1. POPULATION PROJECTIONS
Session 8 - Projections for sub-national and sectoral populations
Ben Jarabi
Population Studies & Research Institute
University of Nairobi 1

2. The need for sub-national projections
Sub-national structures need projected population to quantify likely need and to help plan for services
National governments need projected pop. to allocate resources throughout the country
Ministries need projected population to implement and monitor programmes
The private sector may be interested in the growth of specific age groups and in the growth of households, both of which drive consumption 2 2 2

3. Sub-national projections
Subnational projections - use either a cohort-component procedure or one of a number of simpler, less data-demanding methodologies
Cohort-component projection, which requires all of the inputs of national projections plus internal migration by age and sex
Mathematical or ratio projection, with or without age-sex detail 3 3 3

4. Sub-national projections
Because of the data and computational requirements involved, we typically see, and is generally recommended that the cohort-component projection be applied at no lower than first subnational administrative level
Ratio projection suffices at lower than first subnational administrative levels 4 4 4

5. Sub-national projections
Generating sub-national projections that are both internally consistent and consistent with a national projection is usually more challenging than preparing a national projection
Each region presents the same data problems as the national projection but, in addition, preserving consistency across regions and dealing with data problems that are often more severe than those at the national level adds to the challenge 5 5 5

6. Sub-national projections
A national projection can be generated as the sum of a series of sub-national projections, or a national projection may be prepared first, followed by sub-national projections with the region with the largest population serving as residual
For a given country, a hybrid of procedures may be used. For example, projections for major regions can be combined into a national projection and can serve as separate control totals for provincial or district projections 6 6 6

7. Sub-national projections
Data Requirements
Census age-sex structure and
For cohort-component projection:
Fertility and trend
Mortality and trend
Internal migration and trend
International migration, or
For mathematical or ratio projection:
Trend in population totals 7 7 7

8. Sub-national projections
Begin with base year age-sex structure
Adjust for coverage if possible
Adjust for consistency of under-10 population with fertility and mortality
Smooth at ages 10+ if need be
Ensure that regions add to national totals by age and sex
Resulting age-sex structure provides initial population for forcing base-year consistency between fertility and mortality 8 8 8

10. Sub-national projections
Cohort-Component Method - Summary
This method is widely used, relatively easy to explain, and practical
It permits the use of already available data and existing theoretical knowledge on the dynamics of population growth, and it takes into account causal factors, at least at the level of basic components and compositional factors 10 10 10

11. Sub-national projections
Cohort-Component Method - Summary
It has the capability to produce consistent & comparable national & subnational projections that are easy to update on a regular basis
Much of the work required to use this method lies in the in-depth analysis and development of assumptions for each of the components of change
However, it also has its shortcomings and limitations, e.g. it does not explicitly incorporate socioeconomic determinants of population change 11 11 11

12. Sub-national projections
Ratio Method - Summary
This method suffers from several shortcomings
They do not account for differences in demographic composition or for differences in the components of growth
They provide little or no information on the projected demographic characteristics of the population 12 12 12

13. Sub-national projections
Ratio Method - Summary
This method suffers from several shortcomings
Because they have no theoretical content, they cannot be related to theories of population growth, except perhaps the logistic model, which is consistent with a Malthusian view of population dynamics
Consequently, they have limited usefulness for analyzing the determinants of population growth or for simulating the effects of changes in particular variables or assumptions
In addition, they can lead to unrealistic or even absurd results, even over relatively short horizons 13 13 13

14. Sectoral projections
Projections of households, school enrollment, poverty, employment, health, and other population-related characteristics are needed for many types of planning, budgeting, and analysis - for simplicity, these are referred to as socioeconomic projections
Because of the demand for socioeconomic projections and their close link to projections of basic demographic characteristics, the former are often made on the basis of the latter 14 14 14

15. Sectoral projections
Projections of socioeconomic characteristics, however, has two important features that distinguish them from strictly demographic projections
One, some socioeconomic characteristics are directly affected by policy decisions - e.g. enrollment is usually dictated by entrance requirements
In such instances, knowledge of public policy is essential to the production of projections of socioeconomic characteristics 15 15 15

16. Sectoral projections
Two, projections of socioeconomic characteristics involve achieved characteristics - those that can change over one’s lifetime, e.g. marital status, income, educational attainment, occupation
As a result, projections of socioeconomic characteristics involve a variety of assumptions in addition to those for projections of strictly demographic characteristics 16 16 16

17. Sectoral projections
Two fundamental approaches are frequently used to prepare socioeconomic projections
Participation ratio method
Cohort-progression method 17 17 17

18. Sectoral projections Participation ratio method
In this approach, socioeconomic characteristics are related to demographic characteristics through the use of ratios
Current and historical data are used to construct participation ratios – i.e., proportions of the population (stratified by age, sex, and perhaps other demographic characteristics) that have the socioeconomic characteristic of interest 18 18 18

19. Sectoral projections Participation ratio method
Once such ratios are established, they can be projected in a number of ways, e.g. holding them constant at recent levels, extrapolating recent trends, or tying them to ratios found in other areas
The projected ratios are then applied to population projections for the geographic area(s) under consideration to obtain a set of socioeconomic projections 19 19 19

20. Sectoral projections Cohort-progression method
In this approach, projections are developed by “surviving” people with particular socioeconomic characteristics
The numbers with the socioeconomic characteristic or the corresponding participation ratios are projected on a cohort basis using information on changes in the numbers or participation ratios between two previous dates
The conventional form of this method uses ratios of the number of persons aged a with a particular socioeconomic characteristic in year t to the number of persons aged a - y with that characteristic in year t - y 20 20 20

21. Sectoral projections Cohort-progression method
It is important to remember that cohort progression ratios represent net cohort change rather than gross change
This distinction is important because fundamental patterns may be masked without knowing the numbers “entering and exiting” a population
The cohort-progression method in the form of participation ratios is used less often than the version of the method that employs absolute numbers 21 21 21