The Extent to Which the Forward Market Is an Unbiased Predictor of the Spot Market - Literature review Example

A REPORT ON: THЕ ЕХTЕNT TО WHIСH THЕ FОRWАRD МАRKЕT IS АN UNBIАSЕD РRЕDIСTОR ОF THЕ SРОT МАRKЕT Written By (Your name) Institution (Date of submission) Abstract This report seeks to explore the extent to which the forward market is an unbiased predictor of the spot market. The report focused on the exchange rate market of the South African Republic and not on the stock market. The conclusions were reached after evaluating relevant theory and examining earlier works of authors on the issue of forward and spot rates. Central Banks throughout the world in formulating monetary policy always consider the relationship between forward rates, spot rates and interest margins. The forward rate (Ft) and spot rate (St) vary according to the inclusion of new information and the omission of previous observation does serve to create errors in prediction of the spot rate. This and other factors such as economic agent’s behavior and existence of risk premium have made the author of the report to conclude that the forward market is not an unbiased predictor of the spot market. Table of Contents Abstract………………………………………………………………………………………………………………..……………………..2 Introduction………………………………………………………………………………………..…………………………….…………4 Background Information……………………………………………………………………………………..…………………….4 Literature Review……………………………………………………………………………………………………………………….6 Conclusion……………………………………………………………………………………………………………….………………….12 References………………………………………………………………………………………………………………………………….. 13 Introduction Chiang (1985) argues that in deciding which monetary policy to adopt, central banks across the world must consider the existing relationships between forward rates, interest rate margins and spot exchange rates. Interruption of these relationships is likely to cause speculative transactions in the foreign exchange market. The author states that the question on what extent can markets remain efficient in order for the forward market to reliably predict the spot market remains an interesting question that can only be answered in a structural manner by use of econometrical analysis. Fama (1984) argues that new information does play a role on new exchange rate movements and has indicated the random walk hypothesis (RWH) can be used for the predictions of spot rates. This hypothesis has been discussed in relation to forward and spot rates. Background Information Cornell (1987) argues that there is the belief that forward market rate is an unbiased predictor of the spot rate and that if not so, foreign currency speculators could make profits from the bias by maintaining one position on the spot market and an opposite position on the forward market. The author gives the example of the US dollar whereby the if forward rate under-predicts the future spot rate, speculators could end up buying US dollars while at the same time maintaining a short term forward position. He states that the day when the forward rate contract falls in due, the position maintained by the speculators could, be closed at a considerable profit margin, this is because the spot rate would as such be above the rate indicated in the forward rate contract. This argument holds that the existence of such a bias is not consistent with the concept of an efficient market. This report is based on the behavior of the exchange rate market in the Republic of South Africa from the period 1981 to 1998. The report therefore positively adds to the existing body of works on the rationality of forward and spot markets in reference to the exchange rate markets and whether economic agents fully take into account all the relevant information in their predictions of the Rand/US Dollar rate of exchange. There is need to uncover agents’ behavior since their speculative behavior influences the forward and spot rates in the exchange rates markets. To guide analysis in this paper, it is necessary to first distinguish the difference between a forward market and a spot market. Bilson’s model of predicting the spot rate have been brought out and works of other authors on these models have also been discussed. Ramanathan (1995) makes a distinction between forward and spot markets. He argues that in a spot market, the two parties taking part in the transaction make arrangements to exchange currencies within a short term horizon. The contracts are effective immediately and prices are settled on the spot at current market prices. The author further argues that in a forward market an entity has the opportunity to make prior arrangements to purchase foreign exchange for the purpose of making a future international payment whereby the transaction between the two parties involves agreeing on the exchange rate today. He further adds that forward markets are mainly attractive to those engaged in international trade, for example, importers who want to protect themselves against future exchange rate fluctuations. Bilson (1981) brought out an empirical model for predicting the spot rate using a different currency and the US Dollar whereby the exchange rate is determined how much make a dollar. The model comprised of the below equations: St+k = a + bFt + et+k ……………………....(1) St+k = a + gSt + et+k………………….........(2) St+k = a + bFt + gSt +et+k……………….…(3) St+k = a + bFt + cFt-1 +et+k………………..