This chapter keeps the following structure: section 4.2 put forward the research questions, 4.3 present the data sources and hypotheses of the study and section 4.4 relates to the methodology employed explaining the econometric model to be used. While the following chapter concentrates on the empirical findings.

4.1 Introduction

The literature and empirical findings on debt repayment capacity suggested a large emporium of macroeconomic variables which determines the servicing capacity of a country’s external debt and thereby affecting the economic performance of a country. The study here employs the most relevant variables determining the servicing capacity of debt for a developing state economy Mauritius.

4.2 Research Questions

The curbing effect of international trade post the US Subprime crisis on debt repayment capacity is a matter of concern for an economy. The total public debt of the country being around 57.3 % [1] of GDP in 2011 and external debt represent one third of this percentage. Alongside, it is vital to analyze how the different macroeconomic variables following the economic/financial crisis responding to repayment capacity of the economy. In this optic, a country level analysis through a time series econometric analysis is undertaken for the period 1990 to 2011.

The following sets of hypotheses19 are mentioned with their explanation below:

4.3 Data Sources

The study makes use of time series data from 1990 to 2011, thus making use of 21 data points just enough for effective regression analysis. The data used for this study has been taken from the World Bank.

.

4.4 Methodology

Time Series Models describes the historical patterns of data is popular forecasting methods and have often been found to be competitive relative to economic system of equations (particularly in their multivariate forms). These are the work-horse of the forecasting industry. Although time series data are used heavily in econometric studies, they present special problems for econometricians. One common problem is that of serial correlation and another one is that the underlying time series should be stationary. If that is not the case, we encounter the problem of what is known as spurious or nonsense regression (Granger and Newbold (1974)).

The term "spurious regression’ is used to describe regression results, involving time series, that look good (the t-values suggest that there is a significant relationship between the two variables) when in fact the truth is that there can be no significant relationship between the two variables. To avoid the spurious regression problem that arises from regressing a non-stationary on one or more non stationary time series, we have to transform the non stationary times series into a stationary times series. Because if a time series is not stationary, we can study behavior only for the period under consideration. As a consequence, it is not possible to generalize it to other time periods.

Therefore for the purpose of forecasting, non stationary time series may be of little practical value. Stationarity has always played a major role in time series analysis. To perform forecasting, most techniques required stationarity conditions. Loosely speaking, a time series is stationary if it’s mean and variance do not vary systematically over time. One advantage is that Time Series Data are very easily available.

The Econometric Approach

The Economic Model-adapted from Ramlall (2011)

Following the literature review, we cannot conclude which variable mostly affect Debt repayment positively or negatively. The central objective is to determine and analyze the most important factors that affect debt repayment in Mauritius. In this view, the following model is to be used:

The following model is set up according to Ramlall (2011). Debt repayment capacity is influenced by the macroeconomics factor specified in the model below.

DSCR = f (IT, FDI, EXTDEBT, RES, INV)

Where,

DSCR is Debt Service Coverage Ratio

IT stands for International Trade

FDI stands for Foreign Direct Investment

EXTDEBT stands for External Debt

RES stand for Reserves

INV stands for Investment

For analytical purposes and to conduct various tests, the function has been transformed into an econometric model which can be described as follows:

DSCRt = AŽA²0 + AŽA²1ITt + AŽA²2FDIt + AŽA²3ExternalDebtt + AŽA²4RESt + AŽA²5INVt + AŽAµt

Where the unknown parameter "AŽA²" captures the effects of each variable, AŽA²0 is a constant also known as the intercept term and AŽAµt, the random error term. The "t" in subscript denotes the use of time series data.

Where the dependent variable is rate of interest (ROI) and the explanatory variables are the constant (AŽA±), the treasury bills rate (AŽA²1), usa interest rate (AŽA²2), inflation rate (AŽA²3) and money supply (AŽA²4).

The model above, measures the percentage change of interest rate (Y variable) given a percentage change in the macroeconomic factors(X variables). The slope coefficients on the other hand assess the elasticity of rate of interest in respect with the other macroeconomic determinants.

A time series analysis will be performed for the period 1990 to 2010. A set series represent a set of observations on the values which a variable takes at different times. Such data are collected at regular intervals for example daily (such as stock prices, weather reports), weekly (such as money supply), monthly (for e.g. unemployment rates, consumer price index), quarterly (e.g. GDP) and annually (such as government budgets).

Although time series data are heavily used in many econometric studies, they pose many problems to econometricians. Such can take the form of serial correlation (occurs in time series when the errors associated with a given time period carry over into future time periods) or a problem known as spurious or nonsense regression (Granger and Newbold 1974). The term "spurious regression" describes regression results which involve time series that look good (the t values suggest significant relationship between the variables) when in fact there can be no significant association between the two variables.

A time series therefore is stationary when the mean and variance do not vary systematically overtime.

4.2 The Econometric Approach

4.2.1 The Economic Model-adapted from Maysami, Howe and Hamzah (2004)

Stock Market indices are influenced by all the macroeconomic factors specified in the econometric model below.

