ANALYSIS OF THE RETURN ON EQUITY (ROE) FOR … · 2019. 11. 13. · financial performance is represented by Return on Equity (ROE) which is the ratio between Net-Income to shareholder’s

ANALYSIS OF THE RETURN ON EQUITY (ROE) FOR

TELECOMMUNICATION INDUSTRY USING DUPONT ANALYSIS

David Winster Praveenraj* 1, S.Gokul Kumar 2 and V. S. Karthic3

1 Faculty – School of Management Studies,Bannari Amman Institute of Technology,

India 2 Faculty – School of Management Studies,Bannari Amman Institute of Technology,

India 3 School of Management Studies, Bannari Amman Institute of Technology

India 1 [email protected], 2 [email protected], [email protected]

Abstract

The net income of the organization is not enough to determine its efficiency and performance

unless profit margin, asset turnover, financial leverage, etc. are taken into consideration.

examines That is why this study has used the quantitative finance tool: Dupont Analysis to

ascertain whether Indian Tele-communication companies are able to generate positive Return on

Equity for its shareholders by analyzing various trends of the factors influencing Return on

Equity. The research employs a Descriptive research design for this report. The study

implements the five factor DuPont Model to disaggregate Return on Equity into five factors viz.

Financial Leverage, Asset Turnover, Operating Profit Margin, Tax Burden and Interest Burden.

Using these results, the trend in the industry over the past twelve years is analyzed and the effect

of recession is studied using time series analysis. Cross sectional analysis is performed to

understand significant focus points of different companies and is benchmarked with the industry

average. Also, regression analysis is performed to comprehend the relative impact of the five

ratios on Return on Equity of the selected companies. The key finding of the research study

indicates that there is an overall slump in Return on Equity of the industry attributed to various

internal and external factors of the business. Based on the findings, research study suggests

healthy ways to improve Return on Equity, through optimal usage of Financial Leverage and by

increasing Asset Turnover and Operating Profit Margin efficiencies.

Keywords: Return on Equity, Asset Turnover, Operating Margin, Tax Margin, Tax Burden,

Time-Series analysis, Regression analysis, DUPONT analysis

Tierärztliche Praxis

ISSN: 0303-6286

Vol 39, Issue 11, November - 2019

106

1. Introduction Concurring with the decisions of an Organization’s financial management depends on

the question of whether the decision improves the value of the stakeholder’s equity or not.

Basically, every financial, investment and operational decisions impacting productivity, product

costs, tax limits etc. directly affect the income generated for the investors. Also, any decision

which impacts the type of capital employed (viz. debt or equity or both) affects the capital

structure of the firm.

The financial statements, namely, Balance sheet and Income Statement of an

organization contains significant and extremely crucial information about financial health of the

organization. In order to ascertain the financial health of the organization, the stake holders

employ a variety of techniques and financial ratios viz. profitability ratios, turnover ratios and

performance ratios. One such ratio is the Return On Equity (ROE), which indicates the net

income per equity of shareholder’s funds invested in the organization.\

The DuPont model has become one of the most crucial tools of financial analysis for

assessment of the returns and is widely applied across industries to thoroughly analyze the

important factors influencing Return On Equity (ROE). Thus, in this research study, the

researcher employs the extended DuPont model also known as five factor DuPont model to

disaggregate ROE into a product of various components viz. Financial Leverage (FL), Asset

Turnover (ATO), Operating Profit Margin (OPM), Tax Burden (TB) and Interest Burden (IB).

This division of ROE into various components attaches more meaning to ROE and gives various

insights about different drivers influencing ROE of an organization.

Two major telecom companies are taken into consideration for analysis. A brief about

them are presented. Bharti Airtel is India based Tele-com firm operating globally in 20 countries.

The main services provided include GSM, 3G, 4G LTE mobile services and has also diversified

into a variety of allied activities like Airtel DTH, Airtel Broad Band services, Airtel Payment

Banking services etc. In India it has a total market capitalization of about 1,23,879.4 Cr. Rs.

making it the largest private Tele-com firm in India.

Vodafone India merged with Idea cellular to form Vodafone Idea, which is the second

largest Tele-com firm in India with a total market capitalization of about 27,080 Cr. Rs.

Vodafone Idea also offers similar services to Airtel like GSM, 3G, 4G, payment banking etc.

2. Objectives of the paper To apply quantitative mathematical models to assess the financial performance of the

organizations. Perform Time Series analysis of individual firms by comparing ROE and its

factors within the company.

To conduct scatterplot and correlation analysis between ROE and its factors of the

individual firms to establish relationship between ROE and its factors.

To perform correlation analysis among FL, ATO, OPM, TB and IB for individual firms

to verify the existence of multi-collinearity among these independent variables.

To frame regression equation for ROE, based on factors of ROE to establish strength.

To test whether Indian Tele-com companies are able to generate positive ROE for the

shareholders based on the collected secondary financial data.

To suggest improvements in the ROE of Tele-com companies in India using DuPont

model.

3. Literature review Thalassinos and Curtis, (2005), identified that case of original model of DuPont analysis,

financial performance is represented by Return on Equity (ROE) which is the ratio between Net-

Income to shareholder’s Equity. This can also be expressed as product of Return on assets (ROA)

and Financial Leverage (FL). Thalassinos and Liapis, (2013) , has stated that Return on Asset

(ROA) can be further decomposed into product of Profit Margin (PM) and Asset Turnover

(ATO), where profit margin is the ratio between net-income to sales and asset turnover is ratio


ISSN: 0303-6286


107

between sales to total average assets for the period. PM represents firm’s profitability relating to

revenue (which depends on sales quantity and profit margin per unit) and ATO measures how

efficiently and effectively the company’s assets are utilized in generating profits through sales.

