chapter - vi determinants of human...
TRANSCRIPT
- 90 -
CHAPTER - VI
DETERMINANTS OF HUMAN DEVELOPMENT
The objective of this chapter is to analyze the contribution of different factors affecting levels
of human development. The multiple regression analysis technique has been applied to find whether
the effect of different factors differ significantly between the countries. The analysis has been done
for 79 countries for which the data were available. Further countries have been divided into four
groups based on World Bank classification. Group-I contains nineteen high income countries.
Group-II contains twenty high middle income countries. Group-III has twenty countries with low
middle income and Group-IV contains twenty countries with very low income. The analysis has
been carried out at three points of time viz for the years 1985, 1995 and 2005. The purpose behind
this was to study the changes over a long period of time. The chapter has been divided into four
sections. Section-I deals with the analysis of determinants of human development in respect of all
countries. Section-II deals with the analysis of determinants of human development for different
group of countries based on World Bank classification. In section-III analysis of the determinants of
human development in the developing countries have been taken up. The whole discussion has been
concluded in Section-IV.
SECTION - I
Determinants of Human Development:
In this section analysis has been done to study the contribution of different factors affecting
different levels of human development. Human Development Index (HDI) has been used as a proxy
for human development. The analysis has been undertaken to find the relationship of HDI with per
capita income (PPP), gross domestic saving, health expenditure in terms of GDP percentage, and
education expenditure in terms of GDP percentage, labour force as a percent of total population,
urban population as a percent of total population, share of non agriculture sector as a percent of GDP
and percentage of labour force in non agriculture sector.
All Countries Analysis: 1985
The results of the cross country regression analysis for all countries taken together for the
year 1985 are presented in table 6.1. The table shows that 82 percent variations in the dependent
variable are explained by the eight factors under study. The coefficient of per capita income and
percentage of labour force in non agriculture sector are positive and significant at 1 percent level of
- 91 -
significance. The effect of gross domestic saving and health expenditure are positive and significant
at 10 percent level of significance. The effect of coefficient of urban population is negative and
significant at ten percent level of significance.
Table - 6.1
Cross Country Regression Results for the year 1985: All Country Analysis
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .118 3.655*
Gross Domestic Saving .011 1.712***
Health Expenditure as a percent of GDP .050 1.507***
Education Expenditure as a percent of GDP .006 .175
Labour force as a percent of total population .005 .085
Urban population as a percent of total population -.159 -1.573***
Share of Non Agriculture sector as a percent of
GDP
.421 1.164
Percentage of Labour Force in Non Agriculture
sector
.515 4.161*
Constant -2.921 -5.336
R2 .838
2
R .820
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
This can be attributed to the facts that increase in urbanization increases the cost of
sustenance thereby reducing the capability to deploy suitable funds for education and health.
Migration of people from rural to urban areas results in slums which further reduces the human
development level. Gross domestic saving has direct effect on human development and its value of
coefficient is .011 indicating that if there is one percent increase in gross domestic saving the human
development will increase by .011 percent. This is due to the fact that increase in saving will
increase the income of the inhabitants through increased investments which may help the population
to have the capacity to acquire the facilities which are adding to the development of human resource
- 92 -
and thus increase in human development level. The coefficient of share of non agriculture sector as
a percent of GDP turns out to be positive but insignificant.
All Country Analysis: 1995
The cross country regression results for all countries taken together for the year 1995 are
presented in the table 6.2. The table reveals that 86.5 percent variations are explained by the
explanatory variables under consideration. The coefficient of per capita income, share of non
agriculture sector as a percent of GDP and percentage of labour force in non agriculture sector are
positive and significant at 1 percent level of significance. The coefficient of health expenditure in
terms of GDP percentage is positive whereas that of urban population as a percent of total population
is negative and both are significant at ten percent level of significance. The coefficient of education
expenditure as percent of GDP is negative but non-significant. The coefficient of health expenditure
is .05 indicating that if there is one percent increase in the health expenditure, the human
development will increase by .05 percent. The coefficient of share of non agriculture sector as a
percent of GDP and labour force in non agriculture sector is positive and highly significant meaning
thereby that increase in these factors will result into increase in income and hence better opportunity
for education and better health facilities thus raising the human development level.
Table - 6.2 Cross Country Regression Results for the year 1995: All Country Analysis
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .113 4.009*
Gross Domestic Saving .004 .805
Health Expenditure as a percent of GDP .050 1.397***
Education Expenditure as a percent of GDP -.003 -.087
Labour force as a percent of total population -.019 -.224
Urban population as a percent of total population -.154 -1.427***
Share of Non Agriculture sector as a percent of
GDP
1.103 3.206*
Percentage of Labour Force in Non Agriculture
sector
.362 3.779*
Constant -3.840 -6.068
R2 .879
2
R .865
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
- 93 -
All Countries Analysis: 2005
The cross country regression results for all countries for the year 2005 are presented in the
table 6.3. It is clear from the table that 91.7 percent variations are explained by eight explanatory
variables under study. The coefficients of per capita income (PPP), share of non agriculture sector
as a percent of GDP and percentage of labour force in non-agriculture sectors are substantially
positive at one percent level of significance. The coefficient of labour force as a percent of total
population is positive and that of urban population as a percent of total population is negative and
both are significant at 10 percent level of significance. The coefficient of remaining variables like
gross domestic saving, health expenditure, education expenditure are positive but non significant.
