Section III

Features of Indian Rupee NDF market

Trading platform and offshore centre

NDFs are primarily over-the-counter, rather than exchange-traded products, thus making it difficult to gauge the volume of contracts traded, who trades the contracts, and where they are traded. At the international level, New York tends to dominate trading in Latin American NDFs, Singapore (and to a lesser extent Hong Kong) dominate trading in non- Japan Asian NDFs, while London spans these markets. The INR NDF is largely concentrated in Singapore and Hong Kong, with small volumes being traded in the Middle East (Dubai and Bahrain) as well.

NDF Market Regulation

At present, there are no controls on the offshore participation in INR NDF markets. The onshore financial institutions in India, however, are not allowed to transact in the NDF markets. Domestic banking entities are allowed specific open position and gap limits for their foreign exchange exposures and through these limits domestic entities could play in the NDF markets to take advantage of any arbitrage or even speculate. This itself restricts the extent to which domestic banks could participate in NDF markets. The objective has been that allowing domestic banks to participate in the NDF markets would require an enhanced level of intervention from the Reserve Bank of India (RBI) to protect Indian rupee from any speculative attack.

Market Players

NDF market players generally operate with an objective of hedging, speculating and arbitraging. While the INR NDF market has been around for over the last 10 years or so, the characteristics of this market seem to have evolved over this period in tandem with the onshore exchange controls and regulations. In the late 1990s the NDF market was provided liquidity by foreign residents who had a genuine exposure to the Indian rupee but were unable to hedge their exposure in the domestic market due to existing controls. However, with the gradual relaxation of the exchange controls, reasonable hedging facilities are available to offshore non-residents who have exposures to the rupee, especially when compared with the hedging facilities provided by some other competitor Asian countries such as China. Hence, the INR NDF market presently derives its liquidity largely from (i) Non-residents wishing to speculate on the Indian rupee without any exposure to the country, and from (ii) arbitrageurs who try to exploit the differentials in the prices in the two markets without any outlay of capital on their part by two offsetting transactions. (For the Indian rupee it is believed that arbitrage is profitable when there is difference of around 10 paise in the forwards prices. Such opportunities are not very common, but tend to occur whenever speculative actions increase).

The behaviour of NDF market players depends critically on their objective for participation. Foreign investors who participate in the NDF market to hedge their exposures generally take long positions e.g. multinational companies. Speculators, on the other hand, operate mostly in the short end of the market e.g. hedge funds, also corporates entities with an international presence who undertake speculative or arbitrage trades, jewel exporters and manufacturers that constitute another group who are active arbitraging between domestic and NDF markets.

As reported by market participants, some of the foreign banks which trade in the rupee NDFs include Deutsche Bank, UBS AG, Standard Chartered Bank, Citibank, JP Morgan Chase, ABN Amro, Barclays, ANZ Investment bank and BNP.

Settlement period

The settlement period refers to the gap between the day the NDF contract is fixed and the actual delivery date. The fixing date is the day on which the comparison between the NDF rate and the prevailing spot rate is made. The settlement date is the day whereby the difference is paid or received. Depending on the currencies dealt, there are variations whereby for some currencies, the settlement period is one day whereas for others it is two business days. Generally, the spot rate used in the NDF market is based on a reference page on Reuters or Telerate with a fallback of calling four leading dealers in the relevant market for a quote. For the Indian rupee NDF, the RBI reference rate is generally used as the fixing rate.

Market Turnover

Information on the traded volumes in the NDF market is rather difficult. Various estimates are available though they are published with a lag. As per an estimate by HSBC for mid-2003, the daily volumes for INR NDF was about US $ 100 million. The latest available information on NDF turnover indicates a substantial pick up in NDF turnover in Indian rupee in line with the pick up in domestic onshore spot and forward market turnover. The NDF market turnover, however, remains small when compared with onshore market turnover. The daily average turnover in the spot market is about 4.4 times and forward/swap market is about 4.1 times the turnover in the NDF market (Table 2). The turnover is high for 1-month and 1-year maturity. As compared with some other Asian currencies traded in the NDF market such as the Korean won, Chinese yuan and Taiwanese dollar, the turnover in the NDF market is very small for Indian rupee.