(4) Whereby: St = logarithm of spot rate at time t and (say month t) Ft= logarithm of forward rate at time t (say month t) k = the t+3 settlement period i.e. three month settlement period a, b, g, and c are all constants et = error term in the period t+k. Baekart and Hodrick (1993) agree with Bilson (1981) and argue that the first equation is used to test the hypothesis that forward exchange rate is an unbiased and reliable predictor of the spot rate. They further argue that the equation brings out the issue of rational expectation in the absence of risk premium. The authors hold the assumption that those who participate in the market remain neutral with regards to risk, have rational expectations and that expected values of the determinants of the exchange rate are discounted to their present values. Going by their argument, it can be deduced that proper information for predicting the exchange rates is fully incorporated in the current forward market rate. Ukpolo (1995) has a different view and discusses the second equation in a way that brings out the random walk hypothesis of exchange rate behavior. The author states that in the random walk theory, the current foreign exchange spot rate is the most reliable indicator of the forward rate. He argues that like any other asset market, the foreign exchange market is very efficient in that all the relevant information needed to determine the exchange rate is fully incorporated in the current spot rate. This in itself is a supporting argument only that the author overlooked the fact that the second equation transforms into the Martingale Model (Ukpolo 1995) if any changes in the exchange rate portray dependency in a serial manner. It can be deduced that forward market can be an unbiased predictor of spot rate by testing the joint hypothesis of a=0 and g =1 whereby the end result should be a rejection of the joint hypothesis. Longworth (1981) on his part argues that the third equation is more flexible compared to the first two and that the equation brings out the notion that the spot rate is simply a weighted average of the present forward rate and the corresponding or matching spot rate. This, he argues, shows that participants in the exchange rate market acknowledge the importance of the relevant information contained in the logarithm of forward rate at time t (Ft) and logarithm of spot rate at time t (St) in order to predict the spot rate. The author’s argument is mainly supported by two testable hypotheses a=0 and b+g=1 of which are similar in principle to the models presented by Bilson (1981). Both authors are in agreement in their classification whereby should the forward market become efficient, then the hypotheses g=0 and b=1 should not be rejected. The author concludes that should market expectations remain unchanged then g=1 and b=0 should not be rejected. Phillips (1986) argues that the fourth equation is merely an equation that is general in nature, meaning it is an equation that is usually included to the model for determining the spot rate. He further argues that if the exchange rate market is efficient and there exists zero risk premium then (Ft) contains information necessary for the prediction of St+k as portrayed in the first equation. He goes on to state that Ft-1 only serves to test if advanced fall backs in the forward market rate still have unused information and therefore it should not lead to an increase of explanatory power. From his argument it can be deduced that should the hypothesis c=0 be rejected then it would show that the foreign exchange market is inefficient because the forward rate that is lagged still contains relevant unused information that can be relied upon to predict the spot rate. Fama (1984) argues that since the 1970’s, the Bretton Woods structure of fixed parities was abandoned and the forward market has taken up a pivotal role in protecting against currency fluctuations in future spot markets and exchange rates and that the efficiency of hedging against exchange rate fluctuations, is dependent partly on the relationship between the spot and forward markets and or their exchange rates. If, for example, the forward exchange rate is solely considered to forecast spot exchange rates, then instability in the forward-spot rate relationship generates larger prediction errors. The author further argues that various authors’ arguments on monetary policy have rotated on if the forward market comprises of relevant and helpful information about the future direction of the spot market and that if the forecasted variation in the exchange rate is equal to the interest discrepancy among the currencies, and the variation among the spot and forward exchange rates, then it can be said that the forward premium is equal to the interest rate difference. The author states that the symmetrical forward exchange rate acknowledged now in a competent market for supply of foreign exchange in t periods forward, it should be regarded as the best and reliable forecaster of the spot exchange rate yielded at (t) periods later. The author also argues that the forward rate is not the best forecaster of the spot rate because it continually miss predicts the path via which subsequent spot rate changes. This he connects this to the presence of a risk premium and to ineffective processing of information by participants of a market. He therefore concludes by arguing that the forward rate contains little information as to remain an unbiased predictor of spot rates. Chiang (1985) discusses the expectations theory in relation to the forward market model. The author argues that the forward market rate fully portrays all relevant information as regards expectations about the exchange rate and as such, the forward rate of exchange is usually seen to be an unbiased determinant or predictor of the