The Dynamics of Human Development Index

G Model
SOCSCI-1196; No. of Pages 17
ARTICLE IN PRESS
The Social Science Journal xxx (2014) xxxxxx
Contents lists available at ScienceDirect
The Social Science Journal

journal homepage: www.elsevier.com/locate/soscij
The dynamics of Human Development Index

Jack Hou a,b , Patrick Paul Walsh c , Jing Zhang d,
a
b
c
d
School of Economics, Henan University, China

Department of Economics, California State University, Long Beach, USA
SPIRe and Geary Institute, University College Dublin, Ireland
School of Economics, Xiamen University, China
a r t i c l e
i n f o
Article history:
Received 29 April 2014
Received in revised form 15 July 2014
Accepted 15 July 2014
Available online xxx
JEL classication:
L11
F15
a b s t r a c t
In criticizing the nature of the Human Development Index, this paper proposes a different
way of constructing the HDI in terms of capturing the pure ow of human development in
the areas of material well-being, health, and education. Our comparison of the HDIF and
the HDI shows that measuring human development with ow variables provides a better
human development performance in terms of health and education than the traditional
measure which is a mixture of stock and ow variables.
2014 Western Social Science Association. Published by Elsevier Inc. All rights reserved.
Keywords:
Human development
Policy
Education
Health
GDP
Dynamic analysis
1. Introduction
The Human Development Index (HDI) is a composite statistic aimed at measuring the human development
level of any country and to allow cross-country comparison. Based on the HDI, a country is classied into
three tiers of development: developed, still developing, or
underdeveloped. According to the United Nations Development Programme (UNDP), human development is about
We appreciate the nancial support from National Science Foundation of China (71303200) and Ministry of Education of the Peoples
Republic of China (12YJC790263). We are grateful to anonymous referees
for their invaluable suggestion.
Corresponding author at: School of Economics, Xiamen University,
Fujian 361005, China. fax: +86 592 2186366.
E-mail addresses: Jack.Hou@csulb.edu (J. Hou), ppwalsh@ucd.ie
(P.P. Walsh), jzhang1@xmu.edu.cn (J. Zhang).
creating an environment in which people can develop their

full potential and lead productive, creative lives in accord
with their needs and interests. People are the real wealth of
nations. Development is thus about expanding the choices
people have to lead lives that they value (Streeten, 1994).
In order to reect this, the HDI is based on three equally
weighted dimensions: life expectancy, education attainment, and gross domestic product per capita. The HDI has
been published in the Human Development Reports commissioned by the UNDP since 1990.
The introduction of the HDI attracted widespread attention, which reects the general dissatisfaction with the
conventional real per capita GDP or real wages as a measure of changes in living standards. The HDI was hailed
as an improvement as it extended beyond the purchasing power of private incomes to include the measurement
of well-being and the quality of life. In addition, since the
per capita GDP and the other components of HDI life
http://dx.doi.org/10.1016/j.soscij.2014.07.003
0362-3319/ 2014 Western Social Science Association. Published by Elsevier Inc. All rights reserved.
Please cite this article in press as: Hou, J., et al. The dynamics of Human Development Index. The Social Science Journal
(2014), http://dx.doi.org/10.1016/j.soscij.2014.07.003
G Model
ARTICLE IN PRESS
J. Hou et al. / The Social Science Journal xxx (2014) xxxxxx
.2
.2
Education index
.4
.6
Life expectancy at birth

.4
.6
.8
.8
.2
.4
.6
.8
.2
.4
.6
.8
GDP index
GDP index
Fig. 1. (i) GDP index and life expectancy index. (ii) GDP index and education index.
Source: The data for the GDP index, life expectancy index and education index is obtained from various Human Development Reports.
expectancy and educational attainment have generally

advanced at varying rates over time, this allows a country
to evaluate its economic progress in different developmental stages. Consequently, the HDI is increasingly used to
set a countrys human development goals, policy designs
for economic development, and evaluation of such policy
measures. More than one hundred countries have already
constructed their own national or sub-national HDIs.
As widely used as HDI is, however, it is subject to a number of well-grounded criticisms, despite its improvement
over the original income based measures. This study does
not profess to answer all these criticisms. It is modest in
comparison, as it focuses on just one of the four categories
of criticisms the mixture of stock and ow measures in the
construction of the HDI, and proposes an alternative pure
ow-based measure (HDIF). This study compares the current HDI with the proposed HDIF across a host of countries,
encompassing all three levels of development.
2. Human Development Index: background and

criticism
2.1. Background of the HDI
The United Nations started calculating the HDI for its
member in 1975 and published the rst Human Development Report in 1990. The motivation for the Report was
to rene the basis for measuring a countrys development
and prosperity. The UNDP argues that economic prosperity,
as measured by real income per capita, is not the only factor in measuring human development, because an increase
in income per capita does not necessarily mean that the
people of the country are better off as a whole.
Based on Fig. 1(i) and (ii), we can see that countries with
an income level or GDP index of 0.6, the development level
in terms of health or life expectancy index spans from 0.2
to 0.8, and similarly the education index ranges ranging
from 0.4 to close to 1. Clearly, economic growth does not
necessarily lead to the overall development of the human
being. The rst HDR correctly recognizes that development is much more than just the expansion of income and
wealth and denes human development as the process

of enlarging peoples choices (UNDP, 1990, p. 10). Thus,
the rst HDR employs the HDI and examines such concepts as health and life expectancy, education, and work as
well as leisure time. Today, the HDI examines three basic
dimensions to determine a countrys growth and achievements in human development. The rst is health, which is
measured by life expectancy at birth. Those with higher
life expectancies rank higher than those with lower life
expectancies. The second dimension measures a countrys
overall knowledge level via the literacy rate among adults
over 25 years combined with the gross enrolment ratio
of students from primary school through university. The
third and nal dimension is a countrys standard of living, measured as the gross domestic product per capita in
purchasing power parity terms, based on the United States
dollar.
To assure accuracy, a separate index is calculated for
each dimension using raw data specically gathered for
the report. A simple intermediate index was created for
each dimension by setting each indicator on a scale of 01
using maximum and minimum scaling values. Scaling and
the normalized value are required for two reasons. First,
the variables included in the HDI have different units; life
expectancy is in years, adult literacy rate is in percentages,
and the GDP per capita is in PPP$. Normalization removes
the heterogeneity of units by converting them into pure
numbers. Such transformation is necessary as the composite index is accomplished by adding up the intermediate
indices. Second, scaling enables comparisons as to how a
country fared in each dimension, and how far it has to go.
The HDI for each country is then calculated as a simple average of the three indices, including the life expectancy index,
the gross enrolment index, and gross domestic product.
2.2. Criticism of the HDI