Bid Offer Spreads

Markets are generally, perceived as efficient when market prices reflect all available information, so that it is not possible for any trader to earn excess profits in a systematic manner. The efficiency/ liquidity of the foreign exchange market is often gauged in terms of bid/offer spreads. The bid-ask spread refers to the transaction costsand operating costs involved with the transaction of the currency. With the increase in the volume of transaction of the currency, these costs/bid-ask spreads may reduce.

In India, the spread is almost flat and very low in the spot segment of the foreign exchange market. The spread in the NDF segment remains higher than that of the spot and forward market reflecting lower liquidity in the NDF market (Table 3)3 . As compared with other Asian currencies, the spreads for Indian rupee NDFs remain lower than that of Indonesian Rupiah and Philippine peso, but higher than that of Chinese renminbi and Korean won reflecting the higher liquidity available in the latter two currencies (Table 4).

Market Volatility

Looking at the volatility in the INR NDF market, it is observed that NDF volatilities have been consistently higher than their spot counterparts (Table 5). This is essentially attributed to official intervention in the spot and forward markets by the Reserve Bank. This is a feature observed for most other Asian currencies, particularlyChina, Philippines and Taiwan where intervention is an important component of monetary policy (Ma et al. 2004).

Section IV

The Linkage between Onshore (spot and forward) and Offshore (NDF) market for Indian Rupee

Causality between NDF and onshore Spot and Forward market

With an objective to examine the causal link between the spot, forward and NDF market for Indian rupee, Granger causality test was carried out. The results are reported in Table 64 . As expected, two way causation was exhibited between percentage change in spot and forward rupee-dollar rates as the null hypothesis is rejected at very low levels of significance. Further strong unidirectional causality was observed from percentage change in spot to NDF rates and forwards to NDF rates reflecting the fact that the price in the NDF market is determined primarily on the basis of the RBI reference rate. The causality from NDF to spot and forward markets, however, remained insignificant5 . Such a result seems obvious in the Indian context given the fact that the domestic market players participation in offshore NDF market for Indian rupee is limited. A similar result has also been obtained in the case of Chinese renminbi (Higgins and Humpage, 2005). According to the findings of the study, NDF market’s inability to precisely predict the future level of the spot exchange rate was due to the fact that the underlying spot exchange rate is subject to government restrictions, official interventions and large discrete changes.

Spillover effects between NDF and onshore market

To further explore the interrelation and information flows among the various onshore and offshore markets for Indian rupee, empirical tests were conducted to examine whether the higher volatility in the NDF market (as observed in previous section) has some spillover effects on volatility in the spot market. Despite the growing turnover in the Asian NDF market, studies on NDF market analysis are limited. Empirical exercises to explore the operation of the NDF markets are rather negligible. Park (2001) has tried to analyse the information flow between NDF and spot market for Korean won that is one of the highest traded currencies in the Asian NDF.

Following the methodology used by Park (2001) to examine the volatility spillover effects between onshore and offshore market, the augmented GARCH model is estimated. Prior to estimation of augmented GARCH model, the information variable (lagged residuals) is generated from a vanilla type GARCH model since the financial timeseries data are conditionally heteroscedastic and follow a GARCH (1,1). The augmented GARCH model is further extended to include exogenous variables and is utilized to investigate mean and volatility spillover effects between NDF and spot/forward markets. Thus, the GARCH(1,1) is of the form:

where     is the rupee-dollar percentage return in market i (Spot/ Forward/NDF) at time t,

from spot and forward markets to determine the returns in NDF market, thus, supporting the efficient market hypothesis for NDF market.

The coefficient γ_o reflects the volatility spillover impact between the markets. As can be observed this coefficient is significant from spot to NDF which is expected, considering that NDF market draws information from spot market. It is also important to mention here that this effect from spot to NDF is quite strong. The volatility spillover effect doesn’t exist from forward to NDF.

It is also observed that this coefficient is also significant from NDF to spot and forward markets. This indicates that there is volatility spillover from NDF to spot market and NDF to forward market although the spillover effect is not large given the low value of the coefficients in both the cases.

