Efficiency of Currency Derivatives in Price Discovery Process: Evidences from India

The present study investigates the efficiency of currency derivatives market by assessing its contribution towards price discovery process using spot and future prices of four currencies (USD/INR, EURO/INR, GBP/INR and JPY/INR) traded on the National Stock Exchange (NSE), India. As per the investigation, it can be concluded that there is a long run equilibrium relationship between spot rates and future rates, with unidirectional causality running from future rates to spot rates for all currencies under consideration. As the futures markets contribute more to the price discovery, this implies that more investors are attracted to it. This in turn leads to more rational price discovery.


Theoretical Economics Letters
The market participants include retail investors, brokers, multi-national corporations, public and private banks, central banks, hedge funds, etc.
The forex market in India has come a long way in terms of the number of participants, overall market turnover and variety of instruments. Today, Indian forex market is an attractive destination for investors and has seen an increase in the turnover for different foreign currencies. Over the years, the increase in the number of participants and the thriving derivative products for trading and hedging risk have added to the depth and complexity of the market. Due to this increase in the number of the participants and the changing structure of the market, there is a need of increased transparency, stability, and efficiency. This is one of the reasons currency futures were introduced.
In line with the above thought, RBI has been trying to adjust many policies and formulate new ones to make the forex markets more efficient. For that purpose, one needs to better comprehend how the market and the various components in it behave. One of the ways of that is to understand the price discovery process in the spot and futures market.
Furthermore, currently with the development of the forex markets, especially the futures market, most of the research has been done on the lead-lag relationship and the efficiency of currency derivatives in determining the price discovery process in the developed countries. There is dearth of such studies on the developing economies. Also, the major focus of previous studies have been on assessing the efficiency of futures in determining the price discovery process and lead-lag relationship in the stock markets, commodities market and not many efforts have been made in the context of currency derivatives.
The current study talks about the lead-lag relationship between spot and futures prices in the context of Indian forex markets. This information can be used to device a trading strategy so as to earn more returns as compared to the passive buy and hold strategy. This way the results of the price discovery examination can be used not only by the regulators but will also be beneficial to the market participants.
In terms of the limitations of the paper, the study can be expanded to other currencies and nations in order to expand the horizon of price discovery process. Other empirical techniques can be applied in order to carry out the analysis so as to if the final results come out in the same way or not. Section 2 summarizes the important literature on this topic. Data and methodology relevant to this study are discussed in Section 3. The results of the empirical analysis are illustrated and reported in Section 4. The conclusions of the study are given in Section 5. Finally, the future scope of this study is discussed in Section 6.