Since its inception, the HDI has been the focus of public
debate and criticism in three broad groups. The rst group
is about methodology. Weighting the three components of
the index equally suggests a perfect substitution between
G Model
ARTICLE IN PRESS
longevity, knowledge, and living standards and, therefore,

implicit trade-offs between the three dimensions (Desai,
1991; Kelly, 1991; Ravallion, 1997). As a result, countries
ranked very closely together can have signicantly different development indexes in a given dimension. Moreover,
as the absolute value of each component affects the level
of the HDI, the maximum and minimum values selected
affect the value of the index, resulting in a change in the
ranking order (Noorbakhsh, 1998a). Hopkins (1991) states
that there is no a priori rationale why life expectancy and
literacy can even be added together. Most researchers proposed modications to the standard HDI have maintained
the additivity of the components in their indexes; these
include the Modied HDI (Noorbakhsh, 1998a) and the
Rescaled New HDI (Mazumdar, 2003) that use Euclidean
vector distance.
Not all approaches, however, maintain the equal
weighting of components (Despotis, 2005; Lozano &
Gutirrez, 2008; Noorbakhsh, 1998b). For example, Desai
(1991) suggests a log additive formula to restrict substitutability. Sagar and Najam (1998) propose a reformed HDI
by multiplying component indexes instead of averaging
them. In this case, simultaneously high values for all three
component indexes are required to obtain a high value HDI.
The second block of the criticism centers on the choice
of dimensions included in the HDI. Stewart, Ranis, and
Samman (2006) suggest that, as a multi-dimensional process, the HDI should include development aspects such
as peace, security, environmental concerns, cultural freedom, and access to social services. The rst HDR did discuss
human freedom and human rights, stating that human
development is incomplete without human freedom, but
also that while the need for quality judgment is clear, there
is no simple quantitative measure available yet to capture
the many aspects of human freedom. (HDR, 1990). Civil
rights and political freedom are popular dimension (Boer
& Koekkoek, 1993; Dasgupta & Weule, 1992; Desai, 1991;
Hopkins, 1991; Khatib, 1994; Rao, 1991; Trabold-Nbler,
1991).
Both Desai (1991) and Neumayer (2001) include
resource exploitation and environmental problems in their
discussions. Many other researchers have proposed various
alternatives or additional dimensions that can or should be
considered.
There is also the concern of combining stock variables
such as adult literacy and ow variables like the combined
enrolment ratio in the construction of HDI. Because of this
potential for misuse of data, the HDI thus cannot capture
the short-run changes in human development in a country. It is also not very sensitive to policy changes and as
such may frustrate the policy makers.1 Ignorance of this
problem could lead to the inappropriate comparison across
countries. Therefore, to ll the gaps of existing research,
this paper proposes a solution to the existing HDI so that it
1
As a metaphor, what if Doctoral students are admitted based on
their cumulative GPA since kindergarten? That would be what a stock
measure implies. In contrast, a ow-based measure would evaluate the
applicant based on student performance in the last two years of undergraduate studies.
would be more sensitive to policy changes and would also

be able to monitor short-term progress which is important for assessment, policy making, and enhancing lives of
millions.
Additionally, there is an even more basic criticism
that HDI is a mixture of both ow and stock variables
(Ephrenesis, 1994; Hopkins, 1991; Ivanova, 1994; Pyatt,
1992). Criticisms have hinged largely on the time lags associated with the impact of policy changes on adult literacy
and life expectancy and, hence, the HDI partially measures
the outcomes of past efforts and not just the present situation. It is also unclear what the comparison between GDP
and HDI rankings signies, given that GDP is a ow and
the HDI is a combination of stocks and ows. This would
affect the cross-country comparison and mislead policy
makers in setting and implementing policies. Given that
HDI has become a major policy target, above and beyond a
reection of a nations true performance, it is crucial that
we improve the HDI so that it becomes comparable across
countries contemporaneously so as to serve as a clear gauge
of what needs to be done.
To this end, this paper constructs an index consisting
only of indicators that reect current ows in order to capture the pure ow of human development in the areas
of material well-being, health, and education. In contrast
to the existing HDI, which mix stock variables including
life expectancy and literacy rate with ow variables such
as GDP per capita and gross enrolment ratio. We used
the ow variables of under-ve mortality rate to replace
life expectancy and gross primary-school enrolment ratio
instead of literacy rate together with GDP per capita to form
a new index, the HDIF. It is our belief that the HDIF measure
will provide a better human development measure than the
traditional HDI.
3. Data and construction of HDIF
3.1. Construction of the HDIF
The current HDI consists of both stock (life expectancy
and literacy rate) and ow (GDP per capita) measures of different aspects of human development. As a result, it reects
the cumulative past efforts on health and education and
current income levels. Among the three dimensions, literacy rate and life expectancy tend to be more stationary.
A person, once literate, typically does not become illiterate because of any economic or social changes. Likewise,
life expectancy is a summary measure of the health of a
population, which can reect climate, culture, and public investment in preventive care, all of which tend to
change slowly and have lasting effects. For this reason,
life expectancy is not suitable for making comparisons of
health changes between countries at different income levels and at different points in time. While not as volatile as
income, enrolment rates also tend to be more cyclical than
literacy rates. A recession could prevent families from sending their children to school, leading to rapid changes in the
school-enrolment ratio. Compared to life expectancy, the
under-ve mortality rate is one of the most sensitive indicators of human welfare, the comparative health of nations,
and the effectiveness of public policy and is more precise
G Model
ARTICLE IN PRESS
Y min(Y )
max min
where the min = 0 and the max = 1.