Onshore/Offshore Interest Rate Spreads

Another way to look at market efficiency is to analyse the extent of market segmentation between onshore interest rates and offshore interest rates implied by the NDFs. Under the covered interest parity condition, the forward exchange rate of the home currency, in the absence of capital controls, is linked by arbitrage to its spot rate and the interest rate differential between the home currency and the US dollar as set out in equation 5 :

where F is the forward rate, S the spot rate, r the interest rate on the home currency and r$, the US dollar interest rate. When there are no cross-border restrictions, borrowing and lending ensure that the above holds.

However, when capital controls bind, non-residents may not have full access to onshore credit or placements, giving rise to NDFs (equation 6).

where i is the NDF-implied yield on the home currency offshore.

To the extent that the arbitrage between the onshore money market and offshore NDF market is effectively constrained by capital controls, the NDF-implied offshore interest rate, i, can differ considerably from the interest rate prevailing in the onshore money market, r. A large and persistent onshore/offshore spread (r – i) indicates the presence of effective cross-border restrictions.

Further, the sign of the onshore/offshore yield spread can signal underlying market pressure on the currency. An onshore interest rate above its offshore NDF-implied counterpart would indicate underlying appreciation pressure on the home currency but effective capital controls limiting capital inflows into the home currency. An onshore rate below its offshore counterpart would indicate depreciation pressure but effective stemming of capital outflows. A zero spread may suggest the absence of effective capital controls, or the absence of market pressure on the home currency, or both.

Based on the above analogy, the NDF implied yield on the Indian currency offshore was computed using data for the period November 2004 to February 2007. In the absence of a meaningful/comparable onshore money market rate, an equivalent approach generally usedis to estimate the implied onshore yield on the home currency using the onshore deliverable forward premia and then to compare it with the NDF implied offshore yields. Accordingly, the implied forward yield and spread have been computed for the Indian rupee (Chart 1). The non-zero spread between the implied yield in the forward and NDF market seems to be indicative of presence of capital controls in India that weaken and prevent cross border arbitrage to some extent. During this period, the offshore implied yield has also mostly remained lower than onshore rates reflecting the general appreciation pressure on the rupee onsore during this period. On certain occasions, the spread has touched zero implying no underlying pressures on the domestic INR.

It is interesting to compare the above findings with the results of Ma et al. 2004. For the period January 1999 to December 2001, they had observed the spread to be mostly negative implying continuing depreciation pressures operating in the Indian economy. In fact this was true for most Asian currencies in the post East Asian crisis period. The absolute spread was also very high ranging from (-) 400 to (-)1000 basis points during that period. For the period 2004-2007, the spreads seem to have narrowed considerably ranging between 0-400 basis points7 . This feature could be attributed to the growing liquidity in the INR NDF markets and the diminishing capital controls in India during the past 6-7 years8 .

An interesting observation from the recent trend with regard to the implied NDF yields depicts the fact that despite substantial capital controls, Chinese yuan is observed to have a very low implied yield in the NDF market vis-à-vis other Asian currencies including Indian rupee. For example, the one year NDF implied yield for Chinese yuan is presently placed at around 0.2 per cent as against 7-8 per cent for the Indian rupee NDF implied yield. This is essentially attributed to relatively controlled currency of China along with low domestic yuan interest rates.

Section V

Policy Perspective

Policy Implications of NDF market

From a broader perspective, markets for non-deliverable currency forwards (NDFs) are of interest to policy makers because they are a product generally used to hedge exposure or speculate on a currency movement where local market authorities limit such activity. NDF prices are also a useful tool for market monitoring in that these prices reflect market expectations, and supply and demand factors that cannot be fully manifested in onshore prices in a country with capital controls. The difference between onshore currency forward prices, where they are available, and NDFs can increase in periods of heightened investor caution or concern over potential change in the exchange rate regime or a perceived increase in onshore country risk (Lipscomb, 2005).

In the Indian context, as can be observed from the previous sections, the NDF market draws its information mostly from the onshore spot and forward markets and the reverse causality is poor. The NDF market turnover volumes are also not large enough to affect the domestic onshore INR market. Yet it may be important from the policy angle to take cognizance of the developments in the NDF markets. There are evidences of volatility spillover from NDF market to onshore spot market, though not large. Hence, if not in regular market conditions, in volatile market conditions the NDF market transactions can have an impact on the domestic spot markets. It would be pertinent to monitor the NDF prices so as to tap the additional information with the NDF market players especially pertaining to dollar movements in offshore markets. Besides, prices in the NDF market can be a useful informational tool for authorities and investors to gauge market expectations of potential pressures on the exchange rate in future. Gradually, however, once India moves on to fuller capital account convertibility, the NDF transactions are going to perish. Till that time, close monitoring of the NDF market activity could be crucial. The relatively low turnover and marginal volatility spillover impact, however, discounts for the need of regulatory control on these markets.

