**3. Research Methodology**

This chapter explains the procedures used in finding the answer to the research question. The study used quantitative approach with explanatory correlation design. This chapter includes the research population, research design, data collection, research hypothesis, research instruments and data analysis methods used.

## 3.1. Data

An ideal dataset for the analysis will consist of different countries and total equity and debt. The data was thus classified per country to show different geographic locations of entities extending funds to a borrower. Data was collected from the Treasury.gov website, a website consisting of annual securities data. Secondary data was used to measure the level of debts and equity of different countries in the world over the last four years (Anuar & Chin, 2015). The criteria of the analysis for the countries was as follows;

- The countries must have complete equity and debt (short term and long term debts) information for the period of 4 years (2019-2015).
- The countries must have debt financing in their capital structure.
- The countries must have a positive equity since a negative market-to-book would not be beneficial in indicating country growth.

After the outliers were eliminated, a financial sample of 174 countries (n=174) sample size was used.

## 3.2. Research Hypothesis

According to Ajanthan, (2013), capital structure is defined as the total assets to total debt at book value. This study uses the equity value as the dependent variable while debt as independent valuable to test the reliance of a country to debts.

Different countries relies on debts differently compared to the other, which affects increase in national interest interests. According to Huang & Bolton, (2016), the choice of capital structure should be intended to maximise the value of a country. In the context of a country, the level of debt a country should be able to balance the debt to GDP ratio. However, a country must be able to reduce the equity versus debt financing for its investments.

Kagan & Anderson, (2020); Kaminarides & Nissan, (1993) emphasised on the need for this study. The researchers identified debt crisis as the main cause of steep losses for banks both domestic and internationally. This can hit economic growth or turmoil in the international finance markets. A country’s economic slow down may result into an economic slowdown both domestically and sometimes dragging it to other countries. Kagan & Anderson, (2020) identified the role of global financial crisis in 2007 to 2008 in spreading the hurt to all economies in the world. Even though the crisis was majorly participated in Europe, the crisis was also felt in other countries.

International diversification is perfectly correlated to other country’s economic progress. Thus, the hypothesis predicts the relationship between financial leverage and international diversification. Chiang & Chen, (2008) found a negative relationship between country’s capital structure to another.

## 3.3. Variables Measurements

The study investigated two types of variables; dependent variables and independent variables. The variables were categorized into two to estimate their relationships. The debt of a country was used as a measure of how the country depends on debts to finance their expenditures. Since the data provided long term and short-term debts, our study tried to estimate whether different countries were able to balance their capital structures. Country’s equity value was used as the dependent variable while short term and long term debts were used as independent variables. The cross sectional data used in this research was based on Setia, (2016); Echeimberg, Leone, & Zangirolami-Raimundo, (2018).

## 3.4. Research Methodology

This study used panel data analysis since the sample data contained data across various countries over a given period of time (4 years). The data type thus increases the sample size considerably, and therefore making it appropriate to study. This kind of data was therefore useful in studying the dynamics of the subject. To evaluate the effects of independent variables on debt ratio, our research used two estimation methods, the fixed effects model and the ordinary least squares (OLS). Since the panel data had observations on a similar cross-sectional unit over a given period of time, it is likely that there will be a cross-sectional effect on each country. For this kind of a dataset, several techniques are available in handling the fixed effect model problem. The fixed effect model considers each country as an individual of each country or cross-sectional unit included in the sample data. Therefore, our analysis implemented Hausman specific test to estimate the estimation.

In this study, data was analyzed using Stata MP 15 software to clean data and run the regression model. Stata combine technology of the best statistics software for handling data. It is a statistical tool used for handling, analyzing and forecasting. Additionally, the software can be used to estimate and show the level of coefficients and probabilities of the data in a table at the same time. We used Stata to evaluate how the dependent variable is associated with specific regression models. Thus, the regression model was used to find the relationship between country’s equity level to short term and long-term debts over the study period. The regression model was expressed as:

Y= β0+ β1 equity + β2 stDebt + β3 ltDebt + ε

Where y represents the capital structure β1, β2, β3 is the coefficient of the independent variables while ε is the error term of the regression.

## 3.5. Correlation

Since this research intended to find the relationship between international capital structures, correlation method was appropriate. According to Phyllis, (2014), correlation research is a type of nonexperimental research used for measuring statistical relationship between them. Mainlyif two variables shows a high correlation level, it means that they are most likely to affect each other. In such a case, there is no dependent and independ variables.

The degree of association of the two variables will be measured by the correlation coefficient, renoted by r. In this research, Pearson’s correlation method was used. The correlation formula used was as bellow:

The correlation result will be used to show whether the variables had strong, average or weak relationships.

**4. Results and Findings**

## 4.1. Introduction

This chapter presents results and findings of the research. The research attempts to explain the effect of capital structure in the international finance geography over a period of 4 years 2015 to 2019 period. The study used a cross sectional panel data. It pooled ordinary least squares (OLS) regression to estimate the coefficient of the independent variables of long term and short term det. Additionally, the Fixed Effect Model approach was used to evaluate the effect of independent variable on the equity on the basis of international debt. The dependent variable was the country’s equity while the independent variable was the long term and short term debts of the countries.

A descriptive analysis was conducted to understand the data better. Table 1 shows that the max debt for all the countries was 985 in 2019 and 953 in 2018. The countries’ total equity over the two financial years showed an increase 927 to 946 and means of 60.12 and 62. 64 for 2019 and 2018 financial year. The data indicates that on average, the countries used less equity to finance its projects in 2019 compared to the previous year. The long term debt on the other hand declined from 96.66 to 95.30 for year 2018 and 2019 respectively.

Table 1: Descriptive analysis table

Table 1 shows descriptive findings from the variables. The standard deviation shows that the countries’ equity, long term debt and short term debts varies in a great extent. The table also shows that majority of the countries relied on debt finance in 2019 compared to 2018. Similarly, equity financing declined in 2019 compared to 2018.

Correlation

Table 2 presents the correlation metric of the sample countries’ capital structures. It shows a very low correlation between the variables. The result shows very relationship between total and equity for 2019 and 2018. There was a very high relationship between total, long-term, and short-term debts in 2019 and 2018. There was a negative correlation between short-term debt and equity in 2018 (r =-0.0183). There was a negative correlation (r= 0.083) between short-term debt and equity in 2018. The findings thus shows that there is a relationship in debts and equities on countries. It indicates that one country’s capital structure affects another either directly or indirectly.

Table 2: Correlation Statistics Result

Table 3 shows the regression analysis results. The analysis was to find whether the hypothesis should be accepted or rejected. The regression result shows t-statistics, r squared, p value and the f statistics. The significance level was set to be 95%, thus, significance value of 0.05 would indicate that the variables were statistically significant. The result shows showed that the variables were statistically significant. The p value was <0.05 (p= 0.000) for equity, 0.003 for totals in 2018, long-term debt in 2019 and short-term debt. Similarly, there was a statistical significance between the variables. The r squared confirmed that there was a close relationship between the variables (r2 =0.9362).

Table 3: Regression Analysis Result