Print Reference This Reddit This

Study On The Different Types Of Interest Rates Finance Essay


Interest rates have always been a fascination for scholars. This is even more so when the way the interest rates of countries are linked is mentioned. Over time, this has led to a lot of material on the

A lot of empirical research has been carried out on the Unbiased Forward Rate Hypothesis as researchers feel it is the most important aspect in the foreign exchange market. However, it is the opinion of this researcher that the Uncovered Interest Rate is a distinct form and more important than the unbiasedness of forward rates in explaining the way interest rates are linked among countries.




Interest rates are very important in the economy. It acts a macroeconomic stabilizer, determines asset prices and also attracts capital inflow for various economies. The world is now a global village with a lot of transactions and interdependencies across countries. This makes it imperative to understand how the interest rates are integrated especially during the recession and with the single currency in the European Union.

There are several literatures on this topic. This chapter will attempt to discuss these literatures and identify the shortcomings prevalent in their work. Also, it will show the different empirical evidences that have obtained on this subject.


Interest rates are the percent paid or charged for the use of funds. In economic terms, a particular interest rate can be taken to be representative of all interest rates in an economy. The Treasury Bill rate, base interest rate of retail banks and short term money market rate e.g LIBOR rate could be representative of such a rate, (Howells & Bain, 2007). Sometimes the interest rates on various bonds differ particularly, examples are short term Treasury Bill rates and corporate bonds, this will be explored later. Interest rates are important for several reasons. Interest rates can impact the overall state of an economy because it affects business investment decisions and consumers willingness to spend, (Mishkin & Eakins, 2009).

There is the nominal interest rate and the real interest rate. The nominal interest rate is the amount, in money terms, of any interest quoted. This is the rate usually quoted by financial institutions. While the real interest rate is the purchasing power of the interest receipts that have been adjusted for inflation. It measures the anticipated cost of borrowing in terms of goods and services, (Howells, Bain, 2007). Irving Fisher coined the Fisher Equation which states that the nominal interest rate equals the real interest rate plus the expected rate of inflation (Mishkin & Eakins, 2009). We will concentrate more on the real interest rate.

Interest rates are influenced by several factors. Expected level of inflation plays a big role in determining nominal interest rates. It erodes the real purchasing power of the interest receipts while a deflation will lead to the opposite effect. Monetary policy, levels of sovereign debts outstanding, financial stability, foreign exchange market activity and general economic conditions also influence interest rates.

The relationship (equality) among inflation, interest rates and foreign exchange rates is described in the parity relations. The main assumptions to be kept in mind are that there are minimal or no transaction costs and that there is no restriction on the flow of funds among countries. The four parity relations are: Interest Rate Parity, Purchasing Power Parity, International Fisher Effect and Exchange Rate Expectations. Interest Rate Parity relations are very relevant to this paper.

Interest rate parity relations are of two types; covered and uncovered interest rate parity relations. Covered Interest Parity, CIP, relations shows that the interest rate differentials of any two countries should be equal to the foreign exchange forward premium/discount of the country. This is provided that both home and foreign bonds are free of risk. This implies that one pound invested in the UK will be less than one pound converted into any other currency and invested abroad. This paper focuses on the uncovered interest rate parity relationship.


However, Uncovered Interest Rate, UIP, (open arbitrage condition) tries to show that the difference between the domestic interest rate and the foreign interest rate of countries are equal to the expected depreciation of the home currency. UIP arises when interest rates are viewed as similar to other asset prices. Their movements are interpreted as being determined by financial flows in profit-seeking capital markets. The underlying assumptions particular to UIP are that there is risk neutrality and rationality of expectations of market participants. According to Jong-Cook, UIP shows that any arbitrage among the major world currencies in the form of international capital flows should ensure that the interest differential between any two countries is the unbiased predictor of the future change in the spot exchange rate.

The formula is shown by:


= the spot rate at a point in time.

