 dlog(JPY/SF) + b4 dlog(DEM/SF), where SF - Swiss Franc, USD - US Dollar, JPY - Japanese Yen, DEM - German Mark.

Null hypothesis for Wald Coefficient test: coefficient estimate on the US dollar is equal to one.

denotes significance at 1 percent level.

where SF is the Swiss franc, USD is the U.S. dollar, JPY is the Japanese yen, DEM is the German mark, and dlog is log first difference operator.

20. The estimates show that the rupee’s movement can be almost entirely explained by the dollar.9 This is in contrast to most of the other floating exchange rates in the sample, and even some exchange rates that were closely linked to the dollar before the Asian crisis. This tendency for the rupee to move closely with the dollar against other currencies was particularly evident in 2000, when the dollar’s strength contributed to substantial depreciation of the currencies of most of India’s trading partners in Asia and Europe, while the rupee/dollar rate remained relatively stable.

D. Offshore Markets

21. An offshore, nondeliverable forward (NDF) market for the rupee also exists, chiefly in Singapore and Mauritius. These transactions involves forward transactions for rupees that settle solely in dollars, and are reportedly used mainly by high net-worth nonresidents with limited access to the onshore forward market to hedge their rupee exposure.10 The market also provides the opportunity for non-Indian firms with rupee receivables (e.g., from exporting products to India) to sell their rupees forward against the dollar. Foreign banks, which are not allowed to hedge their capital on-shore, also may access the NDF market through their parent operations to balance their rupee exposure.

22. Capital account restrictions and market segmentation means that the NDF and onshore markets are not well integrated. This “inefficiency” can be illustrated statistically. In particular, a simple regression was estimated relating the percent change in the spot rupee/dollar rate—pch(e)—to the lagged change in NDF premium—the difference between the NDF and onshore six-month forward rate as a percent of the current spot rate, (P). In the absence of segmentation, there should be no gap between the offshore and onshore rates and any gap that might result would not provide leading information for movements in the spot exchange rate. In other words, the coefficient on the lagged premium should be insignificantly different from zero.

23. However, the results suggest that the NDF premium provides significant leading information regarding movements the spot rate. In particular, the estimates suggest that the gap between the offshore and the onshore forward premium is significant in the equation explaining the movements in the spot exchange rate.11


The estimates indicate that pressure on the exchange rate may at times be felt first in the offshore market, with a drop in the NDF rate relative to the onshore forward rate. In subsequent days, the spot rate begins to depreciate, and the lag structure suggests an overshooting in response that is only partially offset in subsequent days.

E. Foreign Exchange Market Intervention

24. As noted above, the RBI actively intervenes in both the spot and forward foreign exchange markets. Although in its public statements, the RBI emphasizes that its intervention is geared toward avoiding excessive volatility, the stability of the dollar/rupee rate raises questions about the extent to which intervention has been successful in affecting the level of the exchange rate.

25. A simple regression was estimated to test this proposition. Following the approach used by Bhaumik and Mukhopadhyay (2000), an equation was estimated that related the log difference of the rupee/dollar exchange rate—dlog(e)—to the log difference of reserves in millions of dollars—dlog(R). If intervention were successful, the expectation would be that a decrease in reserves (i.e., intervention in support of the rupee) would have slowed the rate of depreciation, or that the sign on dlog(R) would be positive.

26. The results of this exercise, however, do not suggest that intervention had the expected effect on the exchange rate:12


The reserve variable was only significant at the second lag and the sign of its coefficient was negative, indicating that a decrease in reserves was associated with depreciation two weeks following. This result does not appear to be related to the frequency of the data as it mirrors the conclusions of Bhaumik and Mukhopadhyay, who used monthly data, and tests over various sub-periods revealed similar results.

27. One possible explanation for the anomalous sign is simultaneity. For example, the equation coefficients could reflect the effects of the RBI’s reaction function, in which intervention “generally coincided with conditions of excess demand in the market” (RBI, 2001, p. IV-14). In other words, intervention would normally take place in response to a depreciation of the rupee. In order to examine this possibility, and to explore the relationship between reserves and the exchange rate in more detail, a simple bivariate vector auto-correction model was estimated using the same variables.13

28. The results confirm the existence of a cointegrating relationship between the level of reserves and the exchange rate, but the negative correlation was retained even in this framework. As summarized below, the long-run correlation between the exchange rate and reserves remained negative, and simulations of the model’s response to a one standard-deviation shock confirmed again that the impact of intervention—defined as a change in reserves—has a small and largely negative effect on the exchange rate.


Response of Log(e) to One Standard Deviation Innovations

Citation: IMF Staff Country Reports 2001, 181; 10.5089/9781451818550.002.A010

Vector Error Correction Estimates 1/

article image

T-statistics are in parentheses; EC is the error from the error correction equation.

29. These results contrast somewhat with those described in earlier sections. In particular, in Section C. it was suggested that the relatively high degree of volatility of foreign exchange reserves was related to the stability of the rupee. However, the analysis in this section does not confirm a causal relationship, and does not indicate that the rupee’s stability has been achieved to any significant degree by intervention in the spot market. This could reflect either the tests’ lack of power, the possibility that intervention has only an effect on a day-to-day basis that is not reflected in weekly data, or that the exchange rate responds to a range of instruments at the authorities’ disposal, including its intervention in the forward market and the use of administrative restrictions.

F. Concluding Observations

30. The discussion above suggests that the rupee’s exchange rate exhibits a number of characteristics:

  • The real exchange rate has been relatively stable since 1993, without displaying a trend and consistent with the authorities’ desire to maintain competitiveness.