The study found that the value of R2 and 2
R which was 83.8% and 82% respectively in the
year 1985 has increased to 87.9% and 86.5 percent respectively in 1995. These further increased to
92.6 and 91.7 percent in 2005. This indicates that the effect of the explanatory variables on the
dependent variable have increased over the period of time. The per capita income continues to be
positive and highly significant throughout the study period. The coefficient of gross domestic saving
was significant in the year 1985 but it became insignificant in the year 1995 and 2005 though
remained positive.
The coefficient of health expenditure has remained significant upto the year 1995 and in the
2005 it became insignificant though remained positive. The coefficient of urban population as a
percent of total population has also remained negative and significant throughout the study period.
The coefficient of share of non agriculture sector as a percent of GDP was non significant in the year
1985 but it became significant in the year 1995 as well as 2005.
- 94 -
Table - 6.3
Cross Country Regression Results for the year 2005: All Country Analysis
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .122 7.411*
Gross Domestic Saving .004 .207
Health Expenditure as a percent of GDP .028 1.003
Education Expenditure as a percent of GDP .016 .689
Labour force as a percent of total population .066 1.889**
Urban population as a percent of total population -.229 -2.881**
Share of Non Agriculture sector as a percent of
GDP
.668 3.095*
Percentage of Labour Force in Non Agriculture
sector
.466 5.445*
Constant -3.452 -9.574
R2 .926
2
R .917
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
SECTION - II
In this section analysis of the determinants of human development for different groups of
countries based on World Bank classification has been undertaken.
- 95 -
High Income Countries: 1985
The results of cross country regression analysis for Group-I i.e. High Income countries for
the year 1985 are presented in the table 6.4. The table reveals that 81 percent variations in the
dependent variable are explained by eight explanatory variables under consideration. The
coefficient of education expenditure is positive and significant at one percent level of significance.
This implies that as the expenditure on education is increased more trained and skilled workers are
produced who will work and earn better and their standard of living will increase which ultimately
lead to enhancement in the human development level.
Table - 6.4
Cross Country Regression Results for the year 1985: Group- I High Income Counties
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .035 1.310
Gross Domestic Saving .016 .817
Health Expenditure as a percent of GDP .074 .469
Education Expenditure as a percent of GDP .022 1.564*
Labour force as a percent of total population .010 .638
Urban population as a percent of total population -.019 -.413
Share of Non Agriculture sector as a percent of
GDP
-2.257 -.243
Percentage of Labour Force in Non Agriculture
sector
.700 -1.948**
Constant 2.415 3.13
R2 .894
2
R .810
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
- 96 -
The coefficient of labour force is negative and significant at five percent level of
significance. The coefficient of urban population is also negative but is non significant. The
coefficients of per capita income, gross domestic saving, health expenditure, labour force are
positive but non significant.
High Income Countries: 1995
The results of the regression analysis for the group-I for the year 1995 are presented in the
table 6.5. Table shows that 87.6 percent variations in the dependent variable are explained by the
explanatory variables under study. The coefficient of per capita income is positive and significant at
1 percent level of significance. The coefficient of gross domestic saving and share of non
agriculture sector as a percent of GDP are negative but insignificant. The coefficients of health
expenditure, education expenditure, labour force, urban population, labour force in non agricultural
sector are positive but non significant.
Table - 6.5
Cross Country Regression Results for the year 1995: Group I High Income Countries
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .134 3.467*
Gross Domestic Saving -.011 -.280
Health Expenditure as a percent of GDP .027 1.020
Education Expenditure as a percent of GDP .005 .204
Labour force as a percent of total population .041 .737
Urban population as a percent of total population .040 .427
Share of Non Agriculture sector as a percent of GDP -1.892 -1.353
Percentage of Labour Force in Non Agriculture sector .235 1.096
Constant 1.602 .615
R2 .931
2
R .876
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
- 97 -
High Income Countries: 2005
The results of the cross country regression analysis of group-I countries for the year 2005 are
presented in the table 6.6. Table shows that 91 percent variations in the dependent variable are
explained by set of 8 explanatory variables under study. The coefficients of Per Capita Income
(PPP) and health expenditure are positive and significant at one percent level of significance. The
coefficient of labour force as a percent of total population and urban population as a percent of total
population is positive and significant at 5 percent level of significance. The coefficient of gross
domestic saving, share of non agriculture sector as a percent of GDP is positive but non significant.
