Transcript Burchi Presentation 2016.HD Measurement

Human Development
Measurement
Francesco Burchi
[email protected]
Master HDFS
1
Table of Contents





Measuring Human Development
dimensions
Human Development Index: goal,
components and aggregation
procedure
New HDI
Inequality-adjusted HDI
Gender Inequality Index
Master HDFS
2
How to measure HD?


Education is a core dimension
Educational indicators should be divided in:
1.
Input indicators (quantity and quality)
2.
Output indicators (quantity and quality)
3.
Outcome indicators (education-related
functionings) which are result of both
qualitative and quantitative educational
inputs and outputs
4.
Impact indicators
Master HDFS
3
Input indicators







Public expenditure in education;
Private expenditure in education;
School resources;
Teachers/students ratio,
Class size and instruction
Teaching material;
Quality and adequacy of curriculum
Master HDFS
4
Output indicators
Measures of ACCESS:
 Enrolment rate;
 Attendance rate;
 Dropout rate,
 Repetition rate
Master HDFS
5
Outcome indicators
(functionings)




Completion rates (in between output
and outcome indicator);
Expected number of completed years
of schooling
Literacy rates;
Standardized test measures of student
and adult achievement in terms of
literacy and numeracy
Master HDFS
6
Impact indicators


Effects on other capabilities such as
nutrition and health?
Effects on economic production,
personal earnings or economic
growth?
Master HDFS
7
The Human Development
Index (1990)


It was elaborated following strong
criticisms towards GDP growth
“Any measure that values a gun several
hundred times more than a bottle of
milk is bound to raise serious questions
about its relevance for human progress”
(ul Haq 1995, p. 46).
Master HDFS
8
Old and New HDI




It is a composite indicator that cannot reflect
the whole Human Development Approach
It has the main objective to shift the focus
from the means of development to the ends.
The HDI was introduced to cover both social
and economic choices.
A composite index was constructed rather
than a plethora of separate indices.
Master HDFS
9
Old and New HDI (2)


One of the most important decisions was
to keep the coverage and methodology
of HDI quite flexible – subject to gradual
refinements as analytical critiques
merged and better data became
available.
The old HDI is available for 177
countries, the new HDI 187.
Master HDFS
10
HDI: Which components?



Components should reflect basic capabilities,
which are those universally accepted and
without which people are harmed (Fukuda
Parr 2003, 97-99).
Problem of operationalization: only
functionings.
How many components? Problem of
multicorrelation, double-counting and
simplicity.
Master HDFS
11
HDI: Components (2)


Final components:
1. Long and healthy life
2. Knowledge
3. Decent standard of living
The first two are ends, the third is a means.
It is a proxy for all other variables not
reflected in the first two components.
Master HDFS
12
HDI: Components (3)


Other components? Long
debate…environment, mortality rates,
political freedom…
The index should be taken with caution:
choice of dimensions is also based on
data availability (Sen).
Master HDFS
13
The OLD HDI
Master HDFS
14
Units of Measurement


Each component is measured by one or more
variables, which have different units of measurement.
Standardization:
ActualValue  MinValue
MaxValue  MinValue


Evolution over time: before the minimum was the
existing minimum value, now most extreme result in
the previous 3 decades or forecasted in next 3
decades.
Each component has a value >=0 and <=1.
Master HDFS
15
Component 1: Long and
healthy life



Variable: life expectancy at birth
Unit of measurement: years.
Standardization:
CountryLifeExp.  25
Life Expectancy Index=
85  25
Max= 85
Min=25
Master HDFS
16
Component 2: Knowledge
The dimension “knowledge” is measured by a
weighted mean of the following variables:
- Adult literacy rate (weight 2/3)
- Combined gross enrolment rate (weight 1/3). It
was added later.
 The GER is calculated by expressing the number
of students enrolled in primary, secondary and
tertiary levels of education, regardless of age, as a
percentage of the population of official school age
for the three levels.