  • At the same time, compared to most other floating exchange rates, the rupee has displayed an unusual lack of volatility against the U.S. dollar since the rate was unified in 1993. Although the rupee/dollar rate has adjusted over time, this has occurred principally as discrete, level changes, and on a day-to-day basis, the rupee has been relatively stable.

  • The stability of the rupee has been associated with relatively greater variability in domestic interest rates and reserves. However, there does not appear to be a significant statistical relationship between the authorities’ intervention in the spot market and the exchange rate, at least at weekly or lower frequencies.

  • Evidence of market segmentation can be found by examining the relationship between the spot exchange rate and the premium between the offshore and onshore forward rates. The gap between the NDF and the onshore forward rate appears to provide leading information regarding movements in the spot rate.

31. The experience of the Asian crisis has illustrated the importance of exchange rate flexibility, and suggests the need for continued efforts to broaden and deepen the Indian foreign exchange market. Considerable progress has already been made in liberalizing the foreign exchange market since the unification of the exchange rate in 1993. However, the pressure for further liberalization will grow as restrictions on capital account transactions are progressively eased and derivative and other instruments become increasingly available to financial market participants. Against this background, it will be important to ensure that the foreign exchange market develops in a manner that allows it to absorb shocks and day-to-day volatility in the exchange rate without requiring administrative measures and frequent official intervention to maintain orderly conditions.


  • Bayoumi, Tamim and Barry Eichengreen,Exchange Rate Volatility and Intervention: Implications of the Theory of Optimum Currency Areas,Journal of International Economics, 45 (1998), p. 191209.

    • Search Google Scholar
    • Export Citation
  • Bank of International Settlements, Triennial Central Bank Survey of Foreign Exchange and Derivatives Market Activity, 1998, May 1999.

  • Bhaumik, Sumon Kumar, and Hiranya Mukhopadhay,RBI’s Intervention in Foreign Exchange Market: An Econometric Analysis,Economic and Political Weekly, January 29, 2000, 373376.

    • Search Google Scholar
    • Export Citation
  • Frankel, Jeffrey and Shang-Jin Wei,Yen Bloc or Dollar Bloc? Exchange Rate Policies of the East Asian Economies,” in Takatoshi Ito and Anne Krueger (editors), Macroeconomic Linkage: Savings, Exchange Rates, and Capital Flows, Chicago: University of Chicago Press, 1994.

    • Search Google Scholar
    • Export Citation
  • Glick, Reuven, and Clas Wihlborg,Exchange Rate Regimes and International Trade,in Peter Kenen and Benjamin Cohen (editors), International Trade and Finance: New Frontiers for Research, Cambridge: Cambridge University Press, 1997.

    • Search Google Scholar
    • Export Citation
  • McKinnon, Ronald,After the Crisis, the East Asian Dollar Standard Revisited,Mimeograph, Stanford University, 2000.

  • Muhleisen, Martin,Foreign Exchange Markets: Developments and Issues,” in India Selected Issues, IMF Staff Country Report No. 98/112, October 1998.

    • Search Google Scholar
    • Export Citation
  • Reserve Bank of India, Report on Currency and Finance, 1999-2000, January 2001.


Prepared by Taimur Baig, Alexander Hammer, and Christopher Towe.


For example see “India’s exchange regime rate: need for a rethink,” Perspectives, ICICI, September 9, 2000.


Note that REER calculations for India typically use the WPI as the domestic price index, in view of its timeliness and quality compared to the CPI.


See the RBI’s Annual Report: 1995-96, September 1996, p. 86. Similarly, the 1999-2000 Annual Report states the RBI’s policy objective as “ensuring that the external value of the rupee is realistic and credible as evidenced by a sustainable CAD [current account deficit] and reserve position” (p. 115). The level of the REER at the time of the float of the rupee is often used as a benchmark for gauging the appropriate level of the exchange rate—for example, see the Government’s 1995-1996 Economic Survey.


As reported by BIS (1999), only 22 percent of India’s total forward contracts have a maturity of seven days or less, whereas the world average for the same was 51 percent.


Between 1994 and 2000, the number of days the rupee depreciated was 63 percent more than the days it appreciated. In over 90 percent of the days, the currency moved by less than half a percentage point, but in those cases when the daily movements were greater than 0.5 percent, the ratio was closer to 50 percent.


Note that fluctuations in reserves may reflect valuation adjustments, debt repayments, and other factors that do not necessarily represent foreign exchange market intervention. Moreover, forward market intervention, which is common in some of the countries in the sample, is not fully captured by the gross reserves figures.


The index is calculated using the following formula: Flexibility Index = SDEX / (SDEX+SDREV), where SDEX = standard deviation of exchange rate changes (log difference), and SDREV = standard deviation of the ratio of changes in reserves over lagged stock of base money.


The coefficient on the dollar is statistically indistinguishable from one. Broadly similar results were obtained with estimates that dropped the yen and mark variables from the regressions.


NDF markets have evolved mainly to allow financial market participants in countries with underdeveloped or highly restricted financial markets to hedge their foreign exchange exposure. The NDF market differs from the onshore market in that settlement does not involve the actual exchange of currencies—upon the expiration of the contract, the parties involved simply exchange the dollar equivalent of the difference between the contracted forward rupee/dollar rate and the then-prevailing spot rate.


Absolute value of t-statistics are in parentheses; LM(1) is the Breusch-Godfrey test for first-order serial correlation; and the constant term was not significant and was dropped from the regression. The sample period was June 2, 1999 to August 7, 2000.


The regression used weekly data from April 4, 1995 to January 26, 2001; t-statistics are in parentheses; the ARCH statistic tests for serial correlation.


Augmented Dickey Fuller tests did not reject the hypothesis that the data were non-stationary in levels, and the Johansen test could not reject the hypothesis of cointegration.

India: Recent Economic Developments and Selected Issues
Author: International Monetary Fund