The coefficient of education expenditure as a percent of GDP and percentage of labour force in non
agriculture sector are negative and insignificant.
The study found that over the period from 1985 to 1995 the value of 2
R increased from 81
percent to 87.6 percent and further it increased to 91 percent in 2005. The coefficient of per capita
income was positively insignificant in the year 1985 but it became highly significant in 1995 and
2005. The education expenditure was positive and significant in the year 1985 but it became
insignificant but positive in 1995 and in 2005 its coefficient become negative but non significant.
The coefficient of urban population was negative in the year 1985 but it became positive in the year
1995 and 2005. The share of non agriculture sector as a percent of GDP was negative in the year
1985 and 1995 but became positive in the year 2005. The coefficient of health expenditure was
positive but insignificant in the year 1985 and 1995 but became significant in 2005.
- 98 -
Table - 6.6
Cross Country Regression Results for the year 2005: Group I High Income Countries
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .101 7.398*
Gross Domestic Saving .004 .205
Health Expenditure as a percent of GDP .072 3.592*
Education Expenditure as a percent of GDP -.014 -.962
Labour force as a percent of total population .028 2.474**
Urban population as a percent of total population .227 2.041**
Share of Non Agriculture sector as a percent of GDP .115 .264
Percentage of Labour Force in Non Agriculture sector -.138 -.874
Constant -1.744 -1.906
R2 .951
2
R .911
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
Upper Middle Income Countries: 1985
The result of cross country regression analysis for Upper Middle Income countries (Group-II)
for the year 1985 are presented in the table 6.7. The results of the regression analysis show that only
39 percent variations in the dependent variable are explained by the factors under consideration. The
coefficients of gross domestic saving, labour force as a percent of total population and labour force
in non agriculture sector are positive and that of education expenditure is negative and all are
significant at 10 percent level of significance.
- 99 -
Table - 6.7
Cross Country Regression Results for the year 1985: Group II Upper Middle Income
Countries
HDI Dependent Variable` Regression
Coefficients
T-ratios
Per Capita Income PPP .033 .432
Gross Domestic Saving .012 1.9000**
Health Expenditure as a percent of GDP -.020 -.515
Education Expenditure as a percent of GDP -.091 -2.264**
Labour force as a percent of total population .241 2.563**
Urban population as a percent of total population -.133 -.799
Share of Non Agriculture sector as a percent of GDP .562 1.237
Percentage of Labour Force in Non Agriculture sector .635 2.633**
Constant -3.416 -3.678
R2 .647
2
R .390
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
The coefficient of per capita income, labour force as a percent of total population, share of
non agriculture sector are positive but non significant. The coefficient of urban population is
negative but insignificant.
Upper Middle Income Countries: 1995
The result of the regression analysis for group-II countries for the year 1995 are presented in
the table 6.8. It is clear from the table that only 24 percent variations in the dependent variable are
explained by the explanatory variables under consideration. The coefficients of labour force as a
- 100 -
percent of total population is positive and significant at ten percent level of significance. The
coefficient of per capita income, gross domestic saving, health expenditure, education expenditure
and labour force in non agriculture sector are positive but non significant. The coefficient of urban
population and share of non agriculture sector is negative but non significant.
Table - 6.8
Cross Country Regression Results for the year 1995: Group II Upper Middle Income
Countries
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .050 .970
Gross Domestic Saving .039 .557
Health Expenditure as a percent of GDP .033 .883
Education Expenditure as a percent of GDP .001 .022
Labour force as a percent of total population .138 1.397***
Urban population as a percent of total population -.046 -.343
Share of Non Agriculture sector as a percent of GDP -.345 -.441
Percentage of Labour Force in Non Agriculture sector .355 1.354
Constant -1.280 -.958
R2 .446
2
R .244
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
- 101 -
Upper Middle Income Countries: 2005
The results of the cross country regression analysis for Group-II countries for the year 2005
are presented in the table 6.9. The results of the regression analysis show that 48.6 percent
variations in the dependent variable are explained by the set of eight explanatory variables under
study. The coefficient of per capita income is significant at 1 percent level of significance. The
coefficient of labour force as a percent of total population is positively significant at 10 percent level
of significance. The coefficient of gross domestic saving, health expenditure, education expenditure
and labour force in non-agriculture sector is positive but non significant. The coefficient of urban
population and share of non-agriculture sector is negative but non significant. The coefficient of
health expenditure is .034 meaning thereby that if there is one percent increase in the health
expenditure the human development index will increase by .034 percent. The coefficient of labour
force in non-agriculture sector is .170 meaning thereby that if there is increase of 1 percent in non-
agriculture labour force the human development index will increase by .170 percent.
The study found that over a period of time from 1985 to 1995 the value of 2
R has decreased
from 39 percent to 24.4 and it increased over the period from 1995 to 2005 to the value of 48.6. It
shows that explanatory variable affect the dependant variable and the effect declined from 1985 to
1995 subsequently inclined during the period starting from 1995 till 2005. The effect of per capita
income was non significant in the year 1985 and 1995 but it became significant in the year 2005.
