The Gender Development Index and Gender Empowerment Measure:

An Analysis of Strengths and Limitations in The Context of South

Africa
By: Lauren Milne

Abstract - Research in the area of domestic and international development has historically relied upon
development indicators as a means to determine and rank countries with respect to levels of poverty.
This thesis provides an analysis of two such indicators. Specifically, the Gender Development Index (GDI)
and the Gender Empowerment Measure (GEM) are analyzed to assess their strengths, limitations and
accuracy in determining the wellbeing of women in relation to poverty in the context of South Africa.
These indices differ in terms of their purpose. The GEM is an indicator in measuring female
agency (ability, independence) while the GDI focuses on human development measures. Both indices
penalize the male score accounting for gender inequality and include the estimated earned income
variable. The analysis concludes that these indices are inaccurate and misleading as a means of
measuring the wellbeing of women in the context of South Africa.
INTRODUCTION
!
In the study of International Development, various indicators have been developed in order to
assess and measure levels of poverty and development. Some examples of these indicators include the
Gross Domestic Product per capita, the Physical Quality of Life Index and the Human Development
Index. Over the years, development indicators have been formulated to address current development
issues, re-formulated when the indicators do not measure what was intended and abandoned altogether
when their application is no longer accurate or relevant.
"
This paper discusses the strengths and limitations of two development indices the Gender
Empowerment Measure (GEM) and the Gender Development Index (GDI) with respect to the wellbeing of
women. More specifically, it will examine what the GEM and GDI can highlight with regard to the
wellbeing of women in the context of a middle-income country. South Africa is used as a case study for
the purpose of this analysis because it is a middle income country, and recent analyses from Andy
Sumner indicate that that, theres a new bottom billion meaning three-quarters of the extreme poor,
or up to a billion poor people, live not in the poorest countries but in middle-income countries. This
raises all sorts of questions about which countries need aid, the link between aid and poverty (Villarino,
2011, para. 5). In addition, South Africa has among the best national statistics and data in Africa.
!
The analysis focuses on the equalities the indices reveal and its limitations. Specifically the
thesis will answer the following question: What do these indices actually reveal and conceal when
applied to the measurement of the wellbeing of women?
!
CONTRIBUTION TO THE STUDY OF GDI AND GEM
"
There is a gap in the current state of the literature with respect to the GDI and GEM and the
wellbeing of women. There are general critiques on this subject but more detailed analysis about their
limitations or inadequacies is absent from the literature. This gap will be addressed by examining what
the GDI and GEM indicators reveal and conceal about the wellbeing of women in South Africa. Further,
there has been no attempt to meaningfully understand the contextual implications of these
measurements within a developing country. This has critical implications regarding the utility, accuracy
and relevance in the reporting of womens wellbeing.

29

In addition, there is research required on deconstructing gender measurements to analyze how

and what they measure. This is relevant in gender studies and development because it is necessary to
evaluate whether the measures that account for womens wellbeing are socially and culturally
appropriate. The GDI and GEM still use classical economic methods for evaluating income inequalities,
which can distort the real circumstances occurring.
ACCOUNTING FOR GENDER INEQUALITIES IN DEVELOPMENT
"
An important step toward a holistic approach of addressing poverty has been made through
indices created by the United Nations Development Programme (UNDP). The Human Development Index
(HDI) appeared in 1990 based on the premise, that people and their lives should be the ultimate criteria
for assessing the development of a country, not economic growth (Chant, 2003, p. 11). While the HDI
does not take into account a gender component, gendered aspects of poverty such as income and life
expectancy have been used to calculate UNDPs Gender-related Development Index and Gender
Empowerment Measure. The GDI and GEM theoretical foundations branch out of the HDI premise based
on a holistic approach in measuring poverty with a gender perspective by including unique features such
as the participation of females in top managerial and government positions in the calculations. The GDI
and GEM were created in order to move away from neo-liberal economic development thinking and put
human needs at the core of development plans. Human development indicators (education, life
expectancy) are included in these indices. However, the income component still remains from previous
development thinking and can conceal certain gender inequalities.
"
Producing an internationally comparable measure of the wellbeing of women is important and
useful for several reasons. First, the governments of countries shown to have high levels of gender
inequality may face pressure to reduce it. Second, these indicators help advance understanding of the
relationship between gender and general welfare (Dijkstra & Hanmer, 2000). The UNDPs pioneering work
on GDI and GEM have drawn attention to gender inequality in development policy debates, and drawn
the attention of academics to the problem of measuring this inequality.
EXPLANATION AND CALCULATION OF THE INDICES
"
The human development approach developed in response to increasing criticism of conventional
perspectives of the 1980s that associated national economic growth and the expansion of individual
choices. There was a growing body of evidence that did not support the prominent belief of the trickle
down effect surrounding the power of the economy to spread its benefits to end poverty. The work of
Sen and others provided the conceptual foundation for an alternative and broader approach defining
human development, as a process of enlarging peoples choices and enhancing their
capabilities (Kovacevic, 2010, p. 1).
"
In order to address the different level of capabilities between males and females, Anand and Sen
focused on the concept of an aversion to inequality meaning that, if two societies had the same
average achievement but different levels of inter-group inequality in that achievement (men and women)
aversion to inequality would mean that the society with the lower inter-groups inequality should be
socially preferable to one having the same average achievement but large inequality (Bardham &
Klasen, 1997, p. 2). In other words, if country A and country B have similar scores on the HDI, but
country B has less inequality between men and women, it is socially preferable for women to live in this
country than country A.
"
Based on this notion of aversion to inter-group inequality, the GDI adjusts for the gender gaps to
penalize countries HDI for the existing inter-group inequality between males and females. Since the
HDI ignores inter-group differences and assumes that everyone in a country has reached the average
achievement in the three components of the index (life expectancy, education and income) the
adjustment for gender disparity will lower the GDI relative to the HDI (Bardhan & Klasen, 1997). The
extent the GDI score will differ from the HDI will depend on the size of the gender gap in each of the
30

