3.2.2. DATA COLLECTION TECHNIQUES AND TOOLS
Data collection was done through the secondary Data. Rwanda
Revenue Authority(RRA) provides independent variable data as tax revenue and
National institute of statistics of Rwanda(NISR) provides nominal Gross
Domestic product spanning from 2007Q1 to 2017Q4.
P a g e 27 | 48
3.2.3. DATA PROCESSING
This thesis investigates the impact of tax revenue on economic
growth in Rwanda between 2007Q1 and 2017Q4. Many of economic and social changes
have been taken place in Rwanda during this period. It is very important to
analyze and evaluate the factors that affected the economic growth over this
period. The data for the study have been collected from Rwanda Revenue
Authority(RRA) and National institute of statistics of Rwanda( NISR). To find
out the impact of tax revenue on the GDP and to test the study hypotheses,
regression model were used to test the relation between the dependent and
independent variables. Therefore, multiple regression model were developed to
achieve the above objectives. In our model, GDP is the in dependent variable
while direct tax, tax on good and service and tax on international trade and
transaction are the independent variables. In this study we can analyze the
impact of tax revenue on the economic growth in Rwanda. The study covers Rwanda
Revenue Authority(RRA) and National institute of statistics of Rwanda(NISR)
quarterly data spanning from 2007Q1 to 2017Q4.
In this empirical study, data have been processed and
information related to the hypothesis and objectives of the study was taken
into account and transformed into meaningful data for easy interpretation and
understanding. This has been carried out by the use of e-Views package 8.
3.2.4. DATA ANALYSIS
3.4 TECHNIQUES OF DATA ANALYSIS
In analyzing the data gathered regressions model was employed
to establish the relationship between dependent and independent variables. The
study made use of economic approach in estimating the relationship between tax
revenue and economic growth. Vector error correction model (VECM) was employed
in obtaining the numerical estimates of the co-efficient in different equation.
The ordinary least square method was chosen because it possesses some optimal
properties. Its computational procedure is fairly simple
Page 28 | 48
3.8 LIMITATION
Like any other research, this research was encountered by the
difficulties such as; Inadequate funds to carry out the project, Personal
extremely effort to meet deadline.
3.9 ETHICAL CONSIDERATION
For those who interest to read Bible,Matthew 19:21 we see tax
money having its functions to perform in the society which enables government
authorities to use in providing social services that will be enjoyed by all the
citizens of a country. Therefore, we must have attitude and ethics to pay to
tax on time in order to build our nation.
3.10 MODEL SPECIFICATIONS
The method employed in this study, involves discussion of data
collection analysis techniques. We adopted a quasi-experimental research which
is purely analytical. In this study we used quarterly data covering the period
from 2007 to 2017, from the Rwanda Revenue Authority (RRA) statistical bulletin
and annual reports and National institute of statistics of Rwanda(NISR). The
economic growth variable is nominal GDP at current basic prices. The study uses
three independent variables: direct tax including tax on property, taxes on
goods and services, taxes on international trade and transactions including
others tax.
Authors such as Osoro (1993), Kusi (1998), Muriithi &Moyi
(2003) and Bilquess (2004) used the following models to estimate buoyancy and
elasticity:
Tt= aYt f3 e e t
(1)
The logarithm transformation of the equation (1) give
LnTtt = Lna+ f3LnYt + Et
(2)
Or Tax t = u0 + f3yt + et (3)
Where
Tax t = log (Tt) and Yt = log (Yt) e:
Stochastic disturbance term
P a g e 29 | 48
In this this thesis, I am going to use the following models to
estimate long run tax revenue and
economic development in Rwanda: LnGDP = É(LnDT, LnTGS,
lnTITT,) (4)
The estimable econometric model is shown in equation as
LnGDP = á + f31LnLDT + f32LnLTGS +f33LnLTITT + åt ..
(5) Where
LnGDP = Natural logarithm of Gross Domestic Product
LnDT = Natural logarithm of Direct tax
LnTGS = Natural logarithm of Tax on Goods and Services
LnTITT = Natural logarithm of Taxes on International Trade
Transactions
Dummy Variable for Direct taxes (DM_DT)
Dummy Variable for Taxes on goods and services (DM_TGS)
Dummy Variable for Tax on international trade and transactions
DM_TITT) f31, f32, f33 is regression parameters
åt is : stochastic error
Therefore,In order to capture short run dynamics, we estimate
Vector Error Correction Models (VECM)
3.11 DEFINITION OF VARIABLES AND THEIR EXPECTED
SIGNS
Definition of variables used, their estimation coefficients
and expected signs of each
explanatory variable.
Figure 2.definition of variable and expected
sign
|
Variable
|
Definition
|
Estimation coefficient
|
Expected Sign
|
|
GDPt
|
Gross domestic product can be defined as the
total monetary value of all finished goods and services produced within a
country's borders in a specified time period
|
This is the dependent variable
|
|
|
DTt
|
Direct tax is set of Pay As You Earn (PAYE),
Taxes on Corporations & Enterprises and Tax on property (Property tax on
Vehicles)
.
|
f31
|
Positive
|
|
TGSt
|
Taxes on goods and services is the set of
Value Added Tax (VAT), Excise Duty, Road Fund, Mining Royalties, Strategic
reserves levy)
|
B2
|
Positive
|
|
TITTt
|
Taxes on international trade transactions are indirect taxes
and include custom duties and other taxes on
international trade and transactions. These are
imposed by
the government on trade transactions involving exchange of goods and services
between home country and foreign countries.
|
f33
|
Negative
|
Source: Author's computations
P a g e 30 | 48
P a g e 31 | 48
CHAPTER- 4 DATA ANALYSIS AND INTERPRETATION
| 