The GDP index comes directly from various issues of
HDRs.
Finally, the same formula for the HDI is applied to the
HDIF
HDIF =
MR5 + GEI + GDP

3
Figs. 2 and 3 show the HDI and HDIF trends by region,

and several interesting patterns em.
First, the regional ranking was the same for both HDI
and HDIF. North America has the highest regional average followed by Europe and Central Asia in both measures.
.8
GEI =
4.1. Comparison of levels between the HDIF and the

HDIhou
HDI
As the under-ve mortality rate ranges from 0 to 320 out

of 1,000 births, the minimum value is 0 and the maximum
value is 0.32.
The gross primary-school enrolment ratio (PSER)
ranged from 0 to 100 and could be used to calculate the
Gross Enrolment Index (GEI):
4. The HDIF and the HDI
.6
X min(X)
MR5 = 1
max min
As shown in Table 1, there has been an increasing trend

in the level of all the sub-components of human development, though with regional differences. North America
led in all aspects of human development over the sample
period studied, regardless of how health or education is
measured. This is hardly surprising as this region consists
of only the U.S. and Canada. Following North America was
Europe and Central Asia in 1980, Latin America and the
Caribbean caught up in the following years. Sub-Saharan
Africa was at the bottom in almost all sub-components of
the HDI, with South Asia only fairing slightly better.
.4
in terms of change over time (HDR, 2002). Therefore, this

paper selects gross primary-school enrolment ratio and the
under-ve mortality rate as replacements for the two stock
variables of literacy and life expectancy in the current HDI.
We followed the UNDPs formula, which converts the
raw data into a unit-free index with values between 0 and
1. In contrast to the other two dimensions per capita
GDP and gross primary-school enrolment ratio where
high value corresponds to a high index value, lower underve mortality rate actually implies better performance and
should correspond to a higher index value. Therefore, a
reciprocal version of the formula for the mortality rate is
used.
The corresponding under-ve mortality Index (MR5) is
calculated as:
1980
2
The classication of regions is consistent with the one used in various
Human Development Reports commissioned by UNDP.
3
The main reason the sample period ended at 2007 is to avoid the
disruptive effect of the sub-prime crisis.
2010
East Asia and Pacific

Latin America and the Caribbean
North America
Sub-Saharan Africa
Europe and Central Asia

Middle East and North Africa
South Asia
.7
HDIF
.8
.9
Fig. 2. HDI trend by region, 19802007.

Source: The data for the HDI index is obtained from various Human Development Reports.
.6
The data set in this analysis spans from 1980 to 2005 in

ve-year intervals, together with data for the two consecutive years 2006 and 2007 for up to 170 countries grouped
into seven regions2 (East Asia and Pacic, Europe and Central Asia, Latin America and the Caribbean, Middle East
and North Africa, North America, South Asia, and SubSaharan Africa).3 The full data set is drawn from several
data sources. The trend data for the HDI and its components
is obtained from various Human Development Reports.
The data used to construct the HDIF includes the underve mortality rate from the World Health Organizations
Statistical Information System, the gross primary-school
enrolment ratio from UNESCO (Table 1 for index values),
and the GDP index from various HDRs, which were used to
construct the ofcial HDI.
2000
Year
As the GDP index is not available for all the countries

over the sample period, some missing values are generated
in the HDIF.
3.2. Data
1990
1980
1990
2000
2010
Year
North America
Sub-Saharan Africa

South Asia
Fig. 3. HDIF trend by region, 19802007.

Source: The construction of HDIF is based on data from World Health
Organizations Statistical Information System, UNESCO and Human Development Reports.
G Model
ARTICLE IN PRESS
Table 1
Summary statistics for different measures of health and education indexes by region in 1980, 2000, and 2007.
Under-ve
mortality index
Life expectancy
index
Enrolment
index
Education
index
GDP index
Year = 1980
East Asia and the Pacic
The Middle East and North Africa
North America
South Asia
Sub-Saharan Africa
0.806
0.882
0.810
0.802
0.967
0.594
0.582
0.390
0.760
0.661
0.622
0.825
0.464
0.427
0.958
0.949
0.968
0.881
0.988
0.682
0.739
0.669
0.851
0.723
0.586
0.931
0.495
0.430
0.605
0.827
0.663
0.757
0.920
0.390
0.451
Year = 2000
North America
South Asia
Sub-Saharan Africa
0.890
0.946
0.920
0.917
0.983
0.758
0.676
0.723
0.808
0.768
0.767
0.894
0.599
0.471
0.977
0.980
0.995
0.927
0.997
0.804
0.808
0.799
0.928
0.824
0.720
0.971
0.575
0.557
0.622
0.794
0.687
0.767
0.983
0.520
0.443
Year = 2007
North America
South Asia
Sub-Saharan Africa
0.919
0.966
0.940
0.940
0.983
0.798
0.724
0.759
0.834
0.795
0.792
0.915
0.654
0.498
0.994
0.988
0.991
0.959
0.990
0.969
0.929
0.824
0.940
0.861
0.792
0.980
0.610
0.602
0.652
0.844
0.722
0.796
0.991
0.548
0.471
Regions
Note: The trend data for the HDI and its components is obtained from various Human Development Reports. The data on under-ve mortality rate is from
the World Health Organizations Statistical Information System, the gross primary-school enrolment ratio is from UNESCO.
Using the HDI measure, the middle three regions Middle

East, Latin America, and East Asia, had similar index values
between 1985 and 1995, implying similar levels of overall
human development. However, the Middle East and Latin
America outperformed East Asia after 2000. One can certainly speculate whether the 1998 East Asia Financial Crisis
was the cause of this relative decline.
It is, however, quite evident that when evaluating the
development performance based on pure ow measures
(HDIF), the gap between the three mid-level regions and
Europe and Central Asia are much smaller. This precisely
demonstrates the point that a good governments policy
effects are unjustly penalized in a stock based measure
as it takes much longer for the marginal ow effects to
signicantly move the average stock mistakes of past sociopolitical problems. Similarly, though on a slightly smaller
scale, the gap between Europe and North America is also
narrower with the HDIF compared to the stock-based HDI.
This is probably largely due to the education and health
performances of the Central Asia countries in the past couple of decades which will receive a more detailed analysis
later in this paper.
The most signicant disparity between the HDI and
HDIF measure lies in the lowest ranking regions, South Asia
and Sub-Saharan Africa. Based on the HDI measure (Fig. 2),
South Asias initial score (1980) was slightly higher than
Sub-Saharan Africa, and this positive gap widened over
the years before narrowing after 2000. Furthermore, this
narrowing gap is the result of a stagnated South Asia relative to an improving Sub-Saharan Africa. The picture is
slightly different when evaluated with the HDIF measure.
As shown in Fig. 3, South Asia started at a lower HDIF score
than Sub-Saharan Africa in 1980, but surpassed the latter by
1985 and exhibited accelerated development as indicated