Study by P.Bauman, Mark, (2014) adds to the literature by providing guidance for

improving forecasts of company’s performance. This study employs DuPont analysis, which is a

tool used for decomposing a company’s return on net operating assets (RNOA) into asset

turnover (ATO) and profit margin (PM) which gives a valuable understanding about factors

driving operating profitability.

Work by Gardner, John.C; Mc Gowan, Carl.B; Moeller, Susan.E, (2011) , is a case study

on Coca-Cola Corporation. The main purpose of this literature is to educate students on how to

estimate a firm’s Sustainable Growth by implementation of DuPont system of financial analysis

with Coca-Cola corporation. The term sustainable growth in finance refers to highest sales

growth that a firm can achieve by means of using only its internally generated funds or retained

earnings (RE).

Paper by Loukopoulos, George; Roupas, Theodoros (2014) , aims at performing a

comprehensive financial statement analysis by using DuPont based decomposition scheme of

Nissim and Penman, (2001) . The DuPont decomposition scheme reveals a quick snapshot of

various factors influencing ROE. The descriptive analysis shows how capital structure decisions

(usage of financial leverage) of Hygeia eroded shareholder’s profits.

Key points from Literature Review

The comprehensive study of the Review of Literature provides valuable insights about

the variety of applications of DuPont model viz. establishment of inter relationship between ATO

and PM on RNOA, in ascertaining the sustainable growth, in creating well-organized common

investment indicator, in forecasting future profitability conditions, in development of software

tool to educate investment managers, to understand leverage effect and capital structure decisions

etc.

The literature study showed, a descriptive research methodology being adopted in the

research articles that are reviewed. Literature Review also indicates, the studies on the DuPont

analysis has been extensive and hence there is only a little room for research gap. However, in

context of Indian scenario, research gap exists, since DuPont analysis had not yet been performed

in the Tele-com sector of India.

Thus, from the valuable insights gained from the above Review of Literature, the

researcher now proceeds with Descriptive Research methodology and analysis will be performed

using Five Factor DuPont Model to ascertain the various factors influencing ROE of the selected

Tele-communication firms. A time series analysis for past decade (intra firm analysis) and cross-

sectional analysis (inter firm analysis) of the factors influencing ROE will be performed after

DuPont decomposition of ROE. Additionally, a regression analysis will be carried out between

ROE and its factors to understand the strength with which these factors influence ROE, for each

firm. By understanding their strength, companies can decide upon what factor to improve upon to

contribute for healthy improvement of ROE.

4. Methodology This study uses descriptive research design. The research study here employs the

secondary financial data viz. Income statements and balance sheets of the selected Tele-com

companies between 2007 and 2018. These secondary data are the data originally collected by the

prior researcher which was then utilized by some other researcher for their study. The reliability

of these secondary depends on the purpose and the accuracy with which it was collected by the

prior researcher for the original study. The major sources from which secondary financial data

collected for the current research include various reliable sources of financial statements such

like www.screener.in, www.moneycontrol.com and Company’s official websites etc. Due to the

secondary nature of the data used in this research study, sampling design involving population,

sample size, sampling frame and sampling techniques are irrelevant in this research

methodology.


ISSN: 0303-6286


108

The primary tools used in the study include the Five Factor DuPont model of ROE

disaggregation. The secondary tools used include Time Series and Cross Sectional analyses using

line graph, Correlation analysis and Multiple Linear Regression analysis. All these analyses are

carried out using Microsoft Excel tool.

The DuPont Model Formulae [1 Factor]

ROE = (Net Income/Equity)

[3 Factor]

ROE = (NI/Sales) x (Sales/Total Assets) x (Total Assets / Equity)

[Return on Sales] x [ Asset Turnover] x [Financial Leverage]

[5 Factor]

ROE = (NI/EBT) x (EBT/EBIT) x (EBIT/Sales) x (Sales/Total Assets) x (Total Assets/

Equity)

[Tax Burden] x [Interest Burden] x [Operating Margin] x [Asset Turnover] x [Leverage]

Where,

Tax burden is the proportion of profits retained after paying taxes.

Interest burden is the proportion of profits retained after paying interest

obligations.

Operating profit margin is the operating income per dollar of sales.

Asset turnover shows asset utilization efficiency.

Financial leverage indicates proportions of debt and equity employed in capital

structure.

Regression Analysis Equation

Multiple Linear Regression analysis is carried out to understand the strength of impact

contributed by the independent variables (IV) on the dependent variable (DV). The

general form of multiple linear regression is given by the following equation.

Y = β0 + β1X1 + β2X2 +…+ βnXn + ε

Where,

Y = Dependent Variable

X1, X2…Xn = Independent Variables

β1, β 2… β n = Coefficients of Independent Variables

β0 = Y intercept (Constant) and

ε = Standard Error (avg. distance of scatter plots from regression line)

5. Analysis and Interpretation The analysis adopts the following structure. First, the financial data from the balance

sheets and income statements of the respective companies are consolidated into input tables.

Also, in order to make a meaningful comparison between income statement and balance sheet,

average value (current and previous year) is taken for balance sheet values to honor the matching

concept.

Secondly, a DuPont framework is modeled to ascertain ROE, through disaggregated

components of ROE viz. FL, ATO, OPM, TB and IB. Then, using the disaggregated values of

ROEs each firm, a time series analysis is performed to compare ROE and its factors within each

company between 2007 to 2018.