= the foreign interest rate

= the domestic interest rate

is the one period ahead Spot Rate which is not observable. It is a forecast.

The interest rate parity theory makes a very strong case for the financial assets of any two international countries being substitutes for each other. This is due to there being a high instance of international capital mobility between the two countries; arbitrage would bring the two countries interest rates to be equal as well the forward premium on the two currencies of those countries. The integration of the foreign exchange market is thus very important in establishing this theory.

The Uncovered Interest Rate Parity cannot be directly tested when there is a lack of future exchange rates. Some assumptions would have to be made

The uncovered parity is not directly testable in the absence of market expectations of future exchange rates. Moreover, the above rather simple demonstration assumes no transaction cost, equal default risk over foreign and domestic currency denominated assets, perfect capital flow and no simultaneity induced by monetary authorities. Note also that it is possible to construct the UIP condition in real terms, which is more plausible.


Under given conditions of costless financial and commodity arbitrage, the application of relative Purchasing Power Parity (PPP) in Uncovered Interest Parity (UIP) will result in a condition known as the International Fisher-open Condition. This is when there is equality of interest rates between two nations. UIP indicates that arbitrage arising between countries due to international capital flows should have as an unbiased indicator of any future change in exchange rate, the differential in interest rates between the two countries. This leads to the real interest rates between the countries being almost identical which is known as the Fisher-Open condition (Howells, Bain, 2007).

(Howard, Johnson, 1983) have empirically proven that all these i.e. PPP, Interest rate parity and the International Fisher-Open conditions will not hold simultaneously in a world that includes taxation. Other notable authors have also empirically proved so. (Mishkin, 1984) went even further and empirically proved that that there was a disparity of real Eurorates across countries i.e that they were not equal. However, his test of the Unbiasedness of Forward rate forecasts and relative PPP showed a lot of significance. This made it impossible to rule out the thought that real interest rates of countries tended to be equal over time.

It is important to note that in exchange rate determination, it is very important for real interest rates of the same kind of securities to be equal across countries.


Unbiased Forward Rate Hypothesis (Unbiasedness Hypothesis) has long been considered bedrock of market informational efficiency. This is defined as the extent to which the forward rate tries to accurately predict the spot rate. (Crowther, 1994).

Usually, test rejections on the UFRH have been taken to cast doubt on (contradict) the UIP but this is not always the case. UIP consistently shows that that any interest rate differential can be monitored and managed by monetary authorities. This enables them to stop any rapid change associated with exchange rates. This makes the UIP more important than the UFRH in determining policy response and market informational efficiency.


In creating financial and economic models that work, a working understanding of the relationship among interest rates of various countries is essential. Linkages of interest rates reflect the degree of international capital mobility, which is crucial for international investors in making their portfolio decisions and in pricing interest-rate-sensitive securities (SuZhou, 2003).


The European Monetary System (EMS) is a successful mechanism among European countries to reduce inflation, limit exchange rate volatility and generally be a zone of monetary stability, (Rogoff, 1985; Taylor and Artis,1988).


INTERNATIONAL LINKAGES: The US – European Connection

Robert Cumby & Frederic Mishkin, found that there is a positive relationship between real interest rates in the US and the interest rates in Europe though they do not move exactly in correlation. However, there is observed to be a very strong link thus showing that there may be the possibility of European Monetary policy being influenced by US policies. Using the Fisher model to estimate the movements of the interest rates, it was observed that there was a strong relationship between the US and European interest rates though the direction it moved in were outside the scope of the study. However, there were differences in timing and the extent of the linkages.

Using Granger’s Causality Out-of-sample Model

Granger’s Causality model is a refinement and formalization of previously held theories. It involves evaluating causal relationships between two time series and then predicting if one of the series could be helped by including information from the other time series (Chen, Rangarajan, Feng & Ding, 2004). It looks out for any reduction in variance in the prediction error of one of the time series at a point in time. This reduction is obtained by including past measures from the second time series in the empirical regression model. If this occurs, it means the second time series has a causal relationship with the first time series. Thus, time plays an important role in establishing which direction the causal relationship follows.