by the concavity of its development path, until 2000 when
its develop rate slowed. This slower rate was probably close
to the average or stock measure, and that is why the HDI
showed stagnated development (Fig. 2). In contrast, the
Middle East showed little progress before 19901995 and
gained momentum since, thus narrowing the gap between
the two regions.4 As before, it is also evident that based
on the HDIF measure, the score of the two least developed
regions performed much better than the stock-laden HDI
suggests.
Improvements in human development over time are
evident according to both the HDI and the HDIF measures. However, human development as measured by the
pure ow variables of HDIF exhibit higher index values than the mixed HDI variables, especially for low
human-development regions. As both measures of human
development include GDP per capita, this pattern must
largely be due to the difference in non-income dimensions, such as the ow measure of under-ve mortality
and the gross primary-school enrolment ratio relative to
the stock variables of life expectancy and education level
which do not vary contemporaneously with policies or
social changes.
Thus, this study contends that HDI reects the accumulated impact of policies in the past, and is, thus, hard to
4
The timing of this improved development is intriguing, as the initial
upswing appeared right after the Iraq-Kuwait war, while the acceleration
is more reective of the fall of Saddam Hussein and perhaps leading into
the Arab Spring. More comprehensive analysis of this regions development is certainly worthwhile.
G Model
ARTICLE IN PRESS
.1
0
.05
HDIF-HDI
.15
.2
1980
1990
2000
2010
Year
North America
Sub-Saharan Africa

South Asia
Fig. 4. Difference between the level of HDIF and HDI from 1980 to 2007.
Source: The data for the HDI index is obtained from various Human Development Reports. The construction of HDIF is based on data from World
Health Organizations Statistical Information System, UNESCO and Human
Development Reports.
identify the effectiveness of policies implemented in different years or how soon the policy took effect. In other
words, HDI does not give policy makers a very clear signal.
On the other hand, HDIF provides a more accurate picture
of a countrys performance in real time. HDIF enables policy makers to better gauge the effectiveness of their policies
and make adjustments when necessary. This is especially
true for developing countries as the HDI is heavily stockladen and slow to respond to new policies (Fig. 2).
The high and middle human-development regions
experienced a reduction in the HDIF-HDI gap (Fig. 4),
which implies that the non-income quality-of-life variables are converging. This is not surprising, as a major
factor leading to increased literacy levels has been more
widespread access to basic education. Similarly, increased
life expectancy is a consequence of improved under-ve
survival rate. High human-development countries, albeit
from a higher starting position in terms of both health and
education, not only efciently converted their wealth into
human development in both measures, but nal convergence is a sign that the ow variables have reached the
upper bounds in terms of a 100% enrolment ratio and a
0.04% mortality rate.
On the other hand, in the low human development
countries, such as South Asia and Sub-Saharan Africa, the
gap between the HDIF and the HDI indices become more
pronounced after 2000. Both the stock and ow measures
of these regions started at a lower base compared to the
high human-development countries. The divergence pattern could be the result of rapid improving ow variables,
causing both health and education components of the HDIF
to show high growth rates. The stock variables in the HDI
index will grow also, but at a slower and delayed rate. The
aggregate regional index values discussed above convey
important information about inter-regional development;
they do not give an accurate picture of how each country
within the same HDI group performs. In addition, it would
be of interest to distinguish which of the ow variables
health or education drives the human development
of a particular country. Given the page limitations, it is
impossible to go into details of each nations development

history, but the interested reader can probably verify their
understanding of a particular nation, or nd inconsistencies that may spark their intellectual curiosity and prompt
further investigation.
Figs. 5.15.7 5 illustrate the gap between HDIF and HDI
measures, decomposing the inuence into education and
health. In East Asia and the Pacic (Fig. 5.1), the main driver
seems to be education; this is especially pronounced in
Cambodia, Myanmar, and Lao, and China.6 For countries
from the former Soviet Union (Fig. 5.2), health contributes
more than education to the general differences between
the HDIF and the HDI,7 while education seems to be more
about the cause of uctuations in the gap between the two
indices. In Latin America and the Caribbean (Fig. 5.3), most
countries experienced a decline in both the health and education gap, Peru and Venezuela. This convergence mainly
comes in education for Bolivia, whereas in Grenada, health
is the main contributor. Education contributes the major
effect on overall convergence in the Middle East and North
Africa (Fig. 5.4), countries such as Iran and Yemen.
As should be evident, the lowest human development
regions of South Asia and Sub-Saharan Africa are the most
interesting regions for comparisons between HDI and HIDF.
For the former (Fig. 5.6), the gap in education is generally
higher than that in health with Sri Lanka being the only
exception. The rise in the education gap component of the
HDIF and HDI is nowhere more evident than in Bhutan.
When the Human Development report was rst unveiled in
1990, Bhutan had the lowest literacy rate. With grants from
the Asia Development Bank, Bhutan embarked on an ambitious education expansion plan, and reached 100% primary
enrolment rate in 2007. This is truly the poster child of the
deciency of the stock-ladenned HDI measure compared
to the more real-time nature of the HDIF index. Another
country of interest in the South Asia region is Afghanistan.
With the ouster of the Taliban regime and the restoration of
some resemblance of civil government, the education gap
between the stock and ow measure increased dramatically after 2005.
Sub-Saharan Africa countries were the most heterogeneous (Fig. 5.7), not only in terms of the gap but also in the
time trend. There are three different types of countries. In
the rst group, which includes Rwanda and Uganda, both
education and health had a similar effect on the gap. In
the second group, which includes countries such as Mali,
the gap between the HDIF and the HDI widened mainly
as a result of education. The third group, which includes
countries such as Togo, the convergence of the gap is a
result of education.
5
Selected countries in each group are shown in Figs. 5.15.7. Please
contact authors for more information.
6
As China has gained much notoriety mainly due to her economic performance, interested readers can refer to Hou (2011a, 2011b) for a general
discussion on the success and potential problems down the road. For
some insight into Chinas political governance, refer to Li and Lin (2011)
and Man, Zheng, and Lang (2011). For an introduction to Chinas health
and education issues, Hou and Li (2011) and Hu and Hibel (2013) provide
interesting perspectives.
7
Hsu (2013) provides detailed discussion of this issue.
G Model
ARTICLE IN PRESS
Brunei Darussalam
Cambodia
China
Fiji
Indonesia
0 .2 .4 .6 .8
HDIF-HDI
0 .2 .4 .6 .8
Australia
1980
2000
2010
Kiribati
0 .2 .4 .6 .8
Japan
1990
1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
Graphs by country
health
(i)
Malaysia
Myanmar
New Zealand
Philippines
Samoa
Mongolia
.2 .4 .6
Papua New Guinea
Singapore
.2 .4 .6
HDIF-HDI
.2 .4 .6
Lao People's Democratic Republic
1980
1990
2000
2010 1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
health
Graphs by country
(ii)
Fig. 5.1. Differences in health and education between ow and stock measures in East Asia and the Pacic.
Source: The data for the HDI index is obtained from various Human Development Reports. The construction of HDIF is based on data from World Health
G Model
ARTICLE IN PRESS
Armenia
Austria
Azerbaijan
Belarus
Belgium
-.5 0 .5 1
HDIF-HDI
-.5 0 .5 1
Albania
1980
2000
2010
Bulgaria
-.5 0 .5 1
Bosnia and Herzegovina
1990
1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
health
Graphs by country
(i)
Romania
Russian Federation
Serbia
Slovak Republic
Slovenia
-1 -.5 0 .5
HDIF-HDI
-1 -.5 0 .5
Portugal
1980
2000
2010
Sweden
-1 -.5 0 .5
Spain
1990
1980
1990
2000
2010 1980
1990
2000
2010
Year
Graphs by country
edu
health
(ii)
Fig. 5.2. Differences in health and education between ow and stock measures in Europe and Central Asia.
G Model
ARTICLE IN PRESS
Dominica
Dominican Republic
Ecuador
El Salvador
Grenada
0 .1 .2 .3 .4
HDIF-HDI
0 .1 .2 .3 .4
Costa Rica
1980
2000
2010
Guyana
0 .1 .2 .3 .4
Guatemala
1990
1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
health
(i)
Graphs by country
Honduras
Jamaica
Mexico
Nicaragua
Panama
.1 .2 .3 .4 .5
HDIF-HDI
.1 .2 .3 .4 .5
Haiti
1980
2000
2010
Peru
.1 .2 .3 .4 .5
Paraguay
1990
1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
Graphs by country
health
(ii)
Fig. 5.3. Differences in health and education between ow and stock measures in Latin America and the Caribbean.
G Model
ARTICLE IN PRESS
10
St. Vincent and the Grenadines
Suriname
Trinidad and Tobago
Uruguay
Venezuela
.05
.25
.05
.1
.15
.2
HDIF-HDI
.1
.15
.2
.25
St. Lucia
1980
1990
2000
2010 1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
health
Graphs by country
(iii)
Fig. 5.3. (Continued ).
The analysis above focuses on the general time-trend of