Also, a cross sectional analysis is performed by comparing ROE and its factors, between

two companies with their industrial benchmark values to infer the relative performance of the

companies during this period. These analyses will help to understand the impacts of various

internal and external factors influencing ROE of the companies.

The analysis finally ends with the Regression analysis of ROE, by taking ROE as a

dependent variable and its factors as independent variables. This analysis is performed to

understand the strength of influence of these individual factors on ROE of each company

between 2007-2018.


ISSN: 0303-6286


109

Table 5.1 Consolidated Financial Data – Private Telecom Firms in India - 2007-2018

AIRTEL 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

AVERAGE EQUITY (EQ) 11456 16589 25423 34501 44321 49688 50467 55039 49766 53299 67158 68546

AVERAGE TOTAL ASSETS (TA) 28754 38309 55814 67418 108857 151972 158279 171201 183738 207070 235024 249841

SALES (S) 18420 27012 37352 41829 59602 71506 76947 85864 96101 96532 95468 83688

OPERATING PROFIT (EBIT) 7264 11031 14973 16469 20164 23705 23258 26936 32749 32205 34150 29626

EARNINGS BEFORE TAX (EBT) 4678 7312 8591 10895 7678 6518 4785 7864 10540 12846 7723 3267

NET INCOME (NI) 4062 6395 7859 9163 5752 4257 2258 2773 4621 6077 3800 1099

VODAFONE IDEA 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018



SALES (S) 3927 6720 10131 12398 15438 19489 22407 26519 31571 35949 35576 28279


EARNINGS BEFORE TAX (EBT) 509 1115 939 1075 997 1055 1577 3044 4933 4250 -863 -6499

NET INCOME (NI) 502 1042 882 954 899 723 1011 1968 3193 2728 -400 -4168

RELIANCE COMM. 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018



SALES (S) 17190 18827 22251 21496 22431 19677 20561 21238 24787 25594 6554 4593

OPERATING PROFIT (EBIT) 6334 7827 8722 7147 8411 5785 4439 5674 6225 6265 -383 -1825


NET INCOME (NI) 3531 5401 6045 4655 1345 928 672 1047 714 639 -1403 -23839

TATA COMM. 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018



SALES (S) 8611 8297 9963 11026 11932 14196 17213 17714 19913 18149 17620 16651


EARNINGS BEFORE TAX (EBT) 281 149 423 -681 -707 -718 -431 444 373 243 1467 46

NET INCOME (NI) 15 10 316 -598 -777 -795 -623 101 1 9 1233 -329

NETTLINX 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

AVERAGE EQUITY (EQ) 17.5 18.3 19.0 18.8 18.0 17.3 17.6 17.7 18.0 18.0 18.3 18.9

AVERAGE TOTAL ASSETS (TA) 22.2 29.4 35.0 32.0 29.7 30.0 31.2 29.8 28.4 30.3 36.1 47.7

SALES (S) 10.4 11.4 15.82 12.95 10.52 8.71 11.46 8.63 10.89 21.39 19 15.72

OPERATING PROFIT (EBIT) 2.05 2.57 1.54 1.11 -1.72 -0.96 1.03 -2.09 -0.44 1.99 3.72 2.53

EARNINGS BEFORE TAX (EBT) 6.37 0.66 0.23 -0.72 -2.37 -0.21 0.67 -0.21 -0.17 1.66 3.36 1.91

NET INCOME (NI) 5.4 0.33 0.11 -0.63 -2.33 -0.28 0.58 -0.19 -0.16 1.66 2.34 -6.45

QUADRANT TELE. 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

AVERAGE EQUITY (EQ) 591 591 591 634 757 837 837 837 561.5 286 173.5 61

AVERAGE TOTAL ASSETS (TA) 826 884.5 855.5 773.5 809.5 803.5 731.5 799 627.5 615 557.5 404

SALES (S) 276 249 224 196 236 281 330 406 521 558 339 344

OPERATING PROFIT (EBIT) 37 8 -104 -16 -75 -32 12 -110 -78 27 -26 -28

EARNINGS BEFORE TAX (EBT) -116 -142 -214 -21 -224 -179 -136 -261 -240 -135 385 -301

NET INCOME (NI) -116 -143 -215 -21 -224 -179 -136 -261 -240 -135 385 -301


ISSN: 0303-6286


110

AIRTEL 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018



SALES (S) 18420 27012 37352 41829 59602 71506 76947 85864 96101 96532 95468 83688


EARNINGS BEFORE TAX (EBT) 4678 7312 8591 10895 7678 6518 4785 7864 10540 12846 7723 3267

NET INCOME (NI) 4062 6395 7859 9163 5752 4257 2258 2773 4621 6077 3800 1099

VODAFONE IDEA 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018



SALES (S) 3927 6720 10131 12398 15438 19489 22407 26519 31571 35949 35576 28279



NET INCOME (NI) 502 1042 882 954 899 723 1011 1968 3193 2728 -400 -4168

RELIANCE COMM. 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018



SALES (S) 17190 18827 22251 21496 22431 19677 20561 21238 24787 25594 6554 4593

OPERATING PROFIT (EBIT) 6334 7827 8722 7147 8411 5785 4439 5674 6225 6265 -383 -1825


NET INCOME (NI) 3531 5401 6045 4655 1345 928 672 1047 714 639 -1403 -23839