three variables used to calculate the overall index, combined with penalty applied to the all three male
variables.
!
While the HDI measures average achievement, the GDI adjusts the average achievement to
reflect the inequalities between men and women in the following three dimensions: a long and healthy
life measured by life expectancy at birth, knowledge as measured by the adult literacy rate and the
combined primary, secondary and tertiary gross enrolment ratio and a decent standard of living
measured by estimated earned income (UNDP, 2008). !
!
For a further explanation of GDI calculation, the indices used, and results see Appendix C and D.
!
"
The GEM and the GDI were developed and adopted by the UNDP. However, they differ with
respect to their theoretical foundation and calculation; the GDI focuses on the loss of human
development due to gender gaps and the GEM measures womens agency to access certain resources
and their roles within society (Bardhan & Klasen, 1997). The GEM measures womens political
participation and decision-making power measured by, womens and mens percentage shares of
parliamentary seats (UNDP, 2008, p. 360). The second feature includes womens economic participation
and decision-making power measured by two indicators. The first indicator measures womens and
mens share by percentage of positions of legislators, senior officials and managers. The second
indicator evaluates womens and mens percentage shares of professional and technical positions. The
final indicator measures womens access to economic resources by assessing womens and mens
estimated earned income.
"
Bardhan and Klasen argued that certain aspects of GEM are an effective tool in evaluating
opportunities that are open to women when they state that, a society that neglects the economic and
political potential of half its population is likely to perform worse than a society that draws on all its best
talent, regardless of gender(1997, p. 21).
"
There are several fundamental elements similar to the GEM in the calculation of the GDI formula.
Similar to the GDI, the GEM also penalizes all of the three male indicators for aversion to inequality to
adjust for the already present inter-group inequalities between males and females. GEM and GDI
calculate estimated earned income in the exact method by using the, ratio of the female nonagricultural
wage to the male nonagricultural wage, male and female shares of the economically active population,
total female and male population and GDP per capita (UNDP, 2008, p. 360). For a further detailed
explanation of calculation see Appendix A and B.
SOCIO-ECONOMIC HISTORY OF SOUTH AFRICA
"
In order to understand the implications of utilizing the GEM and GDI with respect to womens
wellbeing in South Africa, it is helpful to discuss the historical background of the country.
!
The main cause of widespread poverty is due to the apartheid regime that institutionalized
racism across every level of society including the access to healthcare, land, employment, education
and political rights. Several laws were passed that systematically reinforced the status quo of the poor
treatment and type of access to services the majority of the population was allowed to receive. The
Group Areas Act passed in 1950, enforced the segregation of the different races to specific areas and
allowed for the two oppressed ethnic groups (Blacks and Coloreds) to only own certain zones of land
(Worden, 2012). In practice, Blacks and Coloreds were forced off their own land, which in many cases
happened to be high quality land due to the type of soil fertility and location, and moved to areas further
from city centers with poor land quality. With poor quality housing, schools and healthcare, combined
with a lack of job opportunities in the areas the Blacks and Coloreds were forced to live, levels of poverty
rose and continue to exist today as these areas or townships struggle to break free from the poverty
cycle.