health and education in shaping the difference between the
HDIF and HDI by country. However, not all countries experience a smooth or consistent time-path. Some countries
are worth noticing as they have experienced chaotic uctuations in terms of the relative performance in ow and
stock measures in health or education dimensions. This
again highlights the superiority of the HDIF compared to
the HDI, as the latter does not have the sensitivity to capture
the wide swings in the ow measures of contemporaneous
dynamics.
The HDIF and HDI gap in the educational dimension
is measured by the difference between gross enrolment
index (HDIF) and the education index (HDI, anchored on
the literacy rate), while the gap for the health dimension is
between the under-ve mortality rate and life expectancy.
For middle or low human development countries, a widening gap most likely means the ow measures are growing
at a relatively fast rate, hence indicating that an effective
development policy is being implemented. In contrast, a
convergence of the gap is probably a signal of problems in
that nation. In countries such as Papua New Guinea, Bosnia
and Herzegovina, Montenegro, Dominican Republic, Haiti,
Afghanistan as well as Sierra Leone, they have experienced
uctuations in terms of the educational dimension. When
it comes to the health dimension, countries have exhibited
a stable performance over the time period studied, perhaps reecting both the fact that health is higher on the
agenda of most governments than education, and world

aid is much more ready to intervene or assist when civilian
lives are at stake, thus acting as a built-in stabilizer.
The largest drop in the educational dimension can be
found in 1990 Haiti. A possible explanation for this could be
the political instability caused when the Haitian President
declared a state of siege in January of 1990, and the ensuing military coup. This most likely had a negative impact
on social policies, leading to the negative performance in
education. With the support of the West, Haiti has moved
toward a democratically elected government, which has
devoted more toward positive human development programs and thus a higher level of enrollment ratio in more
recent years.
The Dominican Republic and Afghanistan have both
experienced constant uctuations in their education
dimension, but for different reasons: economic performance for Dominican Republic, and political instability
for Afghanistan. The economy of the Dominican Republic
depended heavily on sugar exports. In the 1980s, falling
world sugar prices and reduced U.S. consumption of cane
sugar combined with reduced import quotas imposed by
the U.S. government dragged down their sugar earnings.
This sugar crisis, along with the debt crisis that swept
through many Latin America countries during the 1980s,
hit the Dominican Republic hard. The poor economic performance led to a sharp decrease in spending in education
and thus a decline in enrolment ratio. As a condition for aid,
G Model
ARTICLE IN PRESS
11
Bahamas
Bahrain
Djibouti
Egypt
Iran (Islamic Republic of)
-.2 0 .2 .4 .6
HDIF-HDI
-.2 0 .2 .4 .6
Algeria
1980
1990
2000
2010 1980
1990
2000
2010
-.2 0 .2 .4 .6
Iraq
1980
1990
2000
2010
Year
edu
health
Graphs by country
(i)
Qatar
Saudi Arabia
Syrian Arab Republic
Tunisia
United Arab Emirates
-.5 0 .5 1
HDIF-HDI
-.5 0 .5 1
Oman
1980
1990
2000
2010 1980
1990
2000
2010
-.5 0 .5 1
Yemen
1980
1990
2000
2010
Year
edu
health
Graphs by country
(ii)
Fig. 5.4. Differences in health and education between ow and stock measures in the Middle East and North Africa.
G Model
ARTICLE IN PRESS
12
United States of America
HDIF-HDI
.05
.1
.15
Canada
1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
Graphs by country
health
Fig. 5.5. Differences in health and education between ow and stock measures in North America.
the International Monetary Fund (IMF) imposed austerity