TATA COMM. 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018



SALES (S) 8611 8297 9963 11026 11932 14196 17213 17714 19913 18149 17620 16651


EARNINGS BEFORE TAX (EBT) 281 149 423 -681 -707 -718 -431 444 373 243 1467 46

NET INCOME (NI) 15 10 316 -598 -777 -795 -623 101 1 9 1233 -329

NETTLINX 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

AVERAGE EQUITY (EQ) 17.5 18.3 19.0 18.8 18.0 17.3 17.6 17.7 18.0 18.0 18.3 18.9

AVERAGE TOTAL ASSETS (TA) 22.2 29.4 35.0 32.0 29.7 30.0 31.2 29.8 28.4 30.3 36.1 47.7

SALES (S) 10.4 11.4 15.82 12.95 10.52 8.71 11.46 8.63 10.89 21.39 19 15.72

OPERATING PROFIT (EBIT) 2.05 2.57 1.54 1.11 -1.72 -0.96 1.03 -2.09 -0.44 1.99 3.72 2.53

EARNINGS BEFORE TAX (EBT) 6.37 0.66 0.23 -0.72 -2.37 -0.21 0.67 -0.21 -0.17 1.66 3.36 1.91

NET INCOME (NI) 5.4 0.33 0.11 -0.63 -2.33 -0.28 0.58 -0.19 -0.16 1.66 2.34 -6.45

QUADRANT TELE. 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

AVERAGE EQUITY (EQ) 591 591 591 634 757 837 837 837 561.5 286 173.5 61

AVERAGE TOTAL ASSETS (TA) 826 884.5 855.5 773.5 809.5 803.5 731.5 799 627.5 615 557.5 404

SALES (S) 276 249 224 196 236 281 330 406 521 558 339 344

OPERATING PROFIT (EBIT) 37 8 -104 -16 -75 -32 12 -110 -78 27 -26 -28

EARNINGS BEFORE TAX (EBT) -116 -142 -214 -21 -224 -179 -136 -261 -240 -135 385 -301

NET INCOME (NI) -116 -143 -215 -21 -224 -179 -136 -261 -240 -135 385 -301


ISSN: 0303-6286


111

Table 5.2 Five Factor DuPont Model for Disaggregation of ROE

Table 5.2 shows the framework of five factor DuPont Model which is created to

ascertain ROE. It involves disaggregating ROE into various components, known as factors. The

original DuPont equation has only 3 factors namely Profit Margin (PM), Asset Turnover (ATO)

and Financial Leverage (FL). In this model an extended DuPont Model is used which has 5

factors in total, obtained by further drilling down PM into 3 factors namely Operating Profit

Margin (OPM), Tax Burden (TB) and Interest Burden (IB). Using this model ROE values are

disaggregated into respective components to attach more meaning to ROE and a time series

analysis and cross sectional analysis will be performed based on the results from the model.

In the time series analysis, the consolidated financial data gathered is fed into the model

to ascertain the factors contributing to ROE. The results for each company is tabulated for period

between 2007 to 2018. Based on the table values, line graphs are plotted in excel to analyze the

trend of individual factors influencing ROE of the firm. Inferences are derived on the basis of

correlating the trend with the internal and external factors of the business scenario.

The following tables show the financial leverage, asset turnover, operating profit margin, tax

burden and interest burden affecting ROE of Bharti Airtel between 2007 to 2018, both per year

value and average value, calculated using excel sheet separately.

Table 5.3 ROE and its Factors for each year - Airtel - 2007-2018

Five Factor DuPont Model for ROE Formulae

ROE

FL =TA/EQ

ATO = S/TA

ROA X

FL OPM = EBIT/S

TB = NI/EBT

IB = EBT/EBIT

ATO X

PM

Legend

ROE - RETURN ON EQUITY

ROA - RETURN ON ASSET

PM - PROFIT MARGIN

ATO - ASSET TURN OVER

OPM X

TB X

IB FL - FINANCIAL LEVERAGE

OPM - OPERATING PROFIT MARGIN

TB - TAX BURDEN

IB - INTEREST BURDEN

Factor 3

Factor 1

Factor 2

Factor 4

Factor 5

AIRTEL 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

ROE 35% 39% 31% 27% 13% 9% 4% 5% 9% 11% 6% 2%

FL 251% 231% 220% 195% 246% 306% 314% 311% 369% 389% 350% 364%

ATO 64% 71% 67% 62% 55% 47% 49% 50% 52% 47% 41% 33%

OPM 39% 41% 40% 39% 34% 33% 30% 31% 34% 33% 36% 35%

TB 87% 87% 91% 84% 75% 65% 47% 35% 44% 47% 49% 34%

IB 64% 66% 57% 66% 38% 27% 21% 29% 32% 40% 23% 11%


ISSN: 0303-6286


112

Table 5.4 ROE and its Factors for 12 years - Airtel - 2007-2018

Avg. ROE Airtel (2007-2018) 11%

Avg. FL Airtel (2007-2018) 315%

Avg. ATO Airtel (2007-2018) 48%

Avg. OPM Airtel (2007-2018) 34%

Avg. TB Airtel (2007-2018) 63%

Avg. IB Airtel (2007-2018) 34%

The following tables show the financial leverage, asset turnover, operating profit margin,

tax burden and interest burden affecting ROE of Vodafone Idea between 2007 to 2018, both per

year value and average value, calculated using excel separately.