Recent Policy Initiatives

Efforts have been made towards improving the hedging facilities that are available to offshore non-residents who have exposures to the rupee, particularly the foreign direct investors. As far as the foreign institutional investors (FIIs) are concerned, they were not allowed to rebook contracts once cancelled till recently. The Committee on Fuller Capital Account Convertibility (FCAC) recommended that to minimise the influence of NDF markets abroad, the FIIs may be provided with the facility of canceling and rebooking forward contracts and other derivatives booked to hedge rupee exposures. The Mid-Term Review of Annual Policy Statement 2006-07, announced by the Reserve Bank of India in October 2006 proposed to implement this recommendation. Accordingly, AD category – I banks have been permitted to allow FIIs to cancel and rebook forward contracts up to a limit of 2 per cent of the market value of their entire investment in India as at the beginning of the financial year. The outstanding contracts must also be duly supported by underlying exposure at all times.

Besides, recently initiatives have been undertaken to expand the range of hedging tools available to the domestic market participants to hedge economic exposures viz., importers’ customs duty component, price risk on aluminium, copper, lead, nickel and zinc in international commodity exchanges, actual users of aviation turbine fuel (ATF) to hedge their economic exposures in the international commodity exchanges, free cancellation and rebooking of forward contract up to 75 per cent, 100 per cent for corporates with overseas direct investments and small and medium enterprises (SMEs).

Section VI

Concluding Observations and Future Outlook

With the growing interest of the global players in the Indian economy and the consequent increase in capital inflows into India over last few years, there has been an increase in liquidity in the INR NDF market. This market essentially draws information from the onshore spot and forward markets. The bid-offer spread and volatility in this market has, however, generally remained higher than that of onshore counterparts. While the offshore NDF rates do not directly influence the spot and forward rates in the domestic market, volatility in NDF markets is empirically tested to influence the onshore markets.

Looking forward, India is expected to remain an attractive destination for foreign investors in near future. Consequently, transactions in NDF market may rise to hedge against rupee exposure. The activity in the NDF market will also be governed by the following factors: (a) further move towards capital account convertibility, the roadmap for which has been provided by the FCAC (b) further development of financial markets particularly in terms of better fulfillment of interest rate parity conditions (c) introduction of more and more derivative products for both onshore and offshore investors so as to provide them more avenues for hedging in the domestic market, particularly to offshore investors and lastly, (d) complete phasing out of the underlying exposure criteria for booking a forward contract.

With India continuing its initiatives towards further opening up of the capital account, the NDF market activity is expected to decline and finally disappear. However, so long as there is differential treatment between residents and non-residents with regard to their operation in derivatives market, offshore INR NDF market will continue to thrive. Till that point, the NDF markets could be regarded as essential for sustenance of offshore investor’s interest in the domestic economy. Activity in the NDF market needs to be closely monitored to keep track of the pressures operating on the Indian currency as well as to prevent speculative attack on Indian currency during volatile conditions.

Notes:

1 n case of Malaysia, it is also stated that “offshore banks possibly refrained from engaging in  creative transactions such as NDF to circumvent the controls so as  not to risk their local franchise.” In countries where major market players have strong incentives to build and maintain good relationships with regulators, the  effectiveness  of regulations is more likely to be ensured (Ishii, 2001).

2 Korean authorities briefly sought to limit the effect of NDF demand on local markets by restricting local banks’ participation in the NDF market. The presumed intent was to lessen the need for central bank intervention as the effect of foreign demand for long won positions would be limited onshore. The regulations were reportedly seen as adversely affecting local banks and were subsequently largely rescinded.

3   The percentage spread is calculated as: (Ask-Bid)/ [ (Ask+Bid)/2] * 100.

4   The detailed discussion about the estimation is reported in Annexure II.

5   NDFGranger causes spot at 10 per cent level of significance.

6 The constant term in conditional variance equation, ARCH and GARCH coefficients are significant and have expected signs except in the last case where the constant term is insignificant.