Literature Review
Price discovery is a key function of the financial markets [1]. In an efficient market, the effect of all the information available to the market participants is  [5], all the available information should be fully reflected in the asset prices. The purpose of price discovery analysis is to investigate as to which market is providing most useful information for determining the fundamental value of the asset. Unfortunately, so far, the concept of price discovery in the financial markets has not been completely grasped.
To know if the spot or the futures market is dominant in the respect of price discovery and which one of them takes the lead is of vital importance to the market participants as well as the regulators. Due to this, the quantification of the process of price discovery has gained importance in the area of empirical finance. Most widely held view is that the process of price discovery is dominated by the futures markets, across different assets [6] [7]. This is majorly because of the lower transaction costs. The process of price discovery has been widely studied under a number of settings, particularly the price discovery of stocks listed on different exchanges. The role of price discovery and volatility spill-over in informationally linked futures market has been studied extensively. Most of the researches on price discovery and the possible volatility consequences for the spot market are focused on a few developed countries, especially the USA. [8] and [9] have studied the price discovery of the stocks in the USA which are traded on different exchanges, with reference to the relative contributions of the NYSE and the regional exchanges. After analysing the commodity spot markets and futures markets, [10] has stated that about 75 percent of the new information first gets incorporated in the future prices. Articles describing the links between stock markets and stock index futures markets include [11]- [16]. Researchers have also studied other equity derivative products. The links between stock markets and option markets have been studied by [17]- [23]. Some other important studies which focus on the price discovery and volatility spill-overs in commodity and stock market's index futures are [24] [25] and [26]. The price discovery mechanism in India's National Stock Exchange (NSE) spot market and future market was studied by [27]. The study uses Johansen's cointegration test and Vector Error Correction Mechanism (VECM) to empirically conclude that there is a bilateral causality between S&P CNX Nifty index and the index futures, for the period from June 2002 to February 2005. Only few researches have tried to examine the price discovery process in the forex market, especially the forex futures market. [28] [29] [30] [31] [32] are some of the major studies that have analysed the foreign exchange markets in mature and emerging economies. [33] has studied the price discovery process in the currency markets by incorporating new information into the exchange rate dynamics. For the European markets, [34] has studied the impact of macroeconomic news and central bank announcements on the exchange rates. [35] has tried to study as to which market (spot market or future market) has greater contribution to price discovery. These studies conclude that in comparison to the spot market, the futures market have greater contribution to price discovery in the currency markets. Theoretical Economics Letters One major shortcoming of the existing studies on price discovery is that most of the studies focus on measuring price discovery over a relatively short period of time, typically less than a year. As such, variations in price discovery over time have not really been examined in the literature. Additionally, the process of price discovery and volatility spill-over has been studied majorly for the equity and commodity markets in the USA and other developed countries [36] [37] [38] [39]. Not much attention has been given to foreign exchange markets in the emerging countries. However, because of the increasing activity in the currency markets of emerging countries, it is of strong interest to see whether the new information is fully and immediately incorporated in prices of traded currencies or not. As far as the examination of price discovery is concerned, the published evidence on Indian economy is very scarce. There is hardly any study on India's foreign exchange market that has investigated the price discovery, majorly because of the recent origin and smaller size of the market. The present study is an attempt to fill this research gap. Before proceeding for the analysis, the descriptive statistics for all the variables are mentioned below (Table 1). a) Testing for stationarity and co-integration of variables Prior to the analysis, the Augmented Dickey-Fuller test (ADF) was employed to check for the stationarity of the spot and future exchange rates for all currencies [40]. The ADF tests were performed on both the level and first differenced observations and to determine the degree of integration of the data series. This was done to check for the possibility of spurious relationships between spot and future prices.

Data and Methodology
Consequently, in the case of non-stationarity, the maximum likelihood approach of [41] is used to investigate the presence of long-term equilibrium relationship between the spot and future prices. This test reports the maximal eigen-value and the trace statistic tests for each of the variable. The null hypothesis states that there is no co-integrated relationship between the variables. Presence of a co-integrated relationship between two variables is an evidence of the long-run equilibrium relationship between them. This also implies that there is a linear combination of the two variables that is stationary. As the test results are sensitive to the number of lags chosen, information criteria have been chosen to select the best lag length.
b) Causality analysis Granger causality test has been undertaken to test for the possibility of causality relationships between the spot exchange rates and the future exchange rates where Δ, as usual is the first difference operator and e t−1 is the lagged residual term from the co-integrating regression, which is nothing but the error correction term. The above equations are used to evaluate the causality between changes in spot rates and future rates. In Equations (1), (3), (5) and (7), existence of a causal relationship running from change in future prices to spot prices is evaluated. In Equations (2), (4), (6) and (8), existence of a causal relationship running from change in spot prices to future prices is evaluated. If the dependent variables can be driven by the error term obtained in the co-integrating equation, this implies that there exists a long-run causal relationship between the dependent and independent variables. This can be done by testing H 0 :λ = 0 (or H 0 :δ = 0). If the error correction terms are non-significant, this implies that the dependent variable responds only to the short-term shocks. Similarly, short-term causality is confirmed by the rejection of the non-significance of the difference explanatory variables by the Wald test.

