Title: MDG Labour Indicators: Measurement, availability and discrepancies of data
1 MDG Labour Indicators Measurement,
availability and discrepancies of data
- MDG 3.2 Share of women in wage employment in the
non-agricultural sector
ILO Dept. of Statistics
2Introduction
- ILO data gathering mechanisms
- Definitions
- Data sources
- Data availability at international level
- Possible sources of discrepancies
- Treatment of missing values (use of proxy
indicators and imputations) - Regional and Global estimates
- Future challenges
3ILO data gathering mechanisms
- Annual standardised questionnaire, websites, NSP
- Questionnaires pre-filled with the statistics
reported in the previous years (last 9 years) - Meta data collected as well
- Consistency checks, validations, footnoting
- Clarifications with the countries
- Dissemination (YB of Labour Statistics,
http//laborsta.ilo.org/, KILM) - Clear international standards, ILO Resolutions
4ILO data gathering mechanisms
- Data collected on EAP, Employment, Unemployment,
Hours of Work, Wages, CPI, Occupational Injuries,
Strikes and lockouts. - Disaggregation by sex, age, economic activity,
occupation, education, status in employment
(where relevant) - These row data are used to calculate various
indicators, incl. MDG
5MDG labour indicators
- MDG1 Eradicate extreme poverty and hunger
- Target 1B Achieve full and productive
employment and decent work for all, including for
women and young people - Employment-to-population ratios for persons
aged 15 and youth (15-24) by sex - Vulnerable employment rate by sex
- Working poverty rate
- Labour productivity growth rate
- MDG3 Promote gender equality and empower women
- Share of women in wage employment in the
non-agricultural sector
6MDG labour indicators
All five employment indicators (under MDG1b and
MDG3) are explained in detail in the Guide to the
new Millennium Development Goals Employment
Indicators, which is available at http//www.il
o.org/public/english/employment/docu/index.htm Th
e Guide includes definitions, data sources,
calculations and analytical examples. The Guide
also includes the full set of Decent Work
Indicators, which allows for comprehensive
monitoring of decent work.
7Employment-to-population ratios for persons aged
15 and youth (15-24) by sex
- The EPR measures the proportion of a countrys
working age population that is employed - EPR Total employment/working age population
100 - Source of data labour force survey or other
household survey/population census with data on
population and employment - There is no single correct employment-to-populat
ion ratio, but national EPRs are typically
between 50-75 per cent - Disaggergation by sex, age - to identify possible
imbalances and dependency rates, and changes over
time
8Vulnerable employment rate for persons aged 15
by sex
- Indication about the proportion of employed that
are in the more vulnerable statuses of employment - Vulnerable employment rate (number of
own-account workers number of contributing
family workers)/total employment 100 - The rate is typically related to level of GDP per
capita - Source of data labour force survey or other
household surveys with data on status in
employment
9Vulnerable employment rate, selected countries
ranked by GDP per capita
10Working poverty rate for persons aged 15
- The working poor are defined as employed persons
living in a household whose members are estimated
to be below the nationally-defined poverty line - Working poverty rate working poor/total
employment 100 - Source of data household surveys with both
employment and income expenditure data if data
on employment and poverty come from different
sources, an approximation may be used to arrive
at the working poverty rate
11Growth rate of labour productivity
- Labour productivity represents the amount of
output achieved per unit of labour input - Labour productivity GDP measured at constant
market prices in national currency/ total
employment - Labour productivity growth rate is measured as
the annual change in GDP per person employed - Labour productivity can be used to assess to
which extent the economic environment allows for
the creation of decent employment opportunities - Source of data labour force survey or other
household survey/population census with data on
employment, in combination with data from
national accounts
12Share of women in wage employment in the
non-agricultural sector
- Measures (i) the degree to which women have
equally access to paid employment, (ii) the
degree to which labour markets are open to women
in industry and services sectors (iii) the
flexibility of the labour market and the
economys capacity to adapt to changes over time. - Does not (i) cover employment within the
agricultural sector, (ii) reveal any differences
in the quality of the different types of non-ag.
wage employment regarding earning, conditions or
work, or the legal and social protection which
they offer. - The indicator is calculated by dividing the total
number of women in paid employment in the
industrial and service sectors by the total
number of people in paid employment in that same
sector.