Granger’s Causality is actually designed for linear relationships. There are times in which it has to be extended to be useful in predicting causal relationships among multiple non-linear relationships.

In investigating the causal relationship among major currencies, Wang & Li, using Granger’s Causality model, found that there was no dominance of the US interest rates in the Eurocurrency market while there was significant influence of Japanese interest rates. However, the Euro did not have any significant influence on the other interest rates. He found that US did not cause any of the other markets but was caused by the UK and was strengthened by the introduction of the Euro. Interestingly, UK markets rather than the Euro had a greater impact on the other rates. This was an interesting result because it is contrary to conventional studies which had found the US to be a dominant player.

However, the study was done when the Euro had just been introduced into circulation. Also, we need to find if there is a predictive relationship now that the Euro has been around for a longer period. The study did not take into cognizance the macroeconomic and policy coordination problems among EMU countries in its early stages neither did it allow for more rapid information transmission from the US to other markets.


Hutchison and Singh, conducted a study using a vector error correction model and cointegration tests to determine if there is any real interest rate linkage between the US and Japan in an equilibrium setting. They did find a very high degree of linkage between the two. This could indicate that there was a high degree of capital market linkage and might be as a result of the free market policies being implemented by Japan.

Yamada was interested in the long run equilibrium relationship between the real interest rates of Canada and USA and the importance of this relationship is in explaining the movement of the individual real interest rates. He tested this by testing the corresponding linear restriction on the co-integrating vector. He examined the existence of a one on one long term relationship between the interest rates of the countries by testing for stationarity of the real interest rate differential. Also, following a shock in one of the markets, the degree of departure from the long-run equilibrium relationship and the corresponding adjustment between those interest rates were examined.

He found that there actually existed a long run relationship between the real interest rates of the countries due to there being a constant stochastic trend between the two variables. He also found that exogenous shocks to the markets led to departures from the long-run relationship but the degree of departure relative to the shocks were not very large. The departures decayed significantly within a reasonable period of time. This gives empirical evidence that there exists a strong linkage and the markets are closely linked which may explain that the one on one equilibrium relationship may have an important role in explaining the individual interest rates in the countries.

Using a Kalman filter approach to detecting structural change

Barassi, Caporale & Hall, were able to use a Kalman to test the hypotheses of a US worldwide leadership, the disengagement of UK monetary policy from those pursued in the

Eurozone and the German Leadership hypothesis (GLH) within the European Union (EU). Their results were quite interesting. They were able to show that there was a disconnect in the interest rate linkages between UK and the rest of the EU in 1992 and a weak German Leadership within the EU. However, it was consistent in showing that there was a US worldwide leadership in its interest rates.

Previous papers, as we have seen, in analysing interest rate linkages have done so in the context of the Exchange Rate Mechanism (ERM) Framework. The hypothesis of Germany being the dominant force in the EU and the other EU countries having to move in line with the monetary policy set by the German Central Bank (the German Leadership Hypothesis) was tested. Seeing as co-movements in interest rates arose due to policy convergence, under pure arbitrage conditions, expectedly interest rates will be linked in the long run.

It appears that even in a system like the ERM which aims to produce

policy coordination it has been possible for monetary authorities to disengage their

policy from developments elsewhere and pursue an independent policy agenda over long

periods. Such an option should remain available for non-participating countries, like the

UK, after the establishment of the Euro. Therefore the UK authorities will not

necessarily find their freedom of action greatly constrained by what is happening in

the Euro zone. Within the Euro zone the policies of the European Central Bank (ECB)

will not necessarily be as stable or credible as those adopted so far by the German

authorities, since smaller countries will also have an influence on monetary policy