The coefficient of labour force as a percent of total population remained significant throughout the
study period. The coefficient of education expenditure as a percent of GDP was negative and
significant in the year 1985 but it became positive though insignificant in 1995 and continued the
same in the year 2005. The coefficient of urban population as a percent of total population though
insignificant but continued to be negative throughout the period of study. The coefficient of non
agricultural labour force was positive and significant in the year 1985 but it became insignificant in
the year 1995 and 2005 though remained positive. The coefficient of share of non agricultural sector
as a percent of GDP was positive and non significant in the year 1985, it became negative in the year
1995 and 2005 but non significant.
- 102 -
Table - 6.9
Cross Country Regression Results for the year 2005: Group II Upper Middle Income
Countries
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .098 3.446*
Gross Domestic Saving .006 .212
Health Expenditure as a percent of GDP .034 .853
Education Expenditure as a percent of GDP .001 .026
Labour force as a percent of total population .082 1.632***
Urban population as a percent of total population -.001 -.021
Share of Non Agriculture sector as a percent of GDP -.140 -.260
Percentage of Labour Force in Non Agriculture sector .170 1.201
Constant -1.550 -1.755
R2 .702
2
R .486
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
Lower Middle Income Countries: 1985
The results of the regression analysis of Lower Middle income countries (Group-III) for the
year 1985 are presented in the table 6.10. The results of the regression analysis show that 61.6
percent variations in the dependent variable are explained by the factors under consideration.
- 103 -
Table - 6.10
Cross Country Regression Results for the year 1985: Group III Lower Middle Income
Countries
HDI Dependent Variable T-ratios
Per Capita Income PPP .286 2.537*
Gross Domestic Saving .015 1.456***
Health Expenditure as a percent of GDP .070 1.475***
Education Expenditure as a percent of GDP -.004 -0.65
Labour force as a percent of total population .055 .367
Urban population as a percent of total population -.438 -1.330
Share of Non Agriculture sector as a percent of GDP -.508 -.792
Percentage of Labour Force in Non Agriculture sector .700 3.347*
Constant -2.461 -2.110
R2 .778 .616
2
R .616
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
The coefficient of per capita income (PPP) and percentage of labour force in non-agriculture
sector is positive and significant at one percent level of significance. The coefficient of gross
domestic saving and health expenditure as a percent of GDP are positive and significant at ten
percent level of significance. The coefficient of education expenditure as a percent of GDP, urban
population as a percent of total population and share of non-agriculture sector as a percent of GDP
are negative but non-significant. The coefficient of labour force as a percent of total population is
positive but non-significant. The coefficient of labour force as a percent of total population is .055
- 104 -
meaning thereby that if there is one percent increase in it, the human development index will
increase by .055 percent.
Lower Middle Income Countries: 1995
The cross country regression results for the year 1995 for Group-III countries are presented
in the table 6.11. The results of the regression analysis show that 61 percent variations in the
dependent variable are explained by the set of explanatory variables. The coefficient of per capita
income (PPP) is positive and highly significant at one percent level of significance. The coefficient
of labour force in non-agriculture sector is positive and significant at 10 percent level of
significance. The coefficient of gross domestic saving, health expenditure as a percent of GDP,
labour force as a percent of total population and labour force in non-agriculture sector is positive but
non significant.
The coefficient of education expenditure, urban population and share of non-agriculture
sector is negative but non significant. The coefficient of health expenditure as a percent of GDP is
.102 meaning thereby that if there is increase of 1 percent in the health expenditure, the human
development index will increase by .102 percent.
Table - 6.11 Cross Country Regression Results for the year 1995: Group III Lower Middle Income
Countries HDI Dependent Variable Regression
Coefficients t-ratios
Per Capita Income PPP .407 3.002*
Gross Domestic Saving .011 .809
Health Expenditure as a percent of GDP .102 1.128
Education Expenditure as a percent of GDP -.026 -.345
Labour force as a percent of total population .346 1.275
Urban population as a percent of total population -.024 -.075
Share of Non Agriculture sector as a percent of GDP -.521 -.587
Percentage of Labour Force in Non Agriculture sector .326 1.431***
Constant -4.594 -2.497
R2 .774
2
R .609
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
- 105 -
Lower Middle Income Countries: 2005
The results of the cross country regression analysis for Group-III countries for the year 2005
are presented in the table 6.12. The results of the regression analysis show that 93.1 percent
variations in the dependent variable are explained by the set of eight explanatory variables under
consideration. The coefficient of per capita income, labour force as a percent of total population,
share of non- agriculture sector as a percent of GDP are positive and significant at one percent level
of significance. The coefficient of health expenditure as a percent of GDP and labour force in non-
agriculture sector is positive and significant at 5 percent level of significance. The coefficient of
gross domestic saving is negative and significant at 5 percent level of significance. The coefficient
of urban population as a percent of total population is negative and significant at 10 percent level of
significance. The coefficient of education expenditure is negative but non significant.