Empirical Results and Discussion
Co-integration analysis necessitates that the variables under consideration be integrated of the same order. Hence, it is necessary to undertake unit-root tests before co-integration analysis. Table 2 reports the results for ADF test (only   constant). It can be seen that for all currency pairs at level, the null hypothesis of nonstationarity cannot be rejected at 10% level of significance (for both spot rates and future rates). However, when first differences are taken, the null hypothesis of non-stationarity is rejected for spot rates and future rates of all the currency pairs. In other words, these variables are integrated of order 1. This satisfies the necessary condition for co-integration to exist. The next step is to test for the presence of long-term (co-integrating) relationship between spot rates and future rates. The results of Johansen co-integration test are summarised in Table 3 and Table 4, which report the maximal eigen-value and the trace statistic tests for spot rates and future rates for each of the currency pairs (For co-integration analysis, Hannan-Quinn information criteria suggests an optimal lag-length of 2 for all the currency pairs under consideration).
The null hypothesis states that there is no co-integrating relationship between the spot rate and the future rate. The above results reveal that both the trace statistics and max-eigen statistics reject the null hypothesis of zero co-integrating vectors in favour of one co-integrating vector for all the currency pairs at the 5% level of significance. The existence of co-integration relationships among the spot rates and future rates suggests that there must be causality in at least one direction.
To determine the direction of causality, along with being able to distinguish between short-run and long-run causality, Vector Error Correction Model (VECM) is deployed. Table 5 provides the joint Wald F-statistics of the lagged explanatory variables. These tests provide an evidence for or against the significance   of the short-run causal effects. Finally, Table 6 provides the t-statistics for the coefficients of the error correction terms, which give an indication of long-run causal effects. Results in Table 5 imply that in the short run, there is causality running from the future rates to the spot rates for all currency pairs except JPY/INR. F-test statistic is not significant for JPY/INR at 5% level of significance. Table 6 shows that all the error correction terms (those capturing causality relationship from the future rates to the spot rates) are negative and have t-statistics which are significant at 5% level of significance. This establishes the unidirectional causality, running from the future rates to the spot rates, in the long run for all the currency pairs. The error correction term depicts the speed of correction of the short-term rates towards the long-term steady. Adjustment in spot rates is maximum for JPY/INR at a speed of convergence to equilibrium of 82.5%. The negative values of the error correction terms confirm the stability of the system. Also, the speed of adjustment for all the currency pairs is fairly rapid.

Conclusions
To understand the nuances of the market and its components, one needs to understand the price discovery process. The relationship of the currency futures and spot market for the four currencies-USD, GBP, EURO and JPY is analysed.
As per the above results there is a long run equilibrium relationship between the futures and spot markets for all the four currencies considered. This study concludes that the spot market follows the futures market so as to attain the equilibrium in the long run. As the price is first reflected in the futures market, arbitragers and speculators take advantage of this mismatch. This moves the spot market price towards the equilibrium. Futures market being more efficient in incorporating new information, plays a major role in price discovery. This goes to show that the Indian forex market suffers from informational efficiency.
Even when the trading volume is higher in the foreign exchange spot market, the futures market has an important role to pay in price discovery. This is supported by the fact that there is much higher level of transparency in the future market as compared to direct interdealer spot trading. The future prices lead the spot price due to the assimilation of the new information in the short-run. Also, the cash market for the currencies under consideration is very well developed. [42] states that in the case of lower transaction costs and faster trade execution, secondary markets can play an important part in the price discovery process.
This information can be used by regulators like RBI and SEBI to lay down better policies for the growth of the markets. As the futures markets lead price discovery, this goes to say that the participants are more attracted towards the futures market. Thus, it is crucial for the regulators to maintain better inspection and control over the spot markets as well. Because the investors are more attracted to a fair, transparent and efficient market, this leads to greater number of participants in the futures market. Greater number of market participant ensures a more reasonable and democratic price discovery.

Future Scope of Study
The present findings of this paper on the price discovery process can be complemented by examining the volatility spill-overs between spot and futures market. This may help one to understand completely the asymmetric reactions in the forex market. This will help policy makers, portfolio managers, traders and other market participants to identify the source of volatility and to take corrective measures.