The detailed discussion on estimation and empirical results is reported in Annexure II.

7 There was one occasional spike in the spread to 800 basis point on December 26, 2006 associated more with liquidity pressures in the domestic market on that day.

8 Any interpretations of onshore/offshore interest spreads need qualifications. Ideally, the comparison should be between a liquid onshore bank interest rate and a similarly liquid offshore implied rate. But the fact that the domestic money market is most liquid at short maturities, while NDF markets tend to be more liquid at medium to long maturities, makes it hard to find good liquidity at matching maturities.

References:

Higgins, Patrick and Owen F. Humpage (2005), “Nondeliverable Forwards: Can We Tell where the Renminbi Is Headed?” Federal Reserve Bank of Cleveland, September.

Ho, Corrinne, Guonan Ma and Robert N. McCauley (2005), “Trading Asian currencies”, BIS Quarterly Review, March.
Ishii, S, I Otker-Robe and L Cui (2001), “Measures to limit the offshore use of currencies: pros and cons”, IMF Working Paper, WP01/43, April.

Lipscomb, Laura (2005), “An Overview of Non-Deliverable Foreign Exchange Forward Markets”, Federal Reserve Bank of New York, May.

Ma, Guonan, Corrinne Ho and Robert N McCauley (2004), “The markets for non-deliverable forwards in Asian currencies” BIS Quarterly Review, June.

Park, Jinwoo (2001), “Information flows between non-deliverable forwards (NDF)/and spot markets: Evidence from Korean currency”, Pacific Basin Finance Journal, 9(4), pp. 363-377.

Reserve Bank of India (2005), “Report of the Internal Technical Group on Foreign exchange Markets”, www.rbi.org.in.

Reserve Bank of India (2006), “Report of the Committee on Fuller Capital Account Convertibility”, (Chairman Shri S.S.Tarapore).

Singhania, Kamal (2006), “Non-deliverable forwards” in Developments in Money, G-sec and Forex Markets in recent years, The Clearing Corporation of India Ltd.

ANNE X II

Data and some empirical results

Data

The data used in the study are daily series and collected from the Reuters database. The study spans from December 10, 2004 to February 9, 2007. However, the period chosen is different for the tables and empirical analysis as per the data availability and comparability since the longer timeseries data are not available for all the currencies. Also the NDF turnover data are available for very short period.

The exchange rates (spot, forward and NDF) used in this paper are the closing prices for bid and ask. The bid-ask spreads reported in the study are the average percentage spreads, i.e. the difference between ask and bid prices, and divided by the average of the two multiplied with 100 for expressing in percentage terms.

Empirical results

All the variables used for empirical work have been tested for unitroots. However the exchange rates are non-stationary at their levels, they are stationary when taken in percentage return terms in all the cases.

For Granger-causality tests, the variables used are spot, forward and NDF returns and they are stationary. The lag length is selected on the basis of final prediction error (FPE) and log-likelihood criteria and it is 6.

For estimating the augmented GARCH model, i.e. equation (3) & (4), the lagged residuals are derived from the GARCH (1,1) model, i.e. jointly estimating equation (1) & (2). The GARCH (1,1) is estimated to model spot, forward and NDF volatility where a variety of mean equations, i.e. ARMA models are chosen on the basis of AIC and SC criteria# . All the pre- and post-diagnostic tests like testof ARCH effect (LB-Q test, ARCH-LM test), coefficient stationary test (Wald test) were conducted and all the results were found to be robust. Further, all the coefficients in variance and mean equations were also statistically significant without violating any non-negativity restrictions. The results are reported in Table A1.

* Smt Sangita Misra and Shri Harendra Behera are Research Officers in the Division of International Finance, Department of Economic Analysis and Policy. The authors would like to thank Dr R.K.Pattnaik for being a constant source of encouragement and Dr Rajiv Ranjan for giving the opportunity to get introduced to the concept. The views expressed in the paper are those of the authors and not of the institution to which they belong.

# Descriptive statistics were calculated for all the return series and were found to be fat tailed leptokurtic and non-normal. Further, before estimating GARCH(1,1) models, the unconditional mean equations were also estimated to check the presence of ARCH effect and the non linear autorrelation was found to be significant up to lag length 8 providing evidence for presence of ARCH effect.