Table - 6.12
Cross Country Regression Results for the year 2005: Group III Lower Middle Income
Countries HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .330 6.853*
Gross Domestic Saving -.075 -2.250**
Health Expenditure as a percent of GDP .089 2.378**
Education Expenditure as a percent of GDP -.015 -.476
Labour force as a percent of total population .413 4.157*
Urban population as a percent of total population -.268 -2.070***
Share of Non Agriculture sector as a percent of GDP .971 4.402*
Percentage of Labour Force in Non Agriculture sector .313 2.450**
Constant -6.469 -9.873
R2 .960
2
R .931
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
- 106 -
The study found that the value of 2
R slightly decreased to 0.7 percent in 1995 from 1985.
But it significantly increased by 33 percent in the year 2005. This indicates that the effect of
explanatory variables on the dependent variable over a period of time has significantly increased.
The coefficient of per capita income persistently substantial during the study period. The coefficient
of gross domestic saving which was positive and significant in the year 1985 became non-significant
but positive in the year 1995, and negatively significant in the year 2005. The health expenditure
became non significant in the year 1995 and it again became significant in the year 2005. The
coefficient of labour force in non-agriculture sector remained significant throughout the study period
though at different levels of significance. The labour force as a percent of total population was non
significant in the year 1985 and 1995 but became significant in the year 2005. The coefficient of
education expenditure remained negative though insignificant during the period of study.
Low Income Countries: 1985
The results of cross country regression analysis for low income countries (Group-IV) are
presented in the table 6.13. The results of the regression analysis show that 32.7 percent variation in
the dependent variable are explained by the factors under study. The coefficient of education
expenditure as a percent of GDP and the labour force in the non-agriculture sector is positively
significant at 10 percent level of significance.
The coefficient of urban population as a percent of total population is negative and
significant at 10 percent level of significance. The coefficient of per capita income and health
expenditure as a percent of GDP is negative but non significant. The coefficient of education
expenditure and labour force in non agriculture sector is positive and significant at ten percent level
of significance. The coefficient of urban population is negative and significant at 10 percent level of
significance. The coefficients of the gross domestic saving, labour force as a percent of total
population and share of non-agriculture sector as a percent of GDP are positive but non significant.
The coefficient of health expenditure as a percent of GDP is negative but non significant.
- 107 -
Table - 6.13
Cross Country Regression Results for the year 1985: Group IV Low Income Countries
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP -.127 -.401
Gross Domestic Saving .020 .913
Health Expenditure as a percent of GDP -.098 -.643
Education Expenditure as a percent of GDP .254 1.773***
Labour force as a percent of total population .062 .210
Urban population as a percent of total population -.449 -1.398***
Share of Non Agriculture sector as a percent of GDP .819 .898
Percentage of Labour Force in Non Agriculture sector .444 1.479***
Constant -2.040 -.899
R2 .495
2
R .327
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
Low Income Countries: 1995
The results of cross country regression analysis for group-IV countries for the year 1995 are
presented in the table 6.14. The results of regression analysis show that 27 percent variation in the
dependent variable are explained by the set of explanatory variables under study. The coefficient of
gross domestic saving is positive and significant at 5 percent level of significance. The coefficient
of share of non agriculture sector as a percent of GDP and labour force in non-agriculture sector is
significant at 10 percent level of significance. The coefficient of per capita income, labour force, as
a percent of total population and urban population as a percent of total population are negative but
- 108 -
non significant. The coefficient of health expenditure, education expenditure are positive but non
significant.
The coefficient of health expenditure is .202 meaning thereby that if there is one percent
increase in the health expenditure, the human development index will increase by .202 percent.
Table - 6.14
Cross Country Regression Results for the year 1995: Group IV Low Income Countries
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP -.176 -1.271
Gross Domestic Saving .026 1.888**
Health Expenditure as a percent of GDP .202 .987
Education Expenditure as a percent of GDP .106 1.045
Labour force as a percent of total population -.738 -1.371
Urban population as a percent of total population -.376 -1.224
Share of Non Agriculture sector as a percent of
GDP
1.648 2.116***
Percentage of Labour Force in Non Agriculture
sector
.384 1.569***
Constant -.175 -.062
R2 .577
2
R
.269
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
Low Income Countries: 2005
The results of cross country regression analysis for the year 2005 for Group-IV countries are
presented in the table 6.15. The results of the cross country regression analysis show that 58.7
percent variations in the dependent variable are explained by the factors under study. The
coefficient of gross domestic saving is positive and significant at one percent level of significance.
The coefficient of urban population as a percent of total population is negative and that of labour
- 109 -
force in non-agriculture sector is positive and both are significant at 10 percent level of significance.
The coefficient of health expenditure is negative but non significant. The coefficient of per capita
income, education expenditure, labour force as a percent of total population and share of non-
agriculture sector as a percent of GDP are positive but non significant.
Table - 6.15
Cross Country Regression Results for the year 2005: Group IV Low Income Countries
HDI Dependent Variable Regression
Coefficients
t-ratios
Per Capita Income PPP .070 .880
Gross Domestic Saving .117 3.229*
Health Expenditure as a percent of GDP -.101 -1.328
Education Expenditure as a percent of GDP .082 1.376
Labour force as a percent of total population .569 1.111
Urban population as a percent of total population -.266 -1.771***
Share of Non Agriculture sector as a percent of
GDP
.480 1.224
Percentage of Labour Force in Non Agriculture
sector
.265 1.480***
Constant -4.496 -1.724
R2 .761
2
R .587
Note: * Significant at 1%
** Significant at 5%
*** Significant at 10%
The study found that the value 2
R first decreased to 27 percent in 1995 from 32.7 percent in
1985. It increased to 58.7 percent in the year 2005. This indicates that the effect of the explanatory
variables first decreased over a period of time from 1985 to 1995 and increased over a period of time
from 1995 to 2005. The coefficient of per capita income was negative in the year 1985 and 1995
- 110 -
though insignificant, it became positive in the year 2005 but insignificant. The coefficient of gross
domestic saving was insignificant in the year 1985 but significant in the year 1995 and 2005. The
coefficient of health expenditure was positive but insignificant in the year 1985 but it became
negative in the year 1995 and 2005 though remained insignificant. The education expenditure was
significant in the year 1985 but over a period of time it became non significant. The coefficient of
urban population remained negative throughout the period of study and significant except in the year
1985. The coefficient of percentage of labour force also remained significant throughout the period
of study.
SECTION - III
Developing Countries Analysis:
In this section analysis has been done to study the human capital status of developing
countries. The multiple regression analysis technique was applied to see whether the effect of
different factors differs significantly between the countries. The human development index has been
used as a proxy for judging the human capital level. Attempt has been made to find the relationship
of human development index with per capita income (PPP), gross domestic saving, health
expenditure as a percent of GDP, education expenditure as a percent of GDP, labour force as a
percent of total population, share of non-agriculture sector as a percent of GDP, percentage of labour
force in non-agriculture sector.
Developing Countries Analysis: 1985
The cross country regression result for all the developing countries for the year 1985 are
presented in the table 6.16. The results of the regression analysis as given by equation I shows that
75.3 percent variation in the dependent variable are explained by the set of explanatory variable
under study. The coefficient of per capita income (PPP) is positive and significant at one percent
level. This means that as the per capita income increases the human capital level increased
significantly in the year 1985. The coefficient of percentage of labour force in non-agriculture sector
is positive and significant at one percent level of significance. This is due to the fact that as the
labour force in non agriculture sector increases their income increases. The increase in income lead
to better standard of living of people and hence increase in human development index which
signifies the increase of human capital level. The coefficient of urban population as a percent of
total population is negative and significant at five percent level of significance. The coefficient of
gross domestic saving, health expenditure as a percent of GDP, labour force as a percent of total
- 111 -
population and share of non agriculture sector as a percent of GDP are positive but non significant.
The coefficient of health expenditure as a percent of GDP is .016. This indicates that if there is 1
percent increase in the health expenditure the human development index will increase by .016
percent.
The analysis has also been done to find the relationship by excluding the variable per capita
income. The equation II of the table shows that 68.3 percent variations in the dependent variable are
explained by the explanatory variables under consideration. The coefficient of health expenditure as
a percent of GDP and share of non-agriculture sector as a percent of GDP are positive and
significant at five percent level of significance. The coefficient of percentage of labour force as a
percent of total population remained positive and significant at one percent level and that of urban
population as a percent of total population is negative and significant at five percent level of
significance.
Table - 6.16 Cross Country Regression Results for the year 1985: Developing Countries Analysis
HDI Dependent Variable Eq 1 T-Value Eq II T-Value
Per Capita Income PPP .220 3.979*
Gross Domestic Saving .007 .960 .013 1.714
Health Expenditure as a percent of GDP .011 .277 .076 1.895**
Education Expenditure as a percent of
GDP
.016 .375 .005 .107
Labour force as a percent of total
population
.074 .731 .021 .182
Urban population as a percent of total
population
-.299 -2.243** -.291 -1.925**
Share of Non Agriculture sector as a
percent of GDP
.459 1.110 .993 2.238**
Percentage of Labour Force in Non
Agriculture sector
.508 3.527* .740 4.956*
Constant -3.726 -5.385 -3.349
R2 .787 .721
2
R .753 .683
Note: * Significant at 1% level of significance
** Significant at 5% level of significance
*** Significant at 10% level of significance
- 112 -
Developing Countries Analysis: 1995
The cross country regression results for human capital status for the year 1995 are presented
in the table-6.17. The equation I of the table show that 79.7 percent variations in the dependent
variable are given by the set of explanatory variables under study. The coefficient of per capita
income (PPP) is found to be positive and significant at one percent level of significance. The
coefficient of share of non-agriculture sector as a percent of GDP is positive and significant at five
percent level and the coefficient of percentage of labour force in non- agriculture sector is positive
and significant at one percent level of significance.