programs in 1991 which helped the economy to recover,
and in turn, led to improved education performance.
Both Pakistan and Sierra Leone also experienced multiple uctuations in education and both can be directly
linked to domestic political issues. For Pakistan, education historically favored the wealthy. In the 1990s, under
the leadership of Prime Minister Benazir Bhutto, improvements in education were one of her prime efforts, and it
was evident in the rise of the gap in 1995. However, this

was hampered with her exile and the military coup by General Pervez Musharraf. The improvement did not happen
before Musharraf lost popularity in the last few years of
our sample, resulting in the increase in the HDIF and HDI
gap in the educational dimension. Sierra Leone suffered
from lengthy civil wars for the most part of our sample
period. The increased education gap in 1995 came after
some form of normalcy was achieved in the election of a
Bangladesh
Bhutan
India
Maldives
Pakistan
0 .2 .4 .6
HDIF-HDI
0 .2 .4 .6
Afghanistan
1980
1990
2000
2010 1980
1990
2000
2010
0 .2 .4 .6
Sri Lanka
1980
1990
2000
2010
Year
Graphs by country
edu
health
Fig. 5.6. Differences in health and education between ow and stock measures in South Asia.
G Model
ARTICLE IN PRESS
13
Chad
Comoros
Democratic Rep. of the Congo
Equatorial Guinea
Eritrea
-.2 0 .2 .4 .6
HDIF-HDI
-.2 0 .2 .4 .6
Central African Republic
1980
2000
2010
Gabon
-.2 0 .2 .4 .6
Ethiopia
1990
1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
Graphs by country
health
(i)
Rwanda
Sao Tome and Principe
Senegal
Seychelles
Sierra Leone
0 .2 .4 .6
HDIF-HDI
0 .2 .4 .6
Nigeria
1980
2000
2010
Sudan
0 .2 .4 .6
South Africa
1990
1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
health
Graphs by country
(ii)
Fig. 5.7. Differences in health and education between ow and stock measures in Sub-Saharan Africa.
G Model
ARTICLE IN PRESS
14
Togo
Uganda
United Republic of Tanzania
Zambia
Zimbabwe
0
.8
0
.2
.4
.6
HDIF-HDI
.2
.4
.6
.8
Swaziland
1980
1990
2000
2010 1980
1990
2000
2010 1980
1990
2000
2010
Year
edu
health
Graphs by country
(iii)
Fig. 5.7. (Continued ).
civilian government and the signing of the Abidjan Peace

Accord. This, however, was short-lived and the chaos of
the brutal civil war tore the country apart. It was not until
a renewed UN mandate and full intervention by the United
Kingdom in this former colony and a member of the British
Commonwealth that the civil war nally ended in 2002, and
so there is a subsequent widening in the education gap.
Many other countries also have stories to tell regarding
their patterns in the educational dimension, these include,
though not limited to, Democratic Republic of Congo, Central African Republic, etc. As stated earlier, the reader can
easily examine the HDIF and HDI gap in the education
or health dimension and link them to the social-politicaleconomic issues of any nation to decipher the story behind
the convergence or divergence of these gaps. One purpose
of this study is to provide a more sensitive measurement of
the human development components to generate a greater
interest in more detailed study of each country.
As one last example of such case studies, let us consider the curious case of Papua New Guinea (PNG). PNG
is richly endowed with natural resources, but difcult terrain, hostile land tenure system, and the most ethnically
diverse nation in the world, with 820 different languages,
combined with mismanagement of the government, especially the deciency in public investment in education, we
saw the convergence of the HDIF and HDI gap in terms
of education. In March 2006, the HDI report downgraded
PNG from a developing country to a least-developed country. This, perhaps, served as a wake-up call, as an IMF
evaluation report in late 2008 nds that a combination
of prudent scal and monetary policies, and high global
prices for mineral commodity exports, have underpinned

Papua New Guineas recent buoyant economic growth and
macroeconomic stability.
Part of the problem with the PNG government was the
electoral system, often resulting in a win by chance rather
than by the will of the people. Limited preferential voting
was instituted and the rst election under this system is
held June/July of 2007 where close to 70% of the PNG voters
voted. It is not unreasonable to speculate that the incumbent ruling party under the leadership of Prime Minister
Sir Michael Somare will pre-emptively implement sound
policy to improve its probability of winning the election;
Somare and the National Alliance Party did win the election. If this is the case, then this could explain the sudden
spike in the education dimension gap in 2007.
If this point stands further scrutiny by readers more
familiar with PNG, it could serve as the poster child for
the relevancy of the proposed HDIF and point to the inadequacy of the HDI index. The UNs downgrade of PNG in
2006 must be due to the HDI measure. Given that the educational dimension of the HDIG is gross enrollment rate,
the spike in the education gap in 2007 reects a signicant
increase in primary school enrollment. This, however, will
have minimal effect on the literacy rate. Thus, even with
a robust GDP growth rate, the stable education and health
dimension will not cause the HDI to increase much, at least
when compared to the more sensitive HDIF. Indeed the HDI
relative to HDIF was 0.536 versus 0.628 in 2006 and 0.541
to 0.781 in 2007. As can be seen, the HDIF increased over
24% in one year, yet the HDI only increased less than 1%. In
terms of ranking, based on the HDI, PNG was ranked 134 in
G Model
ARTICLE IN PRESS
15
Table 2
Difference between HDI and HDIF rankings in 1980, 2000, and 2007.
Year
Obs.
Mean
S.D.
Min
Max
1980
2000
2007
114
160
170
0
0
0
9.12
11.46
12.12
36 (Oman)
64 (Montenegro)
52 (Antigua and Barbuda)
26 (Madagascar)
34 (Botswana)
55 (Botswana)
Note: (1) Only nations with both HDI and HDIF measures are included. (2) The ranking for the HDI is obtained from various Human Development Reports.
The construction of HDIF is based on data from World Health Organizations Statistical Information System, UNESCO and Human Development Reports.
2006, but dropped two slots to 136 in 2007. Yet, based on