Previous empirical studies on interest rates, such as Caporale et al. (1996), reported

convergence in European rates after 1986. Artis and Zhang (1998), using rolling window

cointegration techniques1 found that there is widespread cointegration between both US

and German short rates and those on other ERM currencies up to 1995, after which the US

influence on world-wide rates vanishes. In the context of the ERM, with its target zones,

there might be regime shifts owing to the policies pursued by central banks. Specifically,

the stochastic properties of interest rates (volatility, level and speed of adjustment) are

likely to be different in periods when the currency has to be defended from speculative

attacks, compared to periods when the exchange rate is credible. Because of the UIP

relation, switches in the process governing exchange rates are translated into switches in

the process followed by interest rates. Such regime shifts tend to be more frequent and not

to be as long-lived as changes in monetary policy regimes in the US, say. Dahlquist and

Gray (2000) show that a Markov-switching model characterises adequately the behaviour

of a number of EMS short rates.

As already mentioned, in general, one can think of changes in structure as changes

either in the long-run relationships themselves (the cointegrating vectors) or in

causality links (the loading factors). It would be problematic to specify the source

of structural change in a model allowing for both types of changes as such a model

would typically not be identified. In the case of interest rates, as almost any theory

suggests long-run co-movement, it is reasonable to assume the cointegrating vectors

are constant but the direction of causality changes. Hence we concentrate on the latter

source of change, and estimate time-varying parameter models for the loading weights.

This has the advantage that one does not have to impose a priori restrictions on when

the breaks in the relationships might have occurred. Instead, the relationships are

allowed to evolve freely, and the revealed timing of the structural breaks can be very

informative about the effects of policy changes (see, e.g. Haldane and Hall, 1991, who

analyse Sterling’s relationship with the US dollar and the Deutschemark). Kalman

filtering techniques were used by Hall et al. (1992), who found convergence in

inflation and interest rates within the EMS.

In this paper, we use a Kalman filter approach to detect changes in the causal

structure of cointegrated models. In particular we apply this procedure to bivariate

systems linking the G-7 short interest rates as irreducible cointegrating relations (IC)

(Davidson, 1998a) in order to investigate the possibility of breaks in the causal structure

of these linkages or reversals in the direction of causality. The empirical findings of this

research will have important policy implications, as they will provide evidence on

whether countries can still conduct an independent monetary policy despite the

increasing integration of international financial markets (see Caporale and Williams,

Another important result is the weak leadership of Germany within the European

Monetary Union (EMU) highlighted by the fact that German rates do not adjust to

disequilibrium in cointegrating relations with Italian and French rates (Table 5). This is

apparent, as German rates do not share any structural relations with the European ones

and receive feedback from Japan (in a structural IC relation) and all the remaining

rates included the UK. As for the UK rates, they seem to receive feedback from US

and Canadian rates but not from the EMU countries. This may be due to the presence

of a break in the causal relationships with the latter rates after the third/fourth quarter

of 1992, and it represents one of the phenomena we want to test for in our exercise.

Essentially, the US and Canada appear to constitute the fundamental block, UK rates

respond to non-European rather than to other European rates, Italy is probably

The empirical results seem to support the existence of US leadership. The speed of

adjustment coefficients converges towards zero in almost all cases. Exogeneity of US

interest rates is clearly observable in the bivariate systems that link its rate to the

Italian and the German ones (Fig. 7). The linkage with the Canadian rate seems to be the most significant one, although even this coefficient is below 0.1 and is falling. The

same (but with even smaller coefficients) can be said about the speed of adjustment to

disequilibrium in the linkage with the French and Japanese rates. The situation is

slightly different in the error correction equation containing the cointegrating relation

that links US to UK rates. It seems that there was some (decreasing) feedback from

UK to US rates until the collapse of the ERM in the last quarter of 1992. Afterwards,

US rates appear to be exogenous and causality runs from US to UK rates only.

Overall we can conclude that the evidence from the time-varying estimation supports

the idea that the US economy is a constant point of reference for all the countries of

the G-7 group.