Table - 6.17
Cross Country Regression Results for the year 1995: Developing Countries Analysis
HDI Dependent Variable Eq 1 T-Value Eq II T-Value
Per Capita Income PPP .162 3.619*
Gross Domestic Saving .001 .066 .012 2.021**
Health Expenditure as a percent of GDP .026 .557 .100 2.112**
Education Expenditure as a percent of
GDP
.005 .117 .006 .137
Labour force as a percent of total
population
.021 .174 .014 .109
Urban population as a percent of total
population
-.249 -1.641*** -.251 -1.490***
Share of Non Agriculture sector as a
percent of GDP
1.151 2.811** 1.722 4.107*
Percentage of Labour Force in Non
Agriculture sector
.364 3.184* .519 4.413*
Constant -4.270 -5.306 -4.443 -4.981
R2 .825 .780
2
R .797 .750
Note: * Significant at 1% level of significance
** Significant at 5% level of significance
*** Significant at 10% level of significance
- 113 -
The coefficient of gross domestic saving, health expenditure, education expenditure as a
percent of GDP, labour force as a percent of total population are positive but non significant. The
coefficient of labour force as a percent of total population is .021 indicating that if there is one
percent increase in the labour force as a percent of total population the human development index
will increase by .021 percent.
The analysis has also been made to find the relationship by excluding the variable per capita
income (PPP). The equation II of the table shows that 75 percent variations in the dependent
variable are explained by the explanatory variables under study. The coefficient of gross domestic
saving and health expenditure as a percent of GDP are positive and significant at five percent level
of significance. The coefficients of share of non-agriculture sector as a percent of GDP become
positive and significant at one percent level of significance. The coefficient of urban population as a
percent of total population remained negative and significant at 10 percent and that of percentage of
labour force in non- agriculture sector positively significant at one percent level of significance.
Developing Countries Analysis: 2005
The cross country regression results for human capital status for developing countries for the
year 2005 are presented in the table 6.18. The equation I of the table shows that 88.6 percent
variations in the dependent variable are explained by the explanatory variables under study. The
coefficients of per capita income (PPP) and percentage of labour force as a percent of total
population are positive and significant at one percent level of significance. The coefficient of labour
force as a percent of total population and coefficient of share of non-agriculture sector as a percent
of GDP are positive and significant at five percent level of significance. The coefficient of urban
population as a percent of total population is negative and significant at five percent level of
significance. The coefficient of gross domestic saving, health expenditure and education
expenditure are positive but non-significant.
Analysis has also been made to find the relationship by excluding the variable per capita
income. The equation II of the table shows that 83.4 percent variations in the dependent variable are
explained by the explanatory variables. The coefficient of health expenditure as a percent of GDP is
positive and significant at five percent level of significance. The coefficient of labour force as a
percent of total population and share of non agriculture sector became significant at one percent
level of significance.
- 114 -
Table - 6.18
Cross Country Regression Results for the year 2005: Developing Countries Analysis
HDI Dependent Variable Eq 1 T-Value Eq II T-Value
Per Capita Income PPP .139 4.998*
Gross Domestic Saving .011 .542 .003 .138
Health Expenditure as a percent of GDP .034 .920 .090 2.132**
Education Expenditure as a percent of
GDP
.019 .679 .037 1.116
Labour force as a percent of total
population
.247 2.706** .379 3.600*
Urban population as a percent of total
population
-.235 -2.639** -.265 -2.472**
Share of Non Agriculture sector as a
percent of GDP
.616 2.635** .880 3.196*
Percentage of Labour Force in Non
Agriculture sector
.458 4.307* .777 7.580*
Constant -4.177 -8.645 -4.641 -8.098
R2 .902 .853
2
R .886 .834
Note: * Significant at 1% level of significance
** Significant at 5% level of significance
*** Significant at 10% level of significance
SECTION IV
When the analysis for the determinants of human development level is done for all the
countries as a whole the study found that the value of R2 and
2
R which was 83.8 percent and 82
percent respectively in the year 1985 has increased to 87.9 percent and 86.5 percent in 1995. It
further increased to 92.6 and 91.7 percent in 2005. This indicates that the effect of the explanatory
variables on the dependent variables has increased over the period of time. The per capita income
- 115 -
has continued to be highly significant throughout the study period. The coefficient of gross domestic
saving was significant in the year 1985 but it became insignificant in the year 1995 and 2005. The
coefficient of health expenditure remained significant upto the year 1995 and in the year 2005 it
became insignificant. This indicated that no further increase in health expenditure would result into
the enhancement of human development level. The coefficient of urban population as a percent of
total population remained negatively significant throughout the study period. The coefficient of
share of non agriculture sector as a percent of GDP was non significant in the year 1985 but it
became significant in the year 1995 and continued till 2005. The coefficient of labour force in the
non agriculture sector continued to be significant throughout the study period.