the HDIF, the 2007 ranking moved up 31 spots from 159 in
2006 to 128 in the following year.
4.2. HDIF rank and HDI rank by country
Be it in government, business or in academia, that the
right leadership and correct policies can make a tremendous difference in a very short matter of time. The example
of Papua New Guinea above is potentially such a case. The
effect of the policy may be picked up very quickly in the
ow-variable based HDIF, but it will take much longer for
the stock-variable based HDI to catch on. Thus, it may
be interpreted that the HDIF is a short-run measure of
human development, while the HDI is a longer-run measure; though there is still room for debate on this matter.
Nevertheless, a direct comparison of the HDI ranking
versus that of the HDIF may provide certain insights, as it
may be an indication of the short- versus long-run development trends. Table 2 summarizes this rank gap for 1980,
2000, and 2007.8 By denition, the rank difference must
add to zero. However, the variance of the gap seems to be
widening. We have also listed the countries that exhibited the worst performance in the three sample points.
Oman dropped 36 spots if we switch from HDI to HDIF as a
measure of human development in 1980. Montenegro and
Antigua & Barbuda exhibit even larger slips in ranking in
2000 and 2007, respectively. For the positive performances,
Madagascar raised 26 spots in the HDIF ranking. No simple
feat, until we observe Botswana in 2000 and 2007. Based on
a relatively stable political system and the most transparent nancial sector in Africa, Botswanas economy has been
solid, averaging a nine percent GDP growth rate between
1966 and 1999,9 PPP adjusted per capita GDP rose from $70
in 1966 to $16,400 by 2013.10
In the other two dimensions of human development,
Botswana has also made great strides. The adult literacy
rate rose from 69% in 1999 to 83% in 2008. Of course,
this implies that the ow measure must be higher than
8
In case the reader did not immediately recollect why there was a dramatic increase in observations (countries), in 1990s we saw many new
countries declare independence with the fall of the USSR. For example,
in the 1990s, the former Yugoslavia became ve independent countries,
and now it is moving toward eight separate nations. In addition, many
poorer nations lacked data in the early years of the sample period, this
also contributed to the larger sample size in latter years of our sample.
9
Botswana was granted self-governance by the British in 1964, and
gain independence in 1966. The growth rate is comparable both in terms
of duration and magnitude to the economic miracle of China of late.
10
In many cases poor political governance is linked with national culture, and public corruption seems to have the most impact on human
development (Sims, Gong, & Ruppel, 2012; Wu & Zhu, 2011).
the stock value. However, given that the stock variable

(the adult literacy rate) is high, the HDIF and HDI gap are
expected to converge (Fig. 5.7). In the health dimension,
Botswana has also made signicant gains, though there is
still room for improvement. The life expectancy at birth
was 49 in 2002 but improved to 55 in 2009. This sharp rise
in such a short period dictates a sizable improvement in the
health care system of Botswana. This implies that the ow
measure of the health dimension must be growing significantly faster than the stock measures. This can be seen in
Fig. 5.7 by the increasing gap in the health dimension of
Botswana.
The human development progress of Botswana is no less
impressive than that of China, yet it has not received the
recognition it deserved. The location and size of Botswana
is one reason, but perhaps the slow moving HDI measure
is also partially to blame. With our proposed HDIF, the
true performance of Botswana is revealed. The essence of
the Botswana government and her people should be commended. The structure and policies of Botswana should
serve as a model for the poor developing nations, especially
in sub-Saharan Africa and South Asia, perhaps more so than
that of China. Indeed, the ability to identify the short run
effectiveness of government policies is the main purpose of
the proposed HDI. The case of Botswana to be is self-evident
to this end.
We now turn to the dis-aggregate, nation-by-nation
illustration. Fig. 6(i) shows the difference between the HDI
and HDIF ranking of selected countries in 1980. Countries
were categorized into high, medium, and low human
development, based on their HDI scores as shown on the
horizontal axis is in line with the methodology used in
HDRs. Clearly, if the rank differential is positive for a particular country, it means their ow performance (HDIF) is
better than that of stock (HDI).
As can be seen in the graph, there is a slight negative
gradient. This implies that, ceteris paribus, HDI ranking is
higher than the HDIF ranking for the high human development nations, while the reverse is true for the low
human development. This is as we predicted and, indeed,
the main motivation for the development of HDIF concept,
as the effects of good policies for these low development
countries is overshadowed by the stock measures in the
HDI. This pattern is robust and repeated in Fig. 6(ii) and
(iii) for 2000 and 2007, respectively. If we cross reference
with Table 2, the gainers and losers can be clearly seen.
Again, the performance of Botswana stands out.
As one nal comparison, we contrast the rank differential for the three levels of human development across
our sample. This is presented in Table 3. The conclusions
derived from the graphic observations are reafrmed here.
For the high human developing nations, the HDI rank tends
G Model
ARTICLE IN PRESS
16
Fig. 6. Difference between HDI and HDIF rankings in 1980, 2000, and 2007. Note: If the difference is positive, the HDIF ranking is better than the HDI
ranking and vice versa.
Source: (1) The ranking for the HDI is obtained from various Human Development Reports. The construction of HDIF is based on data from World Health
Organizations Statistical Information System, UNESCO and Human Development Reports. (2) Only selected countries are shown in the gure.
to be higher than that of the HDIF. For the medium human

development countries the HDIF rank is slightly higher,
while for the low human development group the HDIF
ranking is signicantly higher than their HDI ranking.
Two additional trends can be observed. First, for the
high and medium human development groups, the rank
differential has been roughly stable. However, this is not
true for the low development group. There seems to be an
increase in the HDI and HDIF rank gap over the sample. If
our hypothesis is true, this indicates that the governments

of the low human development nations have implemented
sound policies in our sample period, as a result, their HDIF
rankings have risen though the HDI did not respond as
quickly.
The second trend relates to within group variation.
For this, we once again look at the standard deviation to
mean in absolute value ratio. Taking the overall average,
for the high human development group this ratio is 8.80,
Table 3
HDI ranking-HDIF ranking by level of human development.
High human development
Medium human development
Low human development
Year
Mean
S.D.
Max
Min
Mean
S.D.
Max
Min
Mean
S.D.
Max
Min
1980
1985
1990
1995
2000
2005
2006
2007
1.27
0.69
0.63
1.98
2.07
1.42
1.16
1.2
9.72
10.56
8.46
12.93
13.31
11.19
12.34
12.99
22
26
28
30
21
29
27
25
24
22
20
60
64
57
57
52
0.32
0.22
0.44
0.33
0.68
0.57
0.37
0.61
9.70
10.71
12.48
13.85
11.44
12.90
12.65
12.79
21
20
21
28
34
53
56
55
36
43
46
28
24
28
23
21
0.48
0.94
1.46
1.61
2.34
2.00
2.42
1.96
7.54
7.34
7.98
7.81
5.91
4.56
4.84
5.23
26
29
23
20
15
11
9
9
13
13
12
10
12
5
8
12
Average
1.34
11.79
30
64
0.30
12.16
56
46
1.59
6.67
29
13
Note: (1) Only nations with both HDI and HDIF measures are included. (2) The ranking for the HDI is obtained from various Human Development Reports.
The construction of HDIF is based on data from World Health Organizations Statistical Information System, UNESCO and Human Development Reports.
G Model
ARTICLE IN PRESS
a much higher than 40.534 the middle group, while for