31

"
Survey results indicate a high level of poverty in South Africa despite legal and political initiatives
implemented to increase the wellbeing of women. According to Hoogeveen and Ozler, there were
approximately 1.8 million (2.3 million) more South Africans in 2000 living with less than $1/day ($2/day)
than there were in 1995. Inequality increased mainly due to a sharp increase among the African
population (2005, p.5).
"
Following the Apartheid regime, the 1996 Constitution was created and was the most
progressive constitution on the continent. The constitution, legalized abortion, giving women equal
power in marriage, banning all gender discrimination and providing women of any skin color with the
same degree of affirmative action in education, employment, politics (The Economist, 2010, para. 2).
"
Theoretically, South Africa has experienced progress in the promotion of womens rights. In
practice, some sectors are progressing faster than others. For example, in the political sphere women
hold 44 percent of parliamentary seats, the third highest proportion in the world and 41 percent cabinet
posts including many of those often assigned to men: defense, agriculture, foreign affairs, mining,
science and technology, and home affairs (The Economist, 2010, para. 3) including many of the cabinet
posts traditionally assigned to men such as defense and home affairs.
"
In other areas of society, the wellbeing of women has not improved as expected. This is
particularly evident in aspects of the public sector but primarily in the private sector. The Employment
Equity Act requires that, companies with over fifty employees, must hire and promote women in
proportion to their representation in the population as a whole at 52%, however in reality white men still
dominate senior management (The Economist, 2010, para. 4). Moreover, according to the Womens
Business Association, a fifth of the countrys private sector boards have no women, and in universities
females are more than half of the graduates, however account for 45 percent of the academic staff (The
Economist, 2010, para. 4). This does not mean that female graduates finding academic positions is not
progress; nonetheless it is a pattern that exists in gender discriminatory countries because teaching is
one of the main traditional employment options for women throughout history.
"
Finally, that women make up almost half of the formal labour force is a positive trend; however, it
is important to look at which roles they play compared to their male counterparts. Women are in lowerwage sectors, particularly domestic service and on average women still get less than two-thirds of a
mans pay packet (The Economist, 2010, para. 5).
"
The level of poverty among women-headed households is higher than the average and women
continue to earn less than men, even though differences education levels have narrowed significantly, as
one can observe in the GDI. South Africas National Planning Commission state, about 61# percent of
women live in poverty, and 31#percent live in destitution, compared with 39#percent and 18#percent of
men respectively (2011, p.2).
CRITICAL ANALYSIS OF THE GDI AND GEM IN THE CONTEXT IN SOUTH AFRICA: LIMITATIONS
OF THE MEASURES
"
The 2007-2008 Human Development UNDP Report provided rationale for valuing the estimated
income in both GEM and GDI and stated that, the income measure in the GDI and GEM indicates a
persons capacity to earn income. It is used in the GDI to capture the disparities between men and
women in command over resources and in the GEM to capture womens economic independence (2008,
p. 361). However, in computing the GDI and GEM, estimated income for females is problematic because
the indicator is based on an estimate of the female share in earned income.
"
Specifically it is a problem of defining what constitutes an economically active population. The
estimated income is calculated from the total female share in the economically active population. For
example, if family enterprises and subsistence activities are not included in defining an economically
active population, many individuals will be excluded, particularly women, as they tend to largely make up
32