LnS = f {INF, RINT, GRW, LnMS, EXR} Equation (5)

where

LnS: Ln SEMDEX, INF: Inflation Rate, RINT: Real Interest Rate, GRW: Growth Rate, LnMS: Ln Money Supply, EXR: Exchange Rate

4.2.2 Variable Description and Justification

The following criteria have been used as determinants for the selection of variables: existing literature on the topic, economic theory, availability of data and whether they fit well in the model in statistical terms. Before conducting the regression analysis, the study give a concise description of their being in the model.

LnS is the SEMDEX (Ln is applied). It is the one of the main indices that tracks the evolution of prices of securities listed on the official market. This variable captures the overall performance of the market and it is the dependent variable in our regression analysis.

INF; inflation rate is generally described as a rise in the general price level. Inflation rate is one of the most important national economic statistics. Increase in inflation rates increases the cost of living and causes a transfer of resources from stock market instruments to consumables. This therefore leads to a decline in the demand for market instruments which tends to reduce the volume of trading and thus value of traded stocks with no price increase. Inflation is widely used in many such studies. For example, Maysami, and Koh (2000) make use of this variable in order find a relationship between the Singapore all index and the macroeconomic variables. In the study a negative relationship was hypothesized, however they concluded a significant positive relationship between the stock prices and inflation.

RINT; Real Interest Rate, is the lending interest rate adjusted for inflation as measured by the GDP deflator.

GRW; represents growth rate of the economy. It is a measure of the value of economic activity within a system.

LnMS is the money supply-M2, as defined by the International Financial Statistics comprise the sum of currency outside banks, demand deposits other than those of the central government, and the time, savings, and foreign currency deposits of resident sectors other than the central government. Changes in money supply may affect share prices in 2 ways. First, changes in money supply may be related to unanticipated increases in inflation and future inflation uncertainty and hence negatively influence the share price. Secondly, changes in money supply may positively influence the share price through its impact on economic activity.

EXR is the exchange rate of the Mauritian rupee vis a vis the US dollar. Exchange rate is an important variable for as a currency depreciates rapidly, capital flight affects the exchange’s trading volume and market index as funds are redirected outside.

4.2.3 Model Specification

To be able to carry out the empirical analysis of the link between the selected macroeconomic indicators and the performance of the stock market index, we develop the following model:

Equation (6)

where

St- the stock market index (SEMDEX), INF is the inflation rate, RINT-Real Interest Rate, LnMS-Money Supply, EXR-Exchange Rate, AŽA²’s are the coefficients of the variables and Aµt is the error term.

Under this chapter, the different hypotheses and variables used are put forward in section 4.2 and 4.3. Section 4.4 presents the data sources, limitations and methodology utilized. The theory of gravity model is explained and the models for our hypotheses are specified in section 4.5. Moreover, a priori report of expected results is set forward in section 4.6. Ultimately, section 4.7 bestows explanation on choice of estimation method.

4.2 METHODOLOGY

In order to estimate the impact of financial infrastructure undertaken in Sub-Saharan Africa on financial sector development and growth, the panel fixed effect model and the random effect model are used respectively. An augmented model has been invoked based on the different components regarding the definition of "Financial Infrastructure" and other financial variables. Firstly, a general equation is estimated. In such an equation, the various variables are likely to interact with each other. In order to choose the most significant variables in explaining financial sector development and growth, a reduced form regression is obtained with the elimination of "Domestic credit to Private Sector as % of GDP". Thus, this implies that the A¢â‚¬Å¾from general-to-specificA¢â‚¬Å¸ method is used.

On the other hand, in order to test the different hypotheses, the A¢â‚¬Å¾general-to-specificA¢â‚¬Å¸ approach had to be used so as to remove the insignificant variables and it is noteworthy that the variables which need to be eliminated differ from regression to regression.

The result will enable in validating one of these hypothesis:

H0: Null hypothesis – the values of each macroeconomic factor is not related to the debt repayment capacity of Mauritius.

H1: Alternate hypothesis – the values of each macroeconomic factor is related at any point to the debt repayment capacity of Mauritius

In the previous chapters, the different views regarding how exchange rates and stock market are related have been mentioned. In this chapter, the data and the methodology used for this study will be explained. The chapter is structured as follows: section 4.1 presents the data which will be used for the time series analysis and their sources; section 4.2 gives a data analysis and finally in section 4.3 the methodology to be used will be explained.

5.1 Model specification

The following model is set up according to Lucio Sarno et al., Shmuel Kandel et al. 2001 Interest rate is modeled in the following specified factors below:

Rate of interest = f (treasury bills rate, Money Supply, Inflation, USA interest rate)

For analytical purposes and to conduct various tests, we have transformed the function into an econometric model which can be described as follows:

ROIt = AŽA± + AŽA²1Treasury billst + AŽA²2Usa interest ratest + AŽA²3inflation ratest + AŽA²4lnMSSt + ut

Where the dependent variable is rate of interest (ROI) and the explanatory variables are the constant (AŽA±), the treasury bills rate (AŽA²1), usa interest rate (AŽA²2), inflation rate (AŽA²3) and money supply (AŽA²4).

The model above, measures the percentage change of interest rate (Y variable) given a percentage change in the macroeconomic factors(X variables). The slope coefficients on the other hand assess the elasticity of rate of interest in respect with the other macroeconomic determinants.