the low development group this ratio is only 4.19. This
implies, relatively speaking, the rank difference for the low
human development nations are not only positive but also
more stable. If we look at within group time trend, there
lack discerning trends for the high and medium human
development nations. For the low development group, this
ratio is clearly declining. In 1980, this ratio was 15.51, but
dropped to a mere 2.67 in 2007. This suggests that not
only of the government implementing sound policies, but
they are also doing so in a consistent manner. Yet, they
did not seem to get the credit that they perhaps deserve.
As for what caused this trend of better governance among
the poorer nations is certainly worthy of further and more
comprehensive inquiry.
Unlike most other disciplines in the social sciences,
economists tend to place too much condence in the market mechanism and its self-adjustments, while neglecting
the role government and institutions play in the process (Hou, 2011a, 2011b). This is especially true for the
lower developed nations, and it is exactly what this paper
attempts to demonstrate and raise awareness of.
5. Conclusions
This paper develops a new way of measuring human
development (the HDIF) to capture the pure ow of human
development in the areas of wealth, health, and education. Human development, as measured by the HDIF,
performs better than the current HDI, which comprises
both stock and ow variables. By examining the differences between the HDIF and the HDI over time, we
showed that the HDIF and the HDI tend to converge for
wealthy countries and diverge for poor countries, especially those with low HDI rankings. The diverse pattern
can be attributed to the initial level of development and
to the transfer of wealth to non-income dimensions. For
high human-development countries, which started from
higher levels for both health and education, convergence
is a sign that the ow variables have reached the upper
bounds for those dimensions, a 100% enrolment ratio and
a 0.04% mortality rate.
However, the main objective of the HDIF measure is to
show the inadequacy of the current stock-ladenned HDI
method in reecting the positive improvement of political governance of the lower human development nations.
As we show, using the ow oriented HDIF measurement,
the ranking of these countries improves on average. In
contrast, the high human development nations exhibited
a downgrade in their rankings. This suggests that they
may be resting on their laurels, as the stock variables are
cumulative averages of their past success, while the superior performance of the low development countries were
not getting the recognition they deserve. As most readers
17
would agree, the high development nations need no further rights, while the low development countries need all
the encouragement we can plausibly provide. The proposed HDIF establishes a logical platform to gauge such
positive governance, and perhaps can lead to a positive
reinforcement for the continued development of these
nations.
References
Hu, A., & Hibel, J. (2013). Educational attainment and self-rated health in
contemporary China: A survey-based study in 2010. The Social Science
Journal, 50(4), 674680.
Boer, L., & Koekkoek, A. (1993). Human development report: Fad or xture? Development Policy Review, 11, 427438.
Desai, M. (1991). Human development: Concepts and measurement. European Economic Review, 35(2:3), 350357.
Ephrenesis, D. (1994). Policy uses of HDI: Goals and strategies. SIDA Planeringssekretariatet.
Hopkins, M. (1991). Human development revisited: A new UNDP report.
World Development, 19(10), 14691473.
Hou, J. W. (2011a). The role of the state in structural transition and economic crisis. Social Science Journal, 48(1), 112.
Hou, J. W. (2011b). Economic reform of China: Cause and effects. Social
Science Journal, 48(3), 419434.
Hou, J. W., & Li, K. (2011). The aging of the Chinese population and the cost
of health care. Social Science Journal, 48(3), 514526.
Hsu, Y.-C. (2013). The efciency of government spending on health: Evidence from Europe and central Asia. Social Science Journal, 50(4),
665673.
Ivanova, I. F. J. (1994). Effects of the human development indexs social and
economic components on country rankings. New Brunswick, Canada:
University of New Brunswick. Mimeographed.
Kelly, A. C. (1991). The human development index: handle with care.
Population and Development Review, 17(2), 315324.
Khatib, H. (1994). The human development index as a policy and planning
tool. A paper prepared for the Human Development Report Ofce. New
York: UNDP.
Li, S., & Lin, S. (2011). The size and structure of Chinas government debt.
Social Science Journal, 48(3), 527542.
Lozano, S., & Gutirrez, E. (2008). Data envelopment analysis of the human
development index. International Journal of Society Systems Science,
1(2), 132150.
Man, J., Zheng, X., & Lang, T. (2011). Taxation and economic performance:
Evidence from China. Social Science Journal, 48(3), 553559.
Mazumdar, K. (2003). A new approach to human development index.
Review of Social Economy, 61(4), 535549.
Noorbakhsh, F. (1998). A modied human development index. World
Development, 26(3), 517528.
Sagar, A. D., & Najam, A. (1998). The human development index: A critical
review. Ecological Economics, 25, 249264.
Sims, R., Gong, B., & Ruppel, C. (2012). A contingency theory of corruption:
The effect of human development and national culture. Social Science
Journal, 49(1), 9097.
Stewart, F., Ranis, G., & Samman, E. (2006). Human Development: Beyond
the HDI (QEH Working Papers qehwps135). Queen Elizabeth House,
University of Oxford.
Streeten, P. (1994). Human Development: Means and Ends, the American
Economic Review. In Papers and Proceedings of the Hundred and Sixth
Annual Meeting of the American Economic Association (pp. 232237),
vol. 84, no. 2.
Trabold-Nbler, H. (1991). The human development index: A new development indicator? Intereconomics, 26(5), 236243.
UNDP (1990), Human Development Report, United Nations Development
Programme.
Wu, Y., & Zhu, J. (2011). Corruption, anti-corruption, and inter-county
income disparity in China. Social Science Journal, 48(3), 435448.

The Dynamics of Human Development Index

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Dynamics of Human Development Index

Uploaded by

Copyright:

Available Formats

G Model

SOCSCI-1196; No. of Pages 17

Contents lists available at ScienceDirect

The Social Science Journal