Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorHunter, J-
dc.contributor.advisorCostantini, M-
dc.contributor.authorMirkin, Lorice-
dc.descriptionThis thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University Londonen_US
dc.description.abstractThis thesis pertains to international finance and models of exchange rate determination as well as efficiency of the market for foreign currency. The first chapter is an introduction where we discuss the advent of flexible exchange rate regimes and the development of monetary models of exchange rate determination as well as present a framework for this thesis. In the second chapter we consider the historical failure of monetary models of the exchange rate and revisit the standard real interest differential (RID) model (Frankel, 1979a). The Great British Pound (GBP) and Canadian Dollar (CAD) vis-à-vis the United States dollar (USD) are examined during the period 1980:Q1 - 2015:Q4, a time characterized by flexible exchange rate regimes and heightened capital mobility across borders. Unit root properties of the sample variables are examined and the Johansen (1995) methodology is applied to test for cointegration. The RID model yields a single cointegrating relation however tests of long-run exclusion (LE) and weak exogeneity (WE) show that the RID model is not a coherent model of the GBP and CAD against the USD. The study is furthered by examination of the hybrid monetary model (Hunter and Ali, 2014). The hybrid model is tested for comparison with Japan, as the post 2007-2009 financial crisis period is branded by zero-lower bound interest rates, a phenomenon first experienced by Japan for any prolonged period of time. The hybrid model in addition yields a single relation however tests of LE and WE show that the long-run projection is reversed and that a coherent relationship exists between the GBP and CAD vis-à-vis the USD and variables related to monetary fundamentals as well as long-run economic activity. In the third chapter we examine efficiency of the market for foreign currency. The lead-lag pricing relationship between spot and futures rates is discussed and a panel employing data for the GBP, Australia Dollar (AUD), CAD, Brazilian Real (BRL) and South II African Rand (ZAR) vis-à-vis the USD is constructed at several intervals prior to expiry. The Johansen (1995) methodology is applied and shows that spot and futures rates cointegrate and that the cointegrating vector is the basis. Unit root properties for the basis are also examined and found to be integrated of order one or I(1). We therefore show that the market for foreign currency functions efficiently and that profitable arbitrage opportunities exist that restore prices to parity levels. This study is of particular significance in view of the markets’ growing share and need for greater transparency to lay down appropriate regulation that limits systematic risk. In the fourth chapter we re-examine monetary models of the exchange rate and consider the USD vis-à-vis the Japanese Yen (JPY) in view of the Japanese economy’s slow growth in the post 2007-2009 financial crisis period. We test the standard RID monetary model as a framework for modelling the USD/JPY exchange rate however tests of WE show that the nominal exchange rate is weakly exogenous so drives the system instead of adapting to it. The hybrid monetary model developed by Hunter and Ali (2014) is adjusted in consideration of the current period of sluggish economic growth in Japan by incorporating differentials related to traded and non-traded goods productivity (Rogoff, 1992). The adjusted hybrid model produces a single cointegrating relation and joint tests of LE and WE show that the nominal exchange rate cannot be long-run excluded and is not weakly exogenous so that the adjusted hybrid model is a coherent long-run model of the USD/JPY nominal exchange rate. In the fifth chapter we conclude and summarize the findings of the three studies presented in this thesis as well as provide practical recommendations for further study such as construction of dynamic error correction models and assessing out-of-sample forecasting performance for the extended monetary models examined in chapters two and four. III Further development of the study for effectively functioning foreign exchange markets as presented in chapter three is in addition discussed in the final chapter. We contribute to the extant literature by showing in chapter two that the conventional RID monetary model of the exchange rate for the GBP and CAD vis-à-vis the USD can be rejected. A single econometric specification can be adapted to explain the long-run exchange rate for the GBP/USD exchange rate while an extended model is effective in providing an explanation of the long-run CAD/USD exchange rate. In chapter three we demonstrate that the spot and futures markets for five bilateral exchange rates function effectively across developed and developing countries. Lastly, we show in Chapter four that the model of the USD/JPY exchange rate due to Hunter and Ali (2014) appears a specific case and that the USD/JPY is not readily distinguished from a random walk in the context of a monetary model that considers traded and non-traded goods productivity differentials.en_US
dc.publisherBrunel University Londonen_US
dc.subjectReal interest differential monetary modelen_US
dc.subjectMarket for foreign currencyen_US
dc.subjectHarrod-Balassa-Samuelson effecten_US
dc.subjectJohansen (1995) methodologyen_US
dc.subjectWeak exogeneityen_US
dc.titleEssays in exchange rates and international financeen_US
Appears in Collections:Economics and Finance
Dept of Economics and Finance Theses

Files in This Item:
File Description SizeFormat 
FulltextThesis.pdf2.67 MBAdobe PDFView/Open

Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.