this sector of the economy (Dijkstra and Hanmer, 2000). These women are still earning an income they
but are excluded from the definition and will not be accounted for in the calculation resulting in an
inaccurate reflection of the active female labour force within the country. As a result the estimated
income for females will most likely be lower than the real income.
"
Defining the informal economy can range from country to country. Statistics South Africa (Stats
SA) defines the informal sector as: businesses that are not registered in any way. They are generally
small in nature, and are seldom run from business premises. Instead, they are run from homes, street
pavements or other informal arrangements (Devey, Skinner and Valodia, 2006, p. 4). In 20002005,
informal economy workers represented between 31 and 34 percent of South Africas labour force (Devey,
Skinner and Valodia, 2006). This is a significant issue when 34 percent of the population is missing from
the estimated earned income because they do not fit into the formal economy definition.
The relationship between the wellbeing of women and the informal economy is that the,
majority of street traders are black women who trade in a range of goods including clothing, fruit and
vegetables and a labour force survey conducted in 2000 estimated that there are 500,000 street vendors
in South Africa (Consultancy Africa Intelligence, 2013, para. 3). This point is critical in calculating GDI
and GEM because many working women who earn an individual and/or family wage are excluded from
the calculation. It is not known how much the estimated earned income indicator would be effected if
informal female workers were included in the GDI and GEM calculation, however it would be interesting
to observe whether the income gap would decrease.
Initial speculation would suggest that a more accurate reflection of the formal and informal
economy would be represented if the women working in the informal sector were to be included in the
calculations, especially due to the large portion of women contributing to this sector. Statistics from
South Africa on the role that women fulfill in the informal economy suggest that the measurements would
increase and would constitute a more accurate reflection of womens wellbeing. However, it is difficult to
determine whether women in South Africa working in the informal sector experience the same income
gap as their male counterparts in the same sector. Further, a more accurate depiction of this situation
would require the addition of both the male and female informal labour force to the estimated incomes.
The result would be a more accurate representation regarding women and may also reveal similar
income gaps in the formal sector. In addition, the development of a separate indicator for evaluating men
and womens income in the informal sector would be helpful in understanding income inequality
between the genders. This would be a useful measure to observe whether there is a similar income gaps
in the formal sector.
"
Another criticism of the income portion of the GDI and GEM formulas is womens share of
earned income is based on the male/female difference in urban wages rather accounting for rural or the
informal sector (Dijkstra and Hanmer, 2000). Specifically, the calculation only uses a nonagricultural
wage. This is problematic when calculating GDI and GEM in South Africa because the rural agricultural
sector is an important source for employment for women (Schuler, 2007). Hart and Aliber found that,
women represent almost two-thirds of those engaged in some form of agriculture, the majority of these
women do it primarily as a main source or extra source of household food (2012, p. 2). Moreover, only in
commercial farming are the number of women and men roughly equal. South Africas agricultural sector
contributes a large portion to the national gross domestic product, at 12 percent compared to the fishing
industry at 0.5 percent (South African Government Information, 2013). With a population of 51 million
people and 8.5 million employees working in the agricultural sector (South African Government
Information, 2013), it is reasonable to expect a change in the estimated earned income if an average
between agricultural and nonagricultural wages were included in the indicator.
"
In response to criticism regarding crude measurements of calculating estimated earned income,
the UNDP responded that, because data on wages in rural areas and in the informal sector are rare, the
Human Development Report has used nonagricultural wages and assumed that the ratio of female
wages to male wages in the nonagricultural sector applies to the rest of the economy (UNDP, 2008, p.
33

361). It is also important to note that the shares of earned incomes by males and females do not
measure the gaps in consumption at the household level. The GDI was created in similar fashion to the
HDI in that it focuses on a more humanistic approach to poverty, moving away from a strong reliance on
economic growth. However, the GDIs income component is given equal weight to education and life
expectancy still highlighting its strong importance in measuring poverty. Importantly, Bardhan and
Klasen point out that, high earned income does not always translate into high female well-being and
consumption levels (1997, p. 14).
!
When a country has similar male and female education and life expectancy levels even after the
aversion to inequality is applied to the male index, the income indicator will impact the overall country
score more significantly, as in most developing countries income inequality is the most unequal. For
example, countries in the Middle East and North Africa have the largest gaps in earned incomes
between males and females. These regions also have very small gaps in life expectancy and education;
however, due to the large size of the income gap compared to the other indices and how all three indices
have equal value, the one with the most inequality will have a larger effect on the calculated value
(Bardhan and Klasen, 1997). In 1999, Bardhan and Klasen calculated the implied penalties for inequality
for the three indicators in the GDI and observed that the gap in income accounts for 85 percent of the
total gender gap on average (Dijkstra, 2001).
"
A criticism that applies to both the GDI and GEM is that cultural and social elements are
impacted by a womens increase in income, suggesting that a rise in income does not necessarily
translate into a higher level of well-being and empowerment. In Chants research authors Bradshaw and
Linneker stated that by working in maquilas (export processing plants), becoming richer materially may
be accompanied with frustration or demeaning situations for women (2003, p. 24). Also, certain local
cultural values are more important to the women than generating more income, such as modesty,
respect, are not included in GEM or GDI (2003, p. 24), therefore disconnecting what the local women
value and what many development actors believe women should value.
"
A limitation of both GDI and GEM is how all females and males are broadly grouped under the
category of their sex when it is also useful to evaluate how different ages and ethnicities experience
poverty (Chant, 2003, p. 25) which is may range from an elderly indigenous woman to a woman
belonging to a majority ethnicity living in a wealthier city. However, this is not a problem unique to GDI
and GEM because when calculating many measurements of wellbeing and human development these
issues are not considered. Finally, the GDI and GEM are criticized for the general inaccuracy of the data
itself. Most of the data are developed from the national census, which in themselves are rarely a reliable
source of information, prone as they are to sporadic collection, poor enumeration and imprecise
definition of key terms, not to mention gender bias (Chant 2003, p. 23).
STRENGTH OF THE INDICES
!
The major strength of the GDI is the access to education indicator (measured as the literacy and
gross enrolment measure) because of the relationship between higher education levels and improved
employment status, income and autonomy for women. As a result it is one of the most important and
most universal indicators for gender equality (Dijkstra, 2001). Another strength is the access to
healthcare (measured as life expectancy at birth) because it demonstrates discrimination in access to
health services and reflects cultural ideas on men and women (Dijkstra, 2001).
"
In the South African context, the prevalence of HIV/AIDS and its impact on longevity affects
calculations of life expectancy. By the end of 2002, the Department of Health estimated there were 5.3
million people infected with HIV/AIDS and reported that, there is solid evidence of this increased
mortality in South Africa and increased mortality was more marked for women than for men. A woman
between the ages of 25-29 was three and a half times more likely to die in 1999/2000 than she would
have been in 1985 (Whiteside & Lee, 2006, p. 385). This mortality rate affects life expectancy at birth but
also has economic consequences, the responsibility for care of orphans and the sick has largely fallen
34

to women who make up 70 percent of caregivers (Whiteside & Lee, 2006, p. 387). This responsibility is
reflected in the gender gap in the estimated income because being the main caregiver of a household
requires women to stay at home or very small income supplements in the informal sector rather than
finding employment outside of the home.
"
The high rate of domestic violence and sexual abuse is another major factor that impacts life
expectancy and womens wellbeing in South Africa. Jewkes studied gender violence in South Africa
stated that, a women is killed by an intimate partner every eight hours, a probable underestimate
because no perpetrator is identified in 20 percent of killings. That is double the rate of such murders in
the United States (Faul, 2013, para. 6). While South Africa has one of the most progressive constitutions
in the continent regarding womens rights, the current legislation has had minimal effect on protecting
women from domestic violence or encouraging the reporting of incidents.
"
The Inter-Ministerial Committee on Violence against Women and Children found that even when
victims reported cases of domestic violence to police or social workers, [...] their plea for help fell on
deaf ears or they were told to resolve the matter with their partners (Faul, 2013, para. 10). Female life
expectancy may actually be significantly lower if the amount of gender violence, especially when it
results in a loss of life, were to be calculated in the GDI.
"
The main strength in the GEM is that it includes the percentage of females in technical,
professional, administrative, and management positions that are included in this measure. It is an
indicator concerning access to economic resources since these jobs are higher paying jobs. Dijkstra
argued that, access to administrative and management positions reflects to some extent decision
making power in society, while access to technical and professional occupations reflect opportunities for
career development (2001, p. 322). This indicator reflects increased employment opportunities for
women, which measures the extent of female power. The higher the share of women in these positions
the more power women will have on average in the household. Dijkstra also suggests that in comparison
to the number of women elected into parliament, this indicator is less vulnerable to the changing political
climates (2001, p. 322).
"
The female share in parliament is a strong indicator for relative female power in society. In
addition, there is data readily available from many countries regarding the number of females in
parliament. Unfortunately, there is little or no information on female elected officials at regional and local
levels (Dijkstra, 2001). As a result, the accuracy of this indicator to provide the reality for most South
African women is limited to perceptions regarding womens power and wellbeing associated with their
representation in parliament. Chant suggested that female representation in parliament may be important
as it indicates several features of womens wellbeing including gender discrimination in the workplace or
changing attitudes to women decision makers. However, this measure excludes the majority of women
in many societies whose daily duties and lives remain generally unaffected by national politics ad who
own local participation in politics are more likely to be restricted to local levels (2003, p.24).
"
For example in South Africa women's representation in local government has climbed steadily
from 19 percent in after the 1995 elections to 29.6 percent after the 2000 local elections to 40 percent
after the 2006 elections (EISA, 2009, para. 3). Another reason for the rising levels of women in local
politics is due to the African National Congress (ANC) commitment to reaching a minimum quota of 30
percent females within the political party. This trend would suggest the importance of incorporating
involvement in local politics as part of this measure.
GENERAL FINDINGS: GDI AND GEM
Table 1: South Africa Human Development Report 2009: Data from 2007

35

Sex

Life Expectancy at
Birth

Adult Literacy

Combined gross
enrollment ratio in
education

Estimated earned
income
($US, PPP)

Female

53

87.2%

77.3%

$7,328

Male

49.8

88.9%

76.3%

$12,273

Source: (UNDP, 2009).

HDI Rank: 129 out of 182 countries
HDI Value: 0.690
GDI Rank: 109 out of 182 countries
GDI Value: 0.680
Source: (UNDP, 2009).
!
According to the statistics in the above table, the GDI and HDI are almost identical in value that
indicates minimal loss of human development due to gender gaps. However, an important indicator in
the GDI and GEM, namely the estimated earned income, demonstrates a large gender gap in income.
This gap can, in part, explain why South Africa ranks lower on the GDI and HDI apart from having a very
low life expectancy at birth for both sexes. Not only is there a significant gap between males and
females in estimated income that subsequently lowers the overall level of GDI and HDI, the life
expectancy at birth for both sexes is also low ranking, which further decreases the indices.
!
Based on this statistic, one could speculate that the income component does not have an
overriding effect on the calculation as suggested in certain countries when it is an outlier (largest gap out
of all three indicators) that lowers the overall score. In contrast, with South Africa, the life expectancy is
already quite low therefore the income gap does not have a significant impact in producing a low score.
If the life expectancy and education levels were higher for both males and females and the income gap
was large between the sexes, it might be argued that the income gap indicator would decrease the score
significantly.
Table 2: Gender Empowerment Measure: Data from 2007

Sex

Legislators,
Ratio
Seats in
senior
Professional
Estimated
estimated
Ministerial
parliament officials and and technical
earned
female to
positions
held
managers
workers
income
male earned
(% of total)
(% of total) (% of total) (% of total)
($US, PPP)
income

Female

34%

34%

Male

66%

66%

.60

$7,328

45%

$12,273

55%

Source: (UNDP, 2009).

GEM: 26 out of 182 countries
GEM Value: 0.687
Source: (UNDP, 2009).
!
As indicated in the table, South Africa scores fairly high on the Gender Empowerment Measure
despite the fact that the income gap is quite large and females earn 60 percent of what males earn. It
can be concluded that the wellbeing of women is improving in the political sphere yet remains males in
the economic sector. However the income indicator does not have a measurable impact on the GEM.
This is similar to the GDI in South Africa as the income component should decrease the GEM due to the
large inequality gap between the incomes.

36

OVERALL FINDINGS
"
The conclusions drawn from both scores of the GDI and GEM in South Africa is that a large
estimated income gap does not play a significant role in the overall score. In the GEM the income gap
appears to have little bearing on the calculation as South Africa still manages to score quite high. This is
very problematic when applying a measure that evaluates the wellbeing of women within a country
because a country can still score relatively high on GEM while having serious economic inequalities
between men and women. The estimated income gap cannot be ascertained within this calculation
unless the individual variables within the GEM are taken into account. Beteta reported a similar
conclusion in the research on Barbados (2006, p. 223). She stated that, high levels of gender equality in
terms of women in parliaments and in decision-making positions- it is ranked 24 in the 2004 Human
Development Report but where women are still the majority of the poor, female unemployment rates
are higher than those of males (2012, p. 223).
!
By neglecting to account for the three separate variables used in the GEM formula one might
conclude, in the South African case, that women experience low levels of inequality since the country
ranks fairly high out of all the countries ranked. This is not accurate because income disparity, high rates
of domestic violence and the high incidence of HIV/AIDS among women contradict the findings that
imply a high level of wellbeing. In the GDI, the income gap has a minimal impact on the calculation
compared to the GEM because the life expectancy is also fairly low. Due to the fact that two out of the
three variables score low, the overall GDI will decrease.
CONCLUSIONS
"
This paper analyzed the strengths and limitations of the GEM and GDI in measuring the
wellbeing of women in the context of South Africa. The analysis provided evidence that suggests that
the application of these indices is problematic when studying the wellbeing of women for a number of
reasons. Specifically, it was demonstrated by the GEM and GDI scores that a country can experience
significant female income inequalities yet have a high overall GEM score and ranking with respect to the
wellbeing of women. Subsequently, continued application of these indices for the purpose of measuring
wellbeing is questionable since it produces an inaccurate perspective of the reality of women in this
country. Future research in this area should examine, in more detail, the direct effects of these indices in
the overall calculation and how the results can be corrected in order to produce a measure that is more
effective in demonstrating an accurate status of the wellbeing of women in South Africa. In addition,
increased attention should be given to developing additional indices that incorporate different variables
in order to more accurately measure the wellbeing of women. With respect to issue of limited
accessibility for women to participate in national politics, it would be useful to examine how the GEM
would be impacted, when taking into account womens participation in provincial and municipal level
politics.

37

REFERENCES
African National Congress. (2011). Effects of Apartheid on the Status of ! Women in South Africa.
"
Retrieved from! http://www.anc.org.za/show.php?id=4668&t=ES
Bardhan, K. & Klasen, S. (1997). UNDPs Gender-Related Indices: A Critical Review. Retrieved from
http://hdrstats.undp.org/sites/default/files/undp-revisions-gender-related.pdf
Beteta, H.C. (2007). What is missing in measures of Womens Empowerment? Journal of Human
!
D e v e l o p m e n t , 7 . h t t p : / / w w w. t a n d f o n l i n e . c o m . l i b p r o x y. u w i n n i p e g . c a / d o i / p d f /
"
10.1080/14649880600768553
Chant, S. (2003). The engendering of poverty analysis in developing regions: progress since the United
"
Nations Decade For Women, and priorities for the future. Retrieved from h
" ttp://eprints.lse.ac.uk/
!
573/1/theEngenderingOfPovertyAnalysis.pdf
Consultancy Africa Intelligence. (2013). The Informal Sector in South Africa: Women street traders in
!
Durban. Retrieved from !http://www.consultancyafrica.com/index.php?
!
option=com_content&view=article&id=1287:the-informal-sector-in-south-africa-women-street!
traders-in-durban-part-2&catid=91:rights-in-focus&Itemid=296
Devey, R., & Skinner, C., & Valodia, I. (2006). Definitions, data and the informal economy in South Africa:
a critical analysis. In Vishnu Padayachee (Ed.) The development decade? Economic and social
change in South Africa, 1994-2004, (pp 302-323). Cape Town, RSA: Human Sciences Research
Council.
Dijkstra, G. (2002). Revisiting UNDPs GDI and GEM: Towards an Alternative. Social Indicators Research,
!
57. Retrieved from http://www.jstor.org/stable/27526995
Dijkstra, G, & Hanmer, L. (2000). Measuring Socio-Economic Gender Inequality Toward an Alternative to
!
the UNDP Gender-Related Development Index. Feminist Economics, 6. Retrieved from !
!
http://www.tandfonline.com/doi/abs/10.1080/13545700050076106
The Economist. (2010). Women in South Africa: Walking several paces behind. Retrieved from http://
www.economist.com/node/17204625
Electoral Institute for Sustainable Democracy in Africa. (2009). South Africa: Womens representation
"
quotas. Retrieved from http://www.content.eisa.org.za/old-page/south-africa-womens"
representation-quotas
Faul, M. (2013). South Africa Violence Against Women Rated Highest In the World. Retrieved from
!
http://www.huffingtonpost.com/2013/03/08/south-africa-violence-against!
women_n_2837804.html
Hart, T & Aliber, M. (2012). Inequalities in Agricultural Support for Women in South Africa. Retrieved from
!
http://www.hsrc.ac.za/uploads/pageContent/3025/InequalitiesLR.pdf
Hoogeveen, J. & Ozler, B. (2005). Not Separate, Not Equal: Poverty and !Inequality in Post-Apartheid
!
South Africa. Retrieved from http://deepblue.lib.umich.edu/bitstream/handle/2027.42/40125/
!
wp739.pdf
Klasen, S. (2006). UNDPs Gender-Related Measures: Some Conceptual Problems and Possible
"
Solutions. Journal of Human Development, 7. Retrieved from http://
"
www.tandfonline.com.libproxy.uwinnipeg.ca/doi/pdf/10.1080/14649880600768595
Kovacevic, M. (2010). Measurement of Inequality in Human Development- A Review. Retrieved from
!
http://hdr.undp.org/en/reports/global/hdr2010/papers/HDRP_2010_35.p!df
Schuler, D. (2006). The Uses and Misuses of the Gender-related Development Index and Gender
!
Empowerment Measure: A Review of the Literature. Journal of Human Development, 7.
!
Retrieved from !http://www.tandfonline.com/doi/abs/
!
10.1080/14649880600768496#.UnwWZhwcTIs
South African Government Information. (2013). Agriculture. Retrieved from http://www.info.gov.za/
aboutsa/agriculture.htm.
South Africa National Planning Commission. (n.d). Human Conditions Diagnostic. Retrieved from
!
http://www.npconline.co.za/MediaLib/Downloads/Home/Tabs/Diagnostic/
!
Diagnostic_Human_conditions.pdf
United Nations Development Programme. (2013). Calculating the Human Development Indices.
!
Retrieved from http://hdr.undp.org/en/media/HDR_20072008_Tech_Note_1.pdf.
38