Table 5.5 ROE and its Factors for each year - Vodafone - 2007-2018

Table 5.6 ROE and its Factors for 12 years - Vodafone - 2007-2018

Avg. ROE Vodafone (2007-2018) 5%

Avg. FL Vodafone (2007-2018) 301%

Avg. ATO Vodafone (2007-2018) 48%

Avg. OPM Vodafone (2007-2018) 29%

Avg. TB Vodafone (2007-2018) 77%

Avg. IB Vodafone (2007-2018) 17%

Using the above data Line graphs are plotted with various combinations of ROE and its

Factors to gain insights from the trend formed by them. The graphs are analyzed and results are

interpreted.

Figure 5.1 Line Graph – Cross Section Analysis – ROE Trend - Airtel and

Vodafone - 2007-2018

VODAFONE 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018

ROE 19% 36% 11% 8% 8% 6% 7% 13% 16% 12% -2% -16%

FL 334% 378% 236% 206% 232% 254% 259% 272% 272% 306% 378% 396%

ATO 45% 62% 51% 49% 56% 60% 63% 63% 59% 50% 39% 27%

OPM 38% 34% 28% 26% 24% 26% 27% 31% 34% 32% 29% 22%

TB 99% 93% 94% 89% 90% 69% 64% 65% 65% 64% 46% 64%

IB 34% 49% 33% 33% 27% 21% 26% 37% 46% 36% -8% -107%

19%

25%20%

15%

7% 5% 3%

5%

8% 8%3%

-25%

35%39%

31%

2…

13%

9%

4%

5%

9% 11% 6%2%

19%

36%

11%8%

8%6%

7%

13%16%

12%

-2%

-16%

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

2 0 0 7 2 0 0 8 2 0 0 9 2 0 1 0 2 0 1 1 2 0 1 2 2 0 1 3 2 0 1 4 2 0 1 5 2 0 1 6 2 0 1 7 2 0 1 8

ROE (Industry) ROE (Airtel) ROE (Vodafone)


ISSN: 0303-6286


113

Table 5.7 Input Data - Regression Analysis of ROE – Airtel – 2007-2018

BHARTI AIRTEL LTD. ROE FL ATO OPM TB IB

2007 35% 251% 64% 39% 87% 64%

2008 39% 231% 71% 41% 87% 66%

2009 31% 220% 67% 40% 91% 57%

2010 27% 195% 62% 39% 84% 66%

2011 13% 246% 55% 34% 75% 38%

2012 9% 306% 47% 33% 65% 27%

2013 4% 314% 49% 30% 47% 21%

2014 5% 311% 50% 31% 35% 29%

2015 9% 369% 52% 34% 44% 32%

2016 11% 389% 47% 33% 47% 40%

2017 6% 350% 41% 36% 49% 23%

2018 2% 364% 33% 35% 34% 11%

Figure 5.2 Evaluation of Correlation btw. DV & IVs - Scatter Plot – Airtel

From the above figures it is clear there exists a strong positive correlation of ROE with

ATO, OPM, TB and IB while FL is negatively correlated. Thus, there is a non-existence of any

non-contributing independent variable in the ROE regression equation for Bharti Airtel.

Table 5.8 Evaluation of Correlation among IVs - Correlation Analysis – Airtel

FL ATO

FL OPM

FL TB

FL 100%

FL 100%

FL 100%

ATO -82% 100%

OPM -65% 100%

TB -88% 100%

FL IB

ATO OPM

ATO TB

FL 100%

ATO 100%

ATO 100%

0%

500%

0% 500% 1000% 1500%

FL VS ROE

R = -0.769, P = .003

0%

100%

0% 10% 20% 30% 40% 50%

ROE VS ATO

R = 0.9193, P = .000

0%

50%

0% 10% 20% 30% 40% 50%

ROE VS OPM

R = 0.874 , P = .000

0%

50%

100%

0% 10% 20% 30% 40% 50%

ROE VS TB

R =0.901 , P = 0.000

0%

50%

100%

0% 10% 20% 30% 40% 50%

ROE VS IB

R = 0.957 , P =0.000


ISSN: 0303-6286


114

IB -78% 100%

OPM 68% 100%

TB 86% 100%

ATO IB

OPM TB

OPM IB

ATO 100%

OPM 100%

OPM 100%

IB 93% 100%

TB 78% 100%

IB 80% 100%

TB IB

TB 100%

IB 87% 100%

The above table reveals the fact that the IVs influencing ROE of Bharti Airtel are in

inter-related, and thus these factors may be redundant variables in the regression equation for

ROE. There exists a strong negative correlation between FL with ATO, OPM and IB, while ATO

has positive relation with OPM, TB and IB. The OPM bears positive relation with TB and IB,

also TB is also strongly proportional to IB. This complex interaction among these 5 IVs makes a

complex or dynamic impact on the DV (ROE).

Hypothesis Testing

Null Hypothesis H0: All the IVs don’t have significant influence on ROE.

i.e., H0: β1 = β2 = β3 = β4 = β5 = 0

Alternate Hypothesis H1: At least one IV have a significant influence on ROE.

i.e., H1: At least one βi ≠ 0, where i = 1 to 5

The significance level chosen for the analysis is, α = 5%.