The study found that for Group I (high income) countries over the period from 1985 to 1995
the value of 2
R increased from 81 to 87.6 percent and further it increased to 91 percent in 2005.
The coefficient of per capita income is insignificant in the year 1985 but it became highly significant
in 1995 and 2005. The education expenditure was significant in the year 1985 but it became positive
and insignificant in 1995 and in 2005 its coefficient became negative though non-significant. The
coefficient of urban population was negative in the year 1985 but it became positive in the year 1995
and 2005. The share of non agriculture sector as a percent of GDP was negative in the year 1985
and 1995 but it became positive in the year 2005. The coefficient of health expenditure was
insignificant in the year 1985 and 1995 but over a period of time in 2005 it became significant. The
coefficient of education expenditure was positive in the year 1985 and 1995 but it became negative
in the year 2005.
The study found that over a period of time from 1985 to 1995 the value of 2
R have
decreased from 39 percent to 24.4 percent and it increased over a period from 1995 to 2005 to the
value of 48.6 for group II (upper middle income) countries. The effect of per capita income was non
significant in the year 1985 and 1995 but it became significant in the year 2005. The coefficient of
labour force as a percent of total population remained significant throughout the study period. The
coefficient of education expenditure as a percent of GDP was negative and significant in the year
1985 but it became positive though insignificant in 1995 and continued to be same in the year 2005.
The coefficient of urban population as a percent of total population though insignificant, but
continued to be negative throughout the period of study. The coefficient of non agricultural labour
force was significant in the year 1985 but it became insignificant in the year 1995 and 2005. The
- 116 -
coefficient of share of non agriculture sector as a percent of GDP was positive and non significant in
the year 1985, it became negative in the year 1995 and 2005 but non-significant.
The study found that for Group III (Lower Middle Income) countries the value of 2
R has
slightly decreased to the tune of 0.7 percent in 1995 from 1985. But it significantly increased by 33
percent in the year 2005. This indicates that the effect of explanatory variables on the dependent
variables over a period of time has significantly increased. The coefficient of per capita income
remains highly significant throughout the period of study. The coefficient of gross domestic saving
which was positive and significant in the year 1985, became non-significant but positive in the year
1995 and negatively significant in the year 2005. The health expenditure became non-significant in
the year 1995 and it became significant in the year 2005. The coefficient of labour force in non
agriculture sector remains significant throughout the study period though at different levels of
significance. The labour force as a percent of total population was non significant in the year 1985
and 1995 but became significant in the year 2005. The coefficient of education expenditure
remained negative though insignificant during the period of study.
While analyzing the Group IV (low income countries), the study found that the value of 2
R
first decreased to 27 percent in 1995 from 32.7 percent in 1985. It increased to 58.7 percent in the
year 2005. This indicates that the effect of the explanatory variables first decreased over a period of
time from 1985 to 1995 and increased over a period of time from 1995 to 2005. The coefficient of
per capita income was negative in the year 1985 and 1995 though non significant, it became positive
in the year 2005 but remained non significant. The coefficient of gross domestic saving was non
significant in the year 1985 but significant in the year 1995 and 2005. The coefficient of health
expenditure has been found to be positive but non significant in the year 1985 but it became negative
in the year 1995 and 2005 though remained non significant. The education expenditure was
significant in the year 1985 but over a period of time it became non-significant. The coefficient of
urban population remained negative throughout the period of study and significant except in the year
1985. The coefficient of percentage of labour force also remained significant throughout the period
of study.
When the analysis is done for developing countries the study found that over a period of time
from 1985 to 2005 the value of 2
R increased to 88.6 percent in 2005 from 75.3 in 1985. The
increase in 2
R indicates that the effects of explanatory variable on the dependent variable have
- 117 -
increased. The coefficient of per capita income and coefficient of percentage of labour force in non-
agriculture sector and coefficient share of non-agriculture sector as a percent of GDP remained
positively significant throughout the study period. The coefficient of urban population as a percent
of total population remained negatively significant throughout the study period. The coefficient of
gross domestic saving was positive but non significant in 1985. It became significant in 1995 when
the variable per capita income was excluded from the list.
The coefficient of health expenditure was positively significant throughout the period of
study when the variable per capita was excluded. The coefficient of education expenditure as a
percent of GDP remained positive but non significant during the period of study. The coefficient of
labour force as a percent of total population was positive but non-significant in 1985 and 1995 but
became significant in the year 2005.
****