United Nations Development Programme. (2011). Measuring inequality: Gender-related Development

!
Index (GDI) and Gender Empowerment Measure (GEM). Retrieved from ! http://hdr.undp.org/en/
!
statistics/indices/gdi_gem/
United Nations Development Programme. (2008). Human Development Report Technical Note 1:
!
Calculating the human development indices. Retrievedfrom http://hdr.undp.org/en/media/
!
HDR_20072008_Tech_Note_1.pdf
United Nations Development Programme. (2009). Human Development Report Gender Empowerment
!
Measure and its Components. Retrieved from http://hdr.undp.org/en/media/
!
HDR_2009_EN_Table_K.pdf
United Nations Development Programme. (2008). Human Development Report Gender-Related
Development Index. Retrieved from http://hdr.undp.org/en/media/HDR_20072008_GDI.pdf
Villarino, E. (2011). Andy Sumner: A voice for the New Bottom Billion. Retrieved from
https://www.devex.com/en/news/andy-sumner/76197
Whiteside, A., & Lee, S. (2006). Economic and development issues around HIV/AIDS. Vishnu
Padayachee (Ed.) The development decade? Economic and social change in South Africa,
1994-2004. Cape Town, RSA: Human Sciences Research Council.
Worden, N. (2012). The Making of Modern South Africa: Conquest, Apartheid, Democracy. West Sussex,
UK: John Wiley & Sons.

39

APPENDIX A:

Figure A1: The indicators and dimensions calculated within the Gender Empowerment Measure (UNDP,
2008, p. 355).
APPENDIX B:
!
The GEM and GDI use the same maximum and minimum values for income earned, however the
values are used in a differently in GEM. The GDI uses the same formula as the HDI to calculate
estimated earned income in order to compare the values, otherwise if a different formula was used it
would be very difficult to draw conclusions from observing the GDI and HDI.
"
Another significant difference in calculating the GDI and GEM in that in the GEM calculation
there is an extra step to compute the equally distributed equivalent percentage (EDEP). The calculated
value for each indicator is divided by fifty and the rationale behind this method, is that in an ideal
society, with equal empowerment of the sexes, the GEM variables would equal 50% that is womens
share would equal mens share for each variable(UNDP, 2008, p. 360). In other words the assumption is
that a 50/50 share of the population for females and males should be the ideal goal in all three indices.
APPENDIX C:

Figure C1: The indicators and dimensions calculated within the Gender Development Index (UNDP,
2008, p. 358).

APPENDIX D:
!
It is important to note the similarities between calculating the GDI and HDI in regards to using
maximum and minimum values in the formula. Each indicator (life expectancy, knowledge and estimated
income) includes a predetermined maximum and minimum value within the formula as well as the actual
value from the specific country. For example the estimated earned income maximum value is $40,000

40

and the minimum value is $100. The main difference in calculating both indicators is that GDI accounts
for the aversion to inequality and penalizes the male calculations for each indicator to make the indicator
more equal to the female and therefore account for gender inequalities. In other words the improvements
in female achievements are given more weight than improvements in male achievements, given females
are at a lower average in the first instance (Dijkstra and Hanmer, 2000).
!
In calculating the life expectancy index, it is based on a given maximum and minimum values
that are different for men and women taking into account on average women live five years longer than
men (UNDP, 2008). The index for knowledge is a weighted average of the achievements on literacy that
makes up two thirds of the formula and primary and secondary enrolment account for one third.
!
The calculation of the GDI is straightforward in that it is the un-weighted average of the three
component indices, the equally distributed life expectancy index, the equally distributed education index
and the equally distributed income index.

41

Copyright of Undercurrent is the property of Undercurrent and its content may not be copied
or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.