spot rate. He argues that the information contained by the forward rate can be quickly processed by economic agents and that the forward rates reflect relevant information through market adjustments and arbitrage activities of economic agents. The author also argues that unexpected variations of several exchange rate determinants may not always be important for the reason that in some economies it may take market participants over a month to come across the new information, which could be because of a delay before announcements of concrete changes in some variables are made. However, he further states, in the market efficiency hypothesis (MEH), the moment ‘new information’ is taken into account in the exchange rate, previous innovations should not have an effect on the spot rate. The author goes on to state that exchange rate markets tend to be subjected to market inefficiencies and that whereas the forward rate is a good forecaster of the spot rate, it does not take into account all of the information that is readily available to economic agents. This variation in exchange rates, he concludes, is most likely caused by new information regarding economic issues for example variations in the domestic and international quantities of currency, real interest rates and real incomes that were missing when the forward rate was initially determined. Fama (1984) argues that a forward rate is often understood to be the total added value of a premium plus the expected future spot rate. Indicatively, the forward rate ft observed for a particular exchange at time t+1 becomes the market determined certainty equivalent of the long run spot exchange rate st+1. The author makes the assumption that the forward market is both economical and rational, and has produced several studies that have found evidence whereby each of the elements of forward rates vary through time. He sums up his argument with two conclusions.  The primary conclusion is that the majority of the variations in forward rates occur because of the variations within the premiums and the second conclusion is that the premium components of the future spot rate elements of forward rates negatively co-related to each other. Ngama (1994) argues on the importance of considering the relative performance of (St) and (Ft) as predictors. He argues that by using the minimum in-sample root mean square error (RMSE), the spot rate becomes superior to the forward rate in predicting future spot rates and that the qualified presentation of these two rates is directly fixed to the character of exchange rate activities and/or the nature of expectations development. Conclusion It can be concluded that if expectations are shaped realistically and, in the same way, then forward rate suitably portrays the expectations of exchange rate determinants noting that in such a case the forward rate is likely to make good in predicting the future spot rate. By differentiation, if expectations are changing at a slow rate or the exchange rate follows a haphazard process, then the spot rate will perform better. If however the expectations are constant, a partiality may transmit to the tendency of movements of the exchange rates all through the complete period in question not forgetting that if there exists an amplified increase in (St), then a more adaptive expectation will always undervalue its future value. A second conclusion is that the coefficients on both (Ft) and (St) are responsive to the addition of new analysis and the omission of the previous observation. The expected coefficients differ with the measurement period time (t) this being suggestive that the implicit invariable element will produce subjective estimates. Noting the point that the expected coefficients from the above equations have a value of less than one, demonstrating that use of forward rates as predictors of the spot rates would result to a descending bias thereby creating errors in prediction. This leads to a conclusion that the unbiased predictor hypothesis (UPH) should be rejected. The likely justification for this bias is therefore the continued existence of risk premium, market participants’ inability to process to process information and irrational formation of expectations. References Baekart, G. and Hodrick, R.J., (1993), On Biases in the Measurement of Foreign Exchange Risk Premiums, Journal of International Money and Finance, Vol. 12, pp 115-138 Bilson, J., (1981), The Speculative Efficiency Hypothesis, The Journal of Business, Vol. 54, pp 435-451 Chiang, T.C., (1985), The Impact of Unexpected Macro-Disturbances on Exchange Rates In Monetary Models, Quarterly Review of Economics and Business, Vol. 25, No. 2, pp 49 - 59 Cornell, B., (1987) Spot Rates, Forward Rates and Exchange Market Efficiency, Journal of Financial Economics, Vol. 5, pp 55-65 Fama, E.F., (1984), Forward and Spot Exchange Rates, Journal of Monetary Economics, Vol. 14, pp 319 - 338 Longworth, D., (1981), Testing the Efficiency of the Canadian - U.S. Exchange Market Under The Assumption of No-Risk Premium, Journal of Finance, Vol. 36, pp 43-49 Ngama, Y.L., (1994), A Re-Examination of the Forward Exchange Rate Unbiasedness Hypothesis, Weltwirtschaftliches Archive, Vol. 130, No. 3, pp 447 - 460 Phillips, P.C., (1986), Understanding Spurious Regressions in Econometrics, Journal of Econometrics, Vol. 33, pp 311 – 340 Ramanathan, R., (1995), Introductory Econometrics, New York, Harcourt Brace College Publishers Ukpolo, V., (1995) Exchange Rate Market Efficiency: Further Evidence From Cointegration Tests, Applied Economic Letters 2, pp 196-198. Read More