Table 5.9 Regression and Collinearity Analyses of ROE and its Factors – Airtel

Regression Statistics

VIF > 5 => High Multi-Collinearity

Multiple R 0.9879

FL 3.25 Accept

R Square 0.9759

ATO 6.24 Reject

Adjusted R

Square 0.9557

OPM 2.25 Accept

Standard

Error 0.0278

TB 5.52 Reject

Observations 12.0000

IB 7.41 Reject

ANOVA

df SS MS F

Significa

nce F

Regression 5.0000 0.1872

0.037

4

48.51

11 0.0001

Residual 6.0000 0.0046

0.000

8

Total 11.0000 0.1919

Coefficie

nts

Standard

Error t Stat

P-

value

Lower

95%

Upper

95%

Lower

95.0%

Upper

95.0%

Intercept -0.7961 0.2200

-

3.618

6

0.011

1 -1.3344 -0.2578 -1.3344 -0.2578

FL 0.0347 0.0290

1.196

3

0.276

7 -0.0362 0.1055 -0.0362 0.1055

ATO 0.4953 0.2373

2.087

5

0.081

9 -0.0853 1.0758 -0.0853 1.0758

OPM 1.2406 0.4423

2.805

1

0.031

0 0.1584 2.3227 0.1584 2.3227


ISSN: 0303-6286


115

TB 0.1279 0.1129

1.132

4

0.300

7 -0.1485 0.4042 -0.1485 0.4042

IB 0.1730 0.1476

1.172

5

0.285

4 -0.1881 0.5342 -0.1881 0.5342

From the above table, the value of Correlation Coefficient (R) is 0.9879, which

shows high positive correlation between DV and all the IVs. R2 indicates the value of

Coefficient of Determination. In case of Multiple Linear Regression, we consider only

adjusted R2 value. Here, adjusted R2 value is 0.9557, which means that the IVs (viz. FL,

ATO, OPM, TB and IB) accounts for 95.57% of variability in the DV, i.e., ROE. In the

ANOVA table F-Significance value is 0.0001 which is much less than α = 0.05, hence it

affirms with 99.99% confidence that the overall model is statistically significant. In the

last section of the table, P -Value indicates probability value of an IV, which when less

than α = 0.05 then it confirms the significant effect of that IV on the DV. In other words,

if P-value for at least one of the IV is less than α = 0.05, then the Null Hypothesis, H0,

will be rejected. From the above table, P-value of OPM is 0.0310, which is less than α =

0.05, hence Alternate Hypothesis H1 is accepted, i.e., at least one IV (OPM) have a

significant influence on ROE. Also, the effects of FL, ATO, TB and IB doesn’t appear to

be statistically significant for the chosen level of significance. However, based on the

VIFs ascertained for each of the 5 IVs separately, the results indicate only FL and OPM

have VIF < 5, indicating very low collinearity. Thus, in order to have a best regression

model we re-run the entire regression considering only OPM and FL as IVs, although P-

value of FL is statistically insignificant in the above regression. This will eliminate the

multi-collinearity effects of ATO, TB and IB on OPM and FL. The Regression Model is

re-run and the results are tabulated in the following table.

Table 5.10 Input Data - Regression Analysis of ROE – Vodafone – 2007-2018

VODAFONE IDEA LTD. ROE FL ATO OPM TB IB

2007 19% 334% 45% 38% 99% 34%

2008 36% 378% 62% 34% 93% 49%

2009 11% 236% 51% 28% 94% 33%

2010 8% 206% 49% 26% 89% 33%

2011 8% 232% 56% 24% 90% 27%

2012 6% 254% 60% 26% 69% 21%

2013 7% 259% 63% 27% 64% 26%

2014 13% 272% 63% 31% 65% 37%

2015 16% 272% 59% 34% 65% 46%

2016 12% 306% 50% 32% 64% 36%

2017 -2% 378% 39% 29% 46% -8%

2018 -16% 396% 27% 22% 64% -107%


ISSN: 0303-6286


116

Figure 5.3 Evaluation of Correlation between. DV & IVs - Scatter Plot – Vodafone

From the scatter plot, it is understood that ATO, OPM and IB have a positive correlation

with ROE, while TB has a medium correlation but its value is statistically insignificant. Though

FL has a weak negative correlation with ROE, even this value is statistically insignificant. Thus

FL and TB are non-contributing IVs in the regression equation of ROE for Vodafone.

Table 5.11 Evaluation of Correlation among IVs - Correlation Analysis – Vodafone

FL ATO

FL OPM

FL TB

FL 100%

FL 100%

FL 100%

ATO -53% 100%

OPM 20% 100%

TB -28% 100%

FL IB

ATO OPM

ATO TB

FL 100%

ATO 100%

ATO 100%

IB -53% 100%

OPM 30% 100%

TB 16% 100%

ATO IB

OPM TB

OPM IB

ATO 100%

OPM 100%

OPM 100%

IB 80% 100%

TB 20% 100%

IB 62% 100%

TB IB

TB 100%

IB 37% 100%

From the above table it is clear that there exists a mild correlation among the IVs. This

correlation leads to redundancy among the IVs influencing ROE. There exists a weak positive

correlation between FL and OPM, ATO and OPM, ATO and TB, TB and IB & OPM and TB. A

strong positive correlation between ATO and IB. A medium correlation between OPM and IB. A

medium negative correlation between FL and ATO & FL and IB. Weak negative correlation

between FL and TB. This multi-collinearity among IVs cause complex or dynamic influence of

IVs on the DV (ROE) for Vodafone.

Hypothesis Testing

Null Hypothesis H0: All the IVs don’t have significant influence on ROE.

i.e., H0: β1 = β2 = β3 = β4 = β5 = 0

0%

500%

-20% 0% 20% 40%

ROE VS FL

R = -0.073 , P = 0.82

0%

100%

-20% 0% 20% 40%

ROE VS ATO

R = 0.668 , P = 0.0174

0%

50%

-20% 0% 20% 40%

ROE VS OPM

R = 0.753 , P = 0.004

0%

200%

-20% 0% 20% 40%

ROE VS TB

R = 0.531 , P = 0.075

-200%

0%

200%

-20% 0% 20% 40%

ROE VS IB

R = 0.82 , P = 0.001


ISSN: 0303-6286


117

Alternate Hypothesis H1: At least one IV have a significant influence on ROE.

i.e., H1: At least one βi ≠ 0, where i = 1 to 5

The significance level chosen for the analysis is, α = 5%.