Master HDFS
17
An example
Albania 2002
85.3  0
 0.853
Adult Literacy Rate 
100  0
69  0
 0.69
Gross Enrolment Index 
100  0
Education Index = 2/3 (0.853) + 1/3 (0.69) = 0.798
Minimum = 0
Maximum = 100
Master HDFS
18
Component 3: Decent
standard of living


GDP per capita is adjusted with PPP (purchasing
power parity) and with the logarithm because
achieving a decent standard of living does not
require unlimited income
It is measured in USD
log( CountryGDP)  log( 100)
GDPIndex 
log( 40,000)  log( 100)
Minimum=100
Maximum=40,000
Master HDFS
19
Aggregation



Weights: there is no evidence regarding
which capability is more important, thus
each of the three dimensions has a weight
equal to 1/3.
Aggregation procedure: simple mean.
HDI=1/3(Life Expectancy Index)+
1/3(Education Index)+ 1/3(GDPIndex)
Master HDFS
20
HDI tables



HDI Value (between 0 and 1)
HDI Ranking
Division of countries in 3 groups:
- High Human Development Countries (HDI>= 0.800)
- Medium Human Development Countries
(0.800>HDI>=0.500)
- Low Human Development Countries (HDI<0.500)
Master HDFS
21
Proposals of alternative HDI





Standard of Living: “median income” in place of
“log per capita GDP” (Brazilian HDR)
Knowledge: school attendance rate instead of
enrolment rate (American HDR)
Inclusion of inequality-adjusted functionings
(Hicks, 1997; Foster et al, 2005).
Aggregation methods different from simple
arithmetic mean.
Context-based indicators
Master HDFS
22
The NEW HDI
Master HDFS
23
HDR 2010 & 2014: changes in the HDI
(http://hdr.undp.org/sites/default/files/hdr14_te
chnical_notes.pdf )
Goalposts are changed: max and min are not
observed, but set on a theoretical basis. Min. mean
years of schooling =0 (societies can subsist without
formal education), max.=15 (projected maximum
of this indicator for 2025.
2. The Knowledge Component is now measured by
two variables (with equal weight): A) Mean years
of schooling; B) Expected years of schooling
- Critique to the binary nature of adult literacy
1.
Master HDFS
24
HDR 2010 & 2014: changes in the HDI (2)


Mean years of schooling of adults (years) = average
number of years of education received by people aged 25
and older in their lifetime based on education attainment
levels of the population converted into years of schooling
based on theoretical durations of each level of education
attended (Barro and Lee, 2010)
Expected Years of schooling of children (years) =
number of years of schooling that a child of school
entrance age can expect to receive if prevailing patterns
of age-specific enrolment rates were to stay the same
throughout the child’s life (UNESCO Institute for
Statistics, 2010).
Master HDFS
25
HDR 2010 & 2014: changes in the HDI (3)
3.
PPP-adjusted per capita GNI replaces PPP-adjusted
per capita GDP in the component “decent standard
of living”: this is because GDP is the monetary
value of goods and services produced in a country,
GNI is the income accrued to residents of a
country, including international flows such as
remittances and aid, and excluding income
generated in the country but repatriated abroad.
- Moreover, natural logarithm replaces that with the
base of 10.
Master HDFS
26
HDR 2010 & 2014: changes in the HDI (4)
4.
Geometric rather than arithmetic mean as
aggregation method for the final HDI. Poor
performance in any dimension is now directly
reflected in the new HDI, which captures how well
a country’s performance is across the 3 dimensions.
No longer perfect substitutability across the
dimensions: a low achievement in one dimension is
not anymore linearly compensated for by high
achievement in another dimension. Penalizes
countries with uneven development across
dimensions!
Master HDFS
27
The New HDI
Master HDFS
28
Formula for the new HDI
(see technical notes HDR 2014)
Education Index = (Mean years of schooling + Exp. years of schooling)/2
New HDI = Indexlife1/3 · IndexEducation
Master HDFS
1/3
· IndexIncome
1/3
29
HDI Vs. GDP






