Better Government Better Life Research Assignment Paper

Better Government Better Life

Yuelin Shi
ECON4850
April 19, 2017

Abstract
This Paper aims at figuring out the relationship between better government and individuals’ levels of good living and happiness. The paper is also aimed at finding how peoples’ lives and happiness can be improved through making better the way the government operates.
The individual happiness evaluation result in 2014 carried on 141 different countries can be used to figure out these countries’ mean value of individual happiness level in current situation. In this paper, I will take government performance in 2014 as an observation.
For government performance, I selected four regressors (by references mentioned in the introduction part): Government Effectiveness, Control of Corruption, Voice and Accountability, and GDP per capita adjusted by PPP (purchase power parity).
Introduction
The paper looks at how individuals’ life pleasure in a country relate to the performance of the government of that economy. This was done by investigating the impact of the value of the government on satisfaction of life in a country. We can find out a way to improve the quality of people’s life by changing or improving government public policies. This is possible if people can get a highly effective government, where their voices are heard and the state is accountable and citizens get a better income, and then their happiness levels can be improved. Changing an individual’s life status is not that easy and calls for an individual’s decision to be self-dependent, however, public policies are more effective to the whole country. The variables used to measure the performance of government are government’s voice, effectiveness, corruption rates, accountability and the GDP of the economy.
A survey to find out what determines satisfaction in life was carried out in the year 2008 by a number of researchers. They observed 90000 features in 70 nations observing the politics, economics, institutions and human development and culture (Bjørnskov, Dreher & Fischer, 2008). The ordered probit method was used to check whether these variables affect satisfaction in life. Robust was also done on the findings using extreme bound analysis method. They claim government effectiveness in micro-econ are more effective to macro-econ. From the conclusion it was noticed that minor factors like openness, business climate, total chambers in government, dominance of Christian and infant mortality have an effect on satisfaction in life (Bjørnskov, Dreher & Fischer, 2008).
Whiteley et al found out the relationship between public policies and satisfaction in life. A survey was done to see if the macro economic variables or government policies affect the microeconomic variables in a country (Whiteley, Clarke, Sanders & Stewart, 2010). Political events and economic changes and how the government managed to tackle them were some of the major variables used to measure government policies. After a linear regression analysis it clearly showed that the policy outcomes determine the performance of the state and in reality the policies formulated affect the satisfaction of life of the individuals in an economy. Effective micro and macro-economic variables in most cases will yield a higher GDP increasing the CPI which enhances satisfaction in people (Whiteley, Clarke, Sanders & Stewart, 2010).
Democracy influences people’s happiness and the end result is satisfaction (Cheung et at). Therefore, government accountability will lead to life satisfaction especially to the less fortunate in the society by helping to curb corruption cases in the economy. In this paper 72 inhabitants were surveyed (Cheung & Leung, 2007). The variables used to measure democracy in this paper include accountability, corruption and inequality. After linear analysis of the variables, it was noticed that tolerance and democracy increase people’s satisfaction in life both in political, economical and on cultural basis. Accountability increases GDP by reducing losses through corruption hence more satisfaction (Cheung & Leung, 2007).
Method
Model Set and Hypothesis Tests
For the Dependent Variable: Life Satisfaction, I take the data from World Happiness Report 2014. Because it is difficult to measure personal happiness, they gave evaluation by using Cantril Ladder (0=worstpossible life and 10=bestpossible life) to let people choose which step of ladder they feel they are standing on currently. The evaluation is based on about 1,000 individual self-assessed in each of 141 countries in 2014.
Regressors data for 2014 available from the World Bank and Kaufmann, Daniel, Aart Kraav and Massimo Mastruzzi (2010): “The Worldwide Governance Indicators: Methodology and Analytical Issues”. Government Effectiveness is the level of public and civil services and degree of government implementation of policies. Control of Corruption represents the degree of control against the use of public power for private gain or private interest of any form. Voice and accountability shows the citizens’ participation in governing a country. Citizens have right to select the governor and freedom to expression and association. The score of Government Effectiveness, Control of Corruption and Voice Accountability both are based on estimating the aggregate indicator by standard normal distribution, whose range is -2.5 to 2.5. As for the GDP, I take various countries from the world as the observation objects. GDP
The following equations represent two models being tested:
Model 1:
LHI=β0+β1GE+β2CC+β3VA+β4GDP+μ
Model 2:
lnLHI=β0+β1GE+β2CC+β3VA+β4VA2+β5GDP+μ
Where LHI is the Life Happiness Index, lnLHI is the natural logarithm of the Life Happiness Index, GE is the Government Effectiveness, CC is the Control of Corruption, VA is the Voice and Accountability, and GDP is GDP per capita adjusted by PPP. By using SAS to analyse data set, we can get Table 1 below.
Table 1: Estimate Model 1

MODEL 1

MODEL 2

GE

0.715***
(0.198)SE

0.165***
(0.039)SE

CC

-0.187
(0.182)SE

-0.085**
(0.038)SE

VA

0.099
(0.111)SE

0.039*
(0.024)SE

VASQ

N/A

0.033**
(0.016)SE

GDP

0.00001***
(0.000005)SE

0.000003***
(0.000001)SE

Number of Observation

141

141

Adjust R2

0.568

0.548

White Test

7.97 (P-value:0.89)

20.13(P-value:0.38)

Condition Number

6.6

7.0

(*) means the p-value is between 10% and 5%. (**) stands for a p-value that is between 5%-1%. (***) represents a p-value that is less than 1%. Standard errors are reported between parentheses. The null hypothesis for the parameters associated with the regressor is that the regressor does not affect obesity. The null hypothesis of the white test is that the error is homoscedastic.
From the Table 1, we can observe that both of these two models do not reject the null hypothesis of the white test, which means there are no heteroscedastic variances and their condition number are all below 20, which indicates there is no severe multicollinearity. However, p-value in Model 1 represents a linear relation, which cannot represent a real relation between dependent variable and regressors. On the other hand, Model2, which represents a non-linear relationship seems to represent the relationship by testing (SE is smaller than in Model 1). By considering all references and the result of test, I chose model 2 as the equation represents the theoretical model.
This model tests my key hypothesis of the relation between individual happiness and government preference. First, government effectiveness affects the public policies formulation and implementation, which has impact on individual happiness; a good policy makes people lives better. The null hypothesis is that β1=0. Secondly, control of corruption makes govern...