Table 5.12 Regression and Collinearity Analyses of ROE and its Factors – Vodafone

Regression Statistics

VIF > 5 => High Multi-

Collinearity

Multiple R 0.9879

FL 2.32 Accept

R Square 0.9760

ATO 2.45 Accept

Adjusted R

Square 0.9560

OPM 3.16 Accept

Standard

Error 0.0260

TB 0.82 Accept

Observations 12.0000

IB 7.29 Reject

ANOVA

df SS MS F

Significan

ce F

Regression 5 0.16436

0.032

87

48.759

74 0.00009

Residual 6 0.00405

0.000

67

Total 11 0.16841

Coefficie

nts

Standard

Error t Stat

P-

value

Lower

95%

Upper

95%

Lower

95.0%

Upper

95.0%

Intercept -0.7569 0.1301

-

5.817

7 0.0011 -1.0753 -0.4386 -1.0753 -0.4386

FL 0.1008 0.0234

4.300

2 0.0051 0.0434 0.1582 0.0434 0.1582

ATO 0.4493 0.1402

3.204

4 0.0185 0.1062 0.7924 0.1062 0.7924

OPM 0.2885 0.3705

0.778

7 0.4658 -0.6181 1.1951 -0.6181 1.1951

TB 0.2778 0.0527

5.266

8 0.0019 0.1487 0.4068 0.1487 0.4068

IB 0.1661 0.0628

2.645

9 0.0382 0.0125 0.3197 0.0125 0.3197

From the above table, the value of Correlation Coefficient (R) is 0.9879, which shows

high positive correlation between DV and all the IVs. Adjusted R2 value is 0.9560, which means

that the IVs (viz. FL, ATO, OPM, TB and IB) accounts for 95.6% of variability in the DV, i.e.,

ROE. In the ANOVA table F-Significance value is 0.0009 which is much less than α = 0.05,

hence, it affirms with 99.99% confidence that the overall model is statistically significant. In the

last section of the table, P -value indicates probability value of an IV, which when less than α =

0.05, it confirms the significant effect of that IV on the DV. In other words, if P-value for at least

one of the IV is less than α = 0.05, then the Null Hypothesis, H0, will be rejected. From the

above table, P-values of FL = 0.005, ATO = 0.018, TB = 0.0019 and IB = 0.0125 which are less

than α = 0.05, Hence Alternate Hypothesis H1 is accepted as 4 out of 5 IVs have a significant

effect on ROE. Also, the effect of OPM doesn’t appear to be statistically significant for the

chosen level of significance. However, based on the VIFs ascertained for each of the 5 IVs

separately, indicates only IB have VIF > 5, indicating high collinearity. Thus, in spite IB’s

significant P-value, we should eliminate IB from further regression analyses due to high multi-

collinearity and although OPM has insignificant P-value we include OPM in the next regression

analysis due to its low multi-collinearity.


ISSN: 0303-6286


118

6. Discussion of findings Based on the time series, cross sectional and regression analyses conducted on ROE and its

factors (decomposed using DuPont model) for Bharti Airtel and Vodafone, a detailed summary

findings are made as follows.

6.1 Findings of Time Series analysis of ROE - Airtel

A downward trend in the ROE from 35% to 2% between 2007- 2018, which

synchronizes with Economic depression, 2G scam’s effect, Reliance Jio’s market

competition and demonetization etc.

It is observed that major portion of capital structure of the firm is financed through debts

or leveraged, increasing from 251% to 364% between 2007-2018, which is observed to

be the major cause for increasing interest payment obligations of the firm.

There is a sort of dynamic inter-relationship that exist among the factors influencing

ROE, i.e., existence of multi-collinearity.

Falling trend in asset usage efficiency is observed from 64% to 33% between 2007-2018.

The operating profit margin of the firm remained constant with minor fluctuations with

an average value of 34% between 2007-2018 which is well above industrial average for

the period.

The tax payment burden between 2007-2018 increased from 13% to 66%, which might

be attributed due to higher profits from operating and non-operating sources and it is

observed that the firm has no control over tax burden to increase ROE, as tax burden is

statutory in nature.

It is found that interest payment commitment of the firm increased from 36% to 89%

between 2007-2018 which might be due to high leveraging effect.

6.2 Findings of Time Series analysis of ROE for Vodafone Idea

A downward trend in the ROE from 19% to -16% between 2007- 2018, which

synchronizes with Economic depression, 2G scam’s effect, Reliance Jio’s market

competition and demonetization and Vodafone Idea merger etc.

It is observed that major portion of capital structure of the firm is financed through debts

or leveraged, increasing from 334% to 396% between 2007-2018, which is observed to

be the major cause for increasing interest payment obligations of the firm.

There is sort of dynamic inter-relationship that exist among the factors influencing ROE,

i.e., existence of multi-collinearity.

Falling trend in asset usage efficiency is observed from 45% to 27% between 2007-2018,

which might be due to poor utilization of acquired assets from Vodafone-Idea merger.

The operating profit margin of the firm fluctuates sinusoidal between 2007-2018 with an

average value of 29% which is slightly below the industry average.