13Share of women in wage employment in the
non-agricultural sector ESCWA-averages by country
14Share of women in wage employment in the
non-agricultural sector ESCWA-trends over time
15Data sources at the national level, and their
limitations
- Labour Force Surveys
- Establishment surveys
- Official estimates
- Administrative records
- Insurance records
- Censuses
- Other household surveys
16(No Transcript)
17Problems of comparability across countries and
over time within countries
- Methodological and conceptual differences
definitions, coverage of the reference
population, coverage of the sectors,
classifications used, sources, etc - (e.g. only public sector, excl. enterprises with
less than 5 employees, excl. informal sector,
etc) - international comparisons difficult
18Data availability by country
19Data availability by year
20ESCWA Bahrain, Egypt, Jordan, Palestine,
Morocco, Oman
21Estimated values for MDG 11
- Estimations based on auxiliary variables
- - Total paid employment
- - Total employment in non-agriculture
- - Employees
- - Total employment
- - Economically Active Population in
non-agriculture - Sensitivity analysis conducted on a selected
number of countries there is strong correlation
between the indicator and the auxiliary variable.
22Data availability (use of proxy indicators)
-
- 10 countries do not provide data but the
information on the economically active population
is used instead as a proxy.
23Sources of discrepancies between national and
global data
- different sources,
- different series from the same source,
- changes in the definitions and classifications
over time (in the same source), - estimates when the national data are not
available for a particular year, - imputations.
24An illustrative example
25An illustrative example
26An illustrative example
27Multiple series/sources
- Where data from multiple sources are available,
the selection of the most appropriate one is
based on a number of criteria, incl. - consistency of concepts, definitions and
classifications with the international standards, - quality of data,
- availability of data/source over time, etc.
28Conceptual variation
- Discrepancies may also exist because of different
definitions and classifications. - for employment status, especially for part-time
workers, students, members of the armed forces,
and household or contributing family workers - classifications over time
- geographical and population coverage
incl.changes over time).
29Treatment of missing values
- Imputations for missing values is unavoidable in
any aggregation process. - Assuming that, if there no data, the value of the
indicator is zero results in biased regional and
global estimates - Imputations
- Implicit assuming the value of the indicator is
the same as the average for the countries with
available data - Explicit (i) carry forward the last observed
value (ii) use the value of the indicator for a
country with similar characteristics, (iii)
predict the value by statistical modelling
30Treatment of missing values in MDG 3.2
- In process of producing regional and global
aggregates for MDG11, ILO uses a methodology for
explicit imputation for missing values - The sole purpose of these imputations is to
produce the regional and global aggregates and
may not be best-fitted for national reports. - The national imputations are best produced
through methodologies that take directly into
account the local specificities of the country
concerned.
31Modelled values for MDG 3.2
- Separate two-level models developed for each
region. The models take into account - between-countries variation over time,
- within-country variation over time.
- Predicted values are based on the assumption that
the data that are available for a given country
are representative of that countrys deviation
from the average trend across time in its region.
32Modelled values for MDG 3.2
- 5 different models developed and their properties
tested. - The data available for the latest year omitted
from the dataset and imputed by using different
models. The modelled data then compared with the
actual observed values. - The quality of the modelled data assessed based
on several criteria (i) mean deviation, (ii)
standard deviation, (iii) maximum positive and
negative deviations. - .
33Modelled values for MDG 3.2
- The quality of the predicted values
- is proportional to the number of years for which
the indicators is available - depends on the quality of the observed values for
a given country and the quality of the data for
the corresponding region. - ? Careful checking is required (outliers, unusual
trends, sources, etc.)
34(No Transcript)
35Observed, estimated and modelled data for MDG 3.2
- Methodological descriptions of the series
disseminated available on the ILO Dept. of
Statistics website. - The estimated values based on proxy indicators
are disseminated on the MDG website. All values
which are estimated are clearly identified. The
modelled data are not disseminated as their sole
purpose is to produce the regional and global
aggregates. - The ILO is making its methodology for imputing
missing values in the process of producing
regional and global aggregates publicly
available.
36Future work
- The ILO will continue to work with countries and
other partners to - (a) enhance the national statistical capacity of
countries to produce the data needed for
estimating the indicator - (b) develop national analytical capacity to
produce good quality imputed country values for
use by countries in their monitoring of the MDGs
and other dev.programmes - (c) ensure that all data available at national
level are collected in a way that will be of
least burden to countries. - (d) Cooperation by the countries much needed.