GDP focuses on the average income owned, while HDI tells us
also about the DISTRIBUTION and the USE of income for
valuable purposes.
GDP cannot be adjusted following people’s diversity (disability,
age, sex, metabolism = conversion factors).
E.g. having weapons is much more valued than having milk
due to monetary terms using only GDP.
HDI shows directly the areas where performances are
eventually low, thus where interventions should be made.
HDI can be disaggregated, by gender, ethnicity, region ….
HDI can also tell us about future economic growth because
having accumulated human capital (education and health) can
lead in the future to the enlargement of economic possibilities.
Master HDFS
30
HDR 2015
HDI 2015 Table 1:
http://hdr.undp.org/en/composite/HDI


Master HDFS
31
The Inequality-adjusted HDI
Master HDFS
32
Inequality-adjusted HDI (2010) (see notes HDR 2014
and http://hdr.undp.org/en/faq-page/inequalityadjusted-human-development-index-ihdi#t293n91 )



The IHDI adjust the HDI for inequality in the
distribution of each dimension across the
population
IHDI=HDI if there is no inequality in all the
dimensions (thus, IHDI<=HDI)
Data source different from HDI data source
because we need information on the
distribution of life expectancy, schooling, and
disposable income/consumption.
Master HDFS
33
IHDI: Step 1

Measuring inequality in each of the 3
dimensions
Geometric mean of the distribution
Arithmetic mean of the distribution


Ax >=0 because geometric mean cannot be
higher than arithmetic mean
The higher the difference between the two
means, the higher inequality is.
Master HDFS
34
IHDI: Step 2

Adjusting the dimension indices for
inequality
Inequality measure
Dimension Index

The inequality-adjusted income index, I*IIncome, is
based on the unlogged gross national income (GNI)
index, I*Income.
Master HDFS
35
IHDI: Step 3

Aggregation through geometric mean
Master HDFS
36
The Gender Inequality Index (2010)


Replaces the Gender-related Development
Index and the Gender Empowerment
Measure
Main drawbacks of past indicators:




Unreliable data especially for genderdisaggregated income
Mix of absolute and relative variables
Main focus on gender bias in elites and urban
areas (in the case of GEM)
Computed for 140 countries
Master HDFS
37
GII: main purpose
To highlight women’s disadvantage in 3
dimensions:
1.
Reproductive health
2.
Empowerment
3.
Access to labour market
Master HDFS
38
GII: Variables
(http://hdrstats.undp.org/en/indicators/24806.html)



Reproductive health: maternal mortality ratio
and adolescent fertility rate (number of births
per 1,000 women age 15-19 years)
Empowerment: share of parliamentary seats
by sex and attainment at secondary or higher
levels
Labour market: labour market participation
rate (Percentage of the working-age
population (ages 15–64) that actively
engages in the labour market, by either
working or actively looking for work)
Master HDFS
39
GII calculation


see technical notes HDR 2010:
http://hdr.undp.org/en/media/HDR_2010_EN
_TechNotes_reprint.pdf and
http://hdr.undp.org/en/statistics/gii/
Master HDFS
40
Policy use of indicators




GDP Index, HDI, IHDI, GII (and previous GEM and GDI) should be
analyzed together and in a historical perspective in order to
understand overall characteristics of a country and to come out
with better development policies.
Comparison across indicators is of easier interpretation with
rankings. A good tool is calculating the difference in a country’s
rankings in two different indicators.
E.g. Support-led or growth-led approach? Sustainable HD or not?
Gender biased policies? Some cultural or religious aspects are a
constraint or a positive engine for HD?
Geographical uniformity? Regions reflect similar human
development levels and trends?
Master HDFS
41
Further tools


HDI Rank – IHDI Rank to measure the loss
of position of a country with the inclusion of
inequality in the 3 dimensions in the
measure of development.
(HDI* – IHDI)/HDI* to measure the
percentage reduction of the IHDI from the
HDI* (i.e, the HDI without logarith of GNI).
Master HDFS
42
PPP
Purchasing power parity in US$ The purchasing
power of a country’s currency, defined as the
number of units of that currency required to
purchase the same (or very similar)
representative basket of goods and services
that a US dollar (the reference currency)
would buy in the United States.
Master HDFS
43