The tax payment burden between 2007-2018 increased from 1% to 36%, which might be

attributed due to higher profits from operating and non-operating sources and it is

observed that the firm has no control over tax burden to increase ROE, as tax burden is

statutory in nature.

It is found that interest payment commitment of the firm increased from 64% to 207%

between 2007-2018 which might be due to high leveraging effect.

6.3 Findings of Cross sectional analysis of ROE

The average ROE for past 12 years shows that Airtel having highest ROE at 11%

followed by the industry with 6% and Vodafone with 5%. This general slump in ROE

may be attributed due to numerous internal factors (firm related decisions viz. ATO

efficiency, Operating efficiency, Capital structure decisions etc.) and external factors

(macro-economic factors viz. Fiscal and Monetary policies, political influences, market

competition etc.).

The average FL for past 12 years shows that Airtel having highest FL at 315% followed

by the industry with 310% and Vodafone with 301%. This increasing FL by Airtel,


ISSN: 0303-6286


119

Vodafone and the industry synchronizes with the 2G scam’s effect that forced the

telecom operators to borrow money.

The average ATO for past 12 years shows that Airtel and Vodafone having similar ATO

efficiency and much higher than average ATO of the industry with 41%.

The average OPM for past 12 years shows that Airtel having highest OPM at 34%

followed by the industry with 30% and Vodafone with 29%.

The average TB for past 12 years shows that Industry having highest burden of tax at

39% followed by the Airtel with 37% and Vodafone with 23%. Tax burden is statutory

in nature and the firms have no control over it to improve their ROE.

Finally, the average IB for past 12 years shows that Vodafone having highest interest

payment obligations at 83% followed by the industry with 75% and Airtel with 66%.

6.4 Findings of Regression analysis of ROE - Airtel

Correlation analysis between ROE and its factors showed FL negatively correlated to

ROE with r = -0.769, while ATO, OPM, TB and IB have positive correlation with ROE

with r being 0.919, 0.874, 0.901 and 0.957 respectively.

Correlation analysis among factors influencing ROE shows

o FL and IB with r = -0.78

o ATO and OPM with r = 0.68

o ATO and TB with r = 0.86

o ATO and IB with r = 0.93

o OPM and TB with r = 0.78

o OPM and IB with r = 0.8 and

o TB and IB with r = 0.87,

which means these IVs are multi-collinear and acts as redundant variable to

one another in ROE regression equation. But multi-collinearity issue is not

addressed (mentioned in the limitations of the study) while framing

regression equation because, the original five factor DuPont model suggests

to factor in all the above variables as they have high degree of correlation

with ROE.

Final regression analysis of ROE for Airtel shows only OPM has a statistically

significant impact on ROE with coefficient value of 3.24, which means for every

percentage of OPM increase, ROE increases 3.24 %.

6.5 Findings of Regression analysis of ROE - Vodafone

Correlation analysis between ROE and its factors showed FL negatively correlated to

ROE with r = -0.073, while ATO, OPM, TB and IB have positive correlation with ROE

with r being 0.017, 0.753, 0.531 and 0.820 respectively.

Correlation analysis among factors influencing ROE shows

o FL and IB with r = -0.53

o ATO and OPM with r = 0.20

o ATO and TB with r = -0.28

o ATO and IB with r = -0.53

o OPM and TB with r = 0.20

o OPM and IB with r = 0.62 and

o TB and IB with r = 0.37,

which means these IVs are multi-collinear and acts as redundant variable to

one another in ROE regression equation. But multi-collinearity issue is not

addressed (mentioned in the limitations of the study) while framing

regression equation because, the original five factor DuPont model suggests

to factor in all the above variables as they have high degree of correlation

with ROE.


ISSN: 0303-6286


120

Final regression analysis of ROE for Airtel shows that except for IB, all other factors viz.

FL, ATO, TB and OPM to have a statistically significant impact on ROE with coefficient

values being 0.058, 0.717, 0.315 and 1.109 respectively. It means that for every unit

increase in values of FL, ATO, TB and OPM, ROE increases by 0.058, 0.717, 0.315 and

1.109 times respectively.

References

[1] Thalassinos, E.J., Curtis, G.P., (2005), “Equity Fund Raising and "Creative

Accounting Practices: Indication from Athens Stock Exchange for the 1999-

2000”, European Research Studies Journal, 3(1-2): 127-135.

[2] Thalassinos, I.E. and Liapis, K. (2013), “A Comparative Analysis for the

Accounting Reporting of“Employee benefits”between IFRS and other Accounting

Standards: A Case Study for the Biggest Listed Entities in Greece”,American

Journal of Economics and Business Administration, 5(4): 153-167.

[3] P. Bauman, Mark (2014). Forecasting operating profitability with DuPont

analysis: Further evidence, 13: 2, 191-205, Retrieved from ProQuest Database.

[4] Gardner, John C; McGowan, Carl B; Moeller, Susan E (2011). Using

Accounting Information for Financial Planning and Forecasting: An Application

of the Sustainable Growth Model Using Coca-Cola, 7: 5, 9-15, Retrieved from

ProQuest Database.

[5] Loukopoulos, George; Roupas, Theodoros (2014). Financial Analysis of the

Greek Private Health Sector over the Last Decade (2002-2012), 17: 2, 3-19,

Retrieved from ProQuest Database.


ISSN: 0303-6286


121

ANALYSIS OF THE RETURN ON EQUITY (ROE) FOR … · 2019. 11. 13. · financial performance is represented by Return on Equity (ROE) which is the ratio between Net-Income to shareholder’s

Documents