Essay: Predictability of exchange rates (dissertation proposal)

1.0 Introduction.
The exchange rate of a country in basic terms is the price of the country’s currency in terms of another currency; it is the rate at which one currency can be exchanged for another. The use of the exchange rate can be seen as a value mechanism for currencies on the international market (Froot and Stein, 1991). The history and development of the exchange rate can be traced to the emergence of national currencies since it facilitated trade between countries. The earliest most structured exchange rate system was the gold standard during the nineteenth century where currencies were converted into their respective gold value. After the collapse of the gold standard in 1914, the Bretton Woods Agreement (BWA) was signed which developed a system where all currencies were pegged to the U.S. dollar. By 1970 the BWA was already under threat from excessive supply of the U.S. dollar leading to its collapse in 1971. The Smithsonian Agreement which came into effect in 1971 under the Nixon led U.S. government lead to the determination of the exchange rates by market forces giving birth to the Floating system.
In present times there exists three main exchange rate regimes these include the Free-floating regime where exchange rates are determined by demand and supply forces. The second regime is the pegged or fixed exchange rate regime where the exchange rate is not allowed to fluctuate by government authority. The hybrid regime is a mixture of the fixed and free floating where the exchange rate is allowed to float within a specific range determined by the Central Bank.
The exchange rate market is a multi-Trillion-dollar market that is estimated by the Bank of International Settlement to have a daily turnover of over 4 trillion U.S. dollars. These estimations make the exchange rate market the largest asset class in trade volume.
The exchange value of a currency affects every part of a nation’s economy. Giving its importance and significance to international trade and national growth, many attempts have been made to predict future exchange rates by economic players in order to make profits or design economic policies. Many Economic models have been developed from the start of the Smithsonian agreement to try and predict exchange rate movements. These models such as the structural models that use economic indicators like interest rates, trade balance were designed in a bid to outperform the random walk movement of exchange rates.
The efficiency of these models had been an issue of great debate in academic circles but the work of Meese and Rogoff (1983) in their seminar paper drew an empirical conclusion on the structural models and the random walk. Using out of sample data they showed that no structural models can outperform the random walk models in predicting exchange rates. In their research Meese and Rogoff used different models like the Flexible-price Monetary model, the sticky-price monetary and the Dornbusch-Frankel model to forecast a twelve-month horizon for the dollar/pound, dollar/mark, dollar/yen and trade-weighted averages.
Cheung, Chinn and Pascual (2005) in their paper “Empirical exchange rate models of the nineties: Are they fit to survive?” examined more newer models like the interest rate parity specification and the composite specification incorporating a number of channels identified in different theoretical models. At the end of their empirical examination, they came to a similar conclusion as Meese and Rogoff and the predictability of these models against the random walk.
Despite the seemingly disappointing conclusion on the predictability ability of the structural models to predict exchange rates many researches have been dedicated to establishing a good fit of economic fundamentals to exchange rate. Engel and West (2005) (EW05) in their paper “Exchange Rates and Fundamentals” established a correlational theoretical frame work between exchange rate movements and Taylor rule fundamentals. The Taylor rule which was introduced by John Taylor in 1993 is a monetary policy tool used by Central banks to control inflation and output gap through the setting of the interest rate. In their 2005 paper, Engel and West derived a discounted present-value model using the Taylor rule. Engel and West, 2006 (EW06) went further to use their developed present value model to predict the Deutschmark/Dollar real exchange rates and found a positive correlation between actual and the predicted exchange rates.
Based on the works of EW05 and EW06 and other empirical works, this paper will perform an empirical and methodological review of other works that seek to underline and establish the relationship between the Taylor rule fundamentals and exchange rate movements. This will help in the mapping of the development of the use of the Taylor Rule as a new direction in the prediction of exchange rates. Four emerging and developing economies namely Russia, South Africa, Indonesia and Brazil will be used for the analysis.
The next part of the paper will carefully examine the different data sets for the selected countries and introduce the methodologies in estimating the Taylor rule fundamental variables. A brief introduction of the various variations of the Taylor rule will be done with emphasis on their main differences. An analytical comparison of the T.R. models would be done to determine the most efficient deterministic factors.
The last part of the paper would focus on using the present value model developed by EW05 to predict the exchange rate moments of the selected economies against the Dollar. The exchange rate predictions would be based on some assumptions:
1. All central banks use the Taylor Rule to set interest rates.
2. All central banks follow a flexible exchange rate regime.
Based on the results, a final conclusion would be made on the predictability of exchange rates using Taylor Rule fundamentals in the case of developing counties.