Jong-Cook Byun, Assistant Professor of Finance, Youngnam University, South Korea; Son-Nan

Chen, Professor of Finance, National Chengchi University, Taipei, Taiwan, and University of

Maryland at College Park.

Global Finance Journal, 7(2): 129-151 Copyright 0 1996 by JAI Press Inc.

ISSN: 1044-0283 All rights of reproduction in any form reserved William Smith



The study uses United States of America (USA), United Kingdom (UK), European Union (EU), Japan and Canada as its case study. This is for several reasons. Firstly, these countries have the highest developed markets in the international financial system. Thus, their indexes are fairly representative of the international system. Secondly, these markets are so developed that to acquire information on these markets through online resources was relatively easy. Also, as has been stated previously, these markets have been fairly consistent.

The basic empirical framework that will be used is one of the vector error correction models, the GARCH Model as compared with the Ordinary Least Squares (OLS) method.



This is a structured approach to time series modelling. Economic theory is used to model the relationship among various variables that are being tested. This particular approach does not capture enough of the dynamic specification to give conclusive information about the relationship. Endogenous variables could be shown on both sides of equations i.e. the left and the right hand sides of the equation thus making whatever estimation or conclusion to be made very difficult. The OLS did not account for the conditional heteroskedasticity in time series data Bollerslev (1986).

These problems lead to an alternative, non-structural approach to time series modelling to capture the relationship among several variables.


The prime motive for using GARCH methods was to capture the ARCH effect i.e. volatility clustering in the market indexes. Risk preferences and falling prices cause informational asymmetry in the market thus giving rise to volatility. GARCH models were later modified in order to capture this asymmetry.

3.2 DATA

The data used in this study covers a 10 year period from June 1, 2000 to May 31, 2010. In selecting this period, we have excluded some events which would have been prejudicial to our study. Such events were the period before the introduction of the Euro in 1999, the German reunification in 1990, the currency crisis in the EMS in 1992, etc. The introduction of the Euro was significant in that it the driving power of Germany as had been postulated cannot be independently verified as the Euro zone now has a single currency.

However, we have included the most significant event and the most recent event which is the global financial recession which began in August, 2008 but has officially been declared over by March 2010. This was adjusted for in the study to show the effect the recession had on the major world financial markets.

The interest rates for the currencies being used in this study were obtained from Datastream. The interest rates were generated from a single market which is the London Eurocurrency market. This is in order to minimise any distortion which may arise due to different tax codes in operation, regulatory interferences, etc. These interest rates obtained from the London Eurocurrency market are recorded simultaneously at the end of a trading day. This preserves the integrity of the data.

In observing the data, it was noted there was very little variation within a week. In keeping with prevalent methodology, all daily data were converted to weekly data. This was to reduce the problem of autocorrelation occurring in the work. From converting the data from daily to weekly observations, we obtained 520 weekly observations.

The data possess the following features over the period of study: (1) prior to the third quarter

of 1999, most Eurocurrency interest rates tended to decline with the exception of the U.S. and

U.K.; (2) all Eurocurrency interest rates have increased from the third quarter of 1999 to the end

of 2000, but they steadily dropped throughout 2002; (3) the Japanese Yen interest rates remained

the lowest over the entire sample period; and (4) before the introduction of the Euro,

the Italian Lira interest rates were the highest for the majority of the time. Several standard

unit root test procedures were applied to test for nonstationarity of the interest rate series in

the two periods. Consistent with previous studies, the results (available on request) show that

a unit root exists in all series in both periods.

Some assumptions were made in obtaining the data analysis. These were;

There are no transaction cost

There is an equal default risk over foreign and domestic currency denominated assets

There is a perfect capital flow

There is no simultaneous result produced by monetary authorities.



(Crowther, 1994)

Need help with your literature review?

Our qualified researchers are here to help. Click on the button below to find out more:

Literature Review Service

Related Content

In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: