ECONOMIC AND SOCIAL COUNCIL. Second Meeting of the Advisory Board. Addis Ababa, Ethiopia ECONOMIC COMMISSION FOR AFRICA UNITED NATIONS

Transcription

1 Distr.: Limited UNITED NATIONS ECONOMIC AND SOCIAL COUNCIL ECA/FSSDD/AB/PAE/00/2 23 October 2000 ORIGINAL: ENGLISH ECONOMIC COMMISSION FOR AFRICA Second Meeting of the Advisory Board on Population, Agriculture and Environment Addis Ababa, Ethiopia October 2000 AFRICA: DEMOGRAPHIC, ENVIRONMENTAL AND AGRICULTURAL INDICATORS, YEAR 2000

2 Africa: Demographic, Environmental and Agricultural Indicators, Year 2000 Introduction (draft) In 2000, African population is estimated to reach over 784 millions. This number, compared to that of 1995, represents an annual growth of 2.4 % during the period Along with the slightly downward trend1 in population growth, the economy of the region performed quite well with some signs of improvement. In 1998, Africa recorded a sustained growth rate of its economy of 3.3% as compared to 2.9% in 1997 and 4.0% in The agricultural sector accounted for 19.4% of the total output in 1997 as compared to 22.3% of 1980, but recorded a consecutive growth of 1.7 and 3.5% respectively in 1997 and Despite all these encouraging figures, there has been little progress in improving the average living standards of the population. Some 80% of low-human development countries -with high population growth rate, low income, low literacy and low life expectancy- are in Africa. Fourty percent of the population still live at or below the line of absolute poverty and the gap between rich and poor keeps on widening. The FAO statistics showed that Africa's per capita agricultural and food production index was 102 ( =100), much lower than the overall production index that was 127. This fact clearly attests that the agricultural performance did not sufficiently keep pace with population. Given disparities among countries, the little change in per capita food production combined also with increasing inequalities in income and food distribution leads to conclude that there has been no change or there has been even an increase in the number of food-insecure and malnourished population in most countries in the past years. There are several factors of the persisting food insecurity of Africa. Among them are: (i) constantly rapid population growth; (ii) low agricultural productivity; (iii) environmental degradation and (iv) weak infrastructure of science and technology. The Food Security and Sustainable Development Division (FSSDD) of the United Nations Economic Commission for Africa (UNECA) has launched a series of activities with the approach that population, environment and agriculture strongly interact to impact on food security and sustainable development in Africa. One of the activities is the PEDA (populationenvironment-development-agriculture) modeling project. This model shows the synergies of those different factors in relation to the goal of ensuring food security and advocates for the holistic approach to development. To support the nexus approach and to help African leaders and development planners make informed policy decisions with regard to food security and sustainable development, the 1 In the precedent five years ( ), Africa recorded an average annual growth of its population of 2.5 %. 2 Source: UNECA, Economic Report on Africa The Challenge ofpoverty Reduction and Sustainability, 1999, Addis Ababa. Ethiopia

3 Division also launched this annual publication on Africa's indicators of population, agriculture and environment. It is the ambition of the Division to make use of its comparative advantages as a division tackling in a comprehensive package the issues relevant to food security and sustainable development - population, agriculture, environment and science & technology - and to cover, by this unique opportunity, the widest range possible of relevant indicators of food security and sustainable development. This publication also aims at providing a quantitative reference tool for analysing the nexus interactions to ail those involved in research and operational activities in the region.

4 Africa: demographic, environmental and agriculture indicators PROPOSED TABLE OF CONTENTS for Africa: Demographic, Environmental and Agricultural Indicators, Year 2000' Introduction I. Background indicators of food security Table 1. Selected background indicators Table 2. Food security indicators Explanatory notes II. Demographic indicators Table 3. Trends in population growth Table 4. Fertility Table 5. Mortality Table 6. Population policies Table 7. Contraceptive uses Table 8. HIV/AIDS prevalence Table 9. Other demographic indicators Explanatory notes III. Agricultural indicators Table 10. Agricultural and food production Table 11. Labour force in agriculture Table 12. Performance of agricultural sector Table 13. Machinery use Table 14. Irrigation Table 15. Fertilizer Explanatory notes IV. Environmental Indicators Table 16. Land area and use Table 17. Forest cover and change

5 Table 18. Water Table 19. Urbanization Explanatory notes V. Science and Technology Indicators Table Explanatory Notes be renamed. * The titles of tables are tentative. According to the data available, they will eventually Annex: Some guiding principles on the Population-Agriculture-Environment (Nexus) analysis

6 ANNEX: Some guiding principles on the Population-Agriculture-Environment (Nexus) analysis In-depth studies on the population, agriculture and environment (nexus) interactions should be carried out. One of the possibilities of these studies is by initialising the PEDA (Population-Environment-Development-Agriculture) model for a particular country and using it for making the medium-to-long term projections by simulating a list of alternative policy scenarios. The model is the unique tool available in the world as of today, specifically designed to demonstrate the complex interrelationships that exist among the nexus issues in terms of attaining food security. To enhance the likelihood of the projection results given by the model and thus provide policy makers a closer-to-reality picture of what will happen in the future given some alternative policies. PEDA preferably requires empirical information that can be collected only by local people. This annex gives a summary of principles related to the PEDA and focuses on data required by the model. Strategies: 1. The first step is to form national expert teams in each country. National teams can include field researchers and experts from various disciplines such as demography, sociology, economics, agriculture, statistics etc. 2. There should be an exploratory phase. In this phase, an inventory of all existing information - including unpublished materials of all sorts - should be formulated. This inventory is a very important tool as it helps assess the data availability situation of the country to draw conclusion on how to organise work to collect further information required. Prior to this process, national governments and institutions should contact the United Nations Economic Commission for Africa (ECA) that will provide them necessary technical advisory services and may eventually raise funds for their activities in cooperation with donors and stakeholders. The exploratory phase will be concluded with a workshop of national teams with ECA and international consultants who produced the model. The workshop will discuss the assessment report of teams and include an intensive training of national experts in programming with Excel and modifying the PEDA software. 3. There should be data collection phase. This can be carried out through: Sending questionnaires to relevant governmental ministries and institutions, NGOs, academic circles Organising sample surveys. 4. There should be PEDA-specific activities phase. Analyse of information gathered through the phase 3 Initialisation of PEDA model with country-specific empirical data Definition of alternative scenarios. Scenarios can be realistic when they are made from the consultations with national planning authorities. Presentation of PEDA simulation results to the high-levels of national governments and to the public at large. Production of national executive reports. Overview on data used in PEDA-: As the PEDA model is a multisectoral model, data used cover a wide range of indicators. They include among others the preparation of the demographic baseline data, the estimation of the agricultural production function elasticities, the estimation of the water saturation curve, the estimation of the land degradation and recovery parameters, the estimation of the food distribution curves, and even the inclusion of new variables that prove to be important in the population, environment, development and agriculture interactions for the particular country. PEDA uses a breakdown of total population of a country by place of residence (urban/rural), literacy and food security status. Each of these eight sub-groups of population is age and sex-specific. Stl: urban, literate, food secure St2: urban. literate, food insecure See also UNECA, PEDA: Technical Manual

7 St3: urban, illiterate, food secure St4: urban, illitrate. food insecure St5: rural, literate, food secure St6: rural, literate, food insecure St7: rural, illiterate, food secure St8: rural, illitrate, food insecure The necessary baseline data include the distribution of the population by sex and single years of age for each of the eight subgroups in the population, the food distribution functions for urban and rural areas separately, fertility and mortality schedules for each of the eight states and the age and sex specific educational transition rates for the illiterate states. The table below illustrates the whole set of variables and parameters used in PEDA model. They are also accompanied with descriptions. Variables/Parameters used by PEDA for the Population-Agriculture-Environment (Nexus) analysis 1. Input variables* Number of population This variable should be ready for all the eight states separately for men and women and by single years of age (0-100). As empirical information found from the national population census doesn't provide the distribution of population by urban/rural areas, literacy and food security status (e.g. stl -8 for men and women), demographic techniques may be necessary to estimate this variable. Here, the expertise and knowledge of people from various disciplines are required to make assumption of literacy rates and proportion of food secure/insecure for all the states. Age-specific mortality rates (ASMRs) Like the number of population, this variable should also be ready for all the eight states separately for men and women and by single years of age (0-100). As empirical information doesn't provide the mortality patterns of all these states, demographic techniques including the MORTPAK software may be used to make life tables for each of the states for men and women. Here, the life expectancy at birth (eo) should be assumed on the basis of the knowledge of the country-specific situation. Age-specific fertility rates (ASFRs) for stl- 8. for females and by single years of age (age 13-55) This variable should be ready for all the eight states for women and by single years of age for the age cohort exposed to fertility event, e.g Like the two variables above, empirical information doesn't provide the fertility patterns of all these states. It may be therefore necessary to make assumption of fertility patterns of each state, based on the overall fertility pattern of the country. Age-specific educational transition rates This variable represents the proportion of girls and boys in each birth cohort that will move over their lifetime from the illiterate to the literate state. The enrollment rate (primary or secondary) may be used. As some countries consider as literate those who ever enter the primary school, it is up to the national expert to decide on the range of age that can be literate. In the present form of the model, these transitions are concentrated around age 10 although more advanced users may specify a particular age pattern for the transition from the illiterate to literate status. Such an age specific pattern for the transition from illiterate to literate status (as opposed to an assumed total proportion to become literate)) will influence the other variables in the model only to a very limited degree. As, like other variables, this information is scarcely available in the real world, assumption is to be made on this variable for all the states.

8 Food distribution function As f^ available in a country is not equally distributed among population, the food distribution function should be assessed separately for the urban and rural areas. It is a Lorenz curve with the cumulated proportion of the population on X-axis vs. the cumulated calories available for distribution on Y-axis. It helps determine the proportion of food secure and food insecure population for both areas. As empirical data on food distribution are hardly found, one may assume that income distribution depicts the same pattern as food distribution and use income data (which could be available from income survey) to draw the food distribution curve. *: Input variables are data required for initialization and relate exclusively to the initial year. In addition to these variables, some others used as scenario variables are also taken as input variables. Assumption made in estimating each variable may be avoided if national experts assess it by conducting sample surveys. 2. Scenario variables or Parameters** Initial year The starting year of the projections. This is the year to which the baseline data apply. This value cannot be changed by the user as it is part of the initialization process. It is only presented as a reference. End of Projection period The end of the projection period. All simulations will be run in single year steps up to that year and all results will be stored up to the end of the specified period. Although there is no direct limit set to the value of this parameter, projection periods of longer than 50 years will slow down calculations and are subject to increasing uncertainty. Production of Kcal per capita in the initial year Refers to the average daily per capita amount of food produced in the starting year. All the agricultural production variables are treated as indexes to increase or decrease upon this initial volume of production. Assumed min kcal per capita to be consumed in order to be food secure In the PEDA model, calorie (energy) requirements are used as a proxy for food requirements. "The minimum energy requirement is the amount ofenergy that is required on average in a population to satisfy the basic physiological needs and the needs for light activities ofadults and the normal energy needs ofchildren and adolescents (including the extra needsfor the growth). Two mainfactors determine the estimation ofthe energy requirement of a population: the distribution by age and sex; and the body ^veight. "* The value of this variable may thus vary under different national conditions, or the user can set the value to define different thresholds to evaluate its effect on the model interactions. Land degradation impact factor Proportion of the cohort moving from rural to urban areas This variable reflects the assumed negative impact of population growth on the natural resource stock. This variable enables the user lo set net rural-urban migration rates. A value of 0.2 means that of every cohort born in rural areas. 20% will move permanently to urban areas over their lifetime. These movements are distributed over the different ages according to standard age-specific migration schedules (see p. *** for more a more precise definition). Sub-Population Parameters Total fertility rale (TFR) Life expectancy at birth This is the mean number of children a woman would get throughout her reproductive life (if she survived to age 50 and were exposed to the age-specific fertility rates observed or assumed for a given year). This is the average number of years a newborn infant can expect to live under current mortality level. It can be set for men and women separately for each of the eight sub groups. This definition is taken from the African Nutrition Database Initiative (AND1) web site, at

9 Educational transition rate HIV/AIDS morbidity rate The number entered in the model for this variable represents the proportion of girls and boys in each birth cohort that will move over their lifetime from the illiterate to the literate state. The enrollment rate (these are usually the one that is used) can be set for men and women separately and is treated as a constant variable over time. In the present form of the model, these transitions are concentrated around age 10 although more advanced users may specify a particular age pattern for the transition from the illiterate to literate status. Such an age specific pattern for the transition from illiterate to literate status (as opposed to an assumed total proportion to become literate)) will influence the other variables in the model only to a very limited degree. HIV/AIDS morbidity rates can be set in a user specified scenario variable that adds an age specific mortality pattern to the mortality schedules for each of the eight subgroups. Just as mortality, HIV/AIDS is treated dynamically but unlike life expectancy, this special agespecific mortality pattern can be specified only for the population as a whole. Variations in the impact of HiV/AIDS on the different subpopulations can only be accounted for through setting different life expectancy trends in these subpopulations. In the HIV/AIDS scenario variable, the user is expected to set an HIV/AIDS related morbidity pattern over time that is defined in terms of proportions of the whole young adult population, aged (similar to the most frequently used HIV prevalence data). A value of 0.15 for this variable means that 15 per cent of the young adult population is assumed to be sick to an AIDS related complaint. In any population this proportion tends to be lower than the HIV prevalence rate because the period of morbidity tends to be short than the incubation period. Agricultural production variables Size of the rural labour force Literacy of the labour force Land Fertilizer use Endogenously determined variable Endogenously determined variable Endogenously determined variable The amount and productivity of fertilizer used in agriculture. The user needs to set a value that expresses a relative improvement/worsening with respect to the conditions in the starting year. Machinery use Technical education Water Irrigation The amount and productivity of machinery used in agriculture. The user needs to set a value that expresses a relative improvement/worsening with respect to the conditions in the starting year. In addition to literacy, the user can give value to specific technical capacity of the rural labour force for agricultural purposes. As with Fertilizer and machinery use, one needs to give a value that expresses a relative improvement/worsening of the conditions in a particular year as compared to the starting year. This covers the change in weather/climate conditions. Its initial value depends on the climate conditions of the country at the time of initialisation. Assuming period in which water is lower than I the initial case, means simulating droughts, if it is higher it means better rainfall, if it is very high it means flooding. This covers the efforts made in irrigation as compared to the starting years, including such things as pipelines, pumps, energy etc. (Reservoir capacity is set with another parameters at a lower level, i.e. directly in Excel) Other scenario variables influencing the availability of food Loss in transport and storage Urban bias factor Individuals will not consume all the food that is produced. Some of the food will be lost during the treatment of the food, the transport or storage. This variable enables the user to take these effects into account. This variable enables the user to allocate food disproportionally between urban and rural areas. A value of 1.0 means distribution according to population size, 1.2 means 20% more to the urban areas.

10 Food imports and exports This variable allows the user to take food imports and exports into account. As ail other variables influencing the availability of food it is treated as an index that has value 1 for the starting year of the projections. Scenario variables are not prepared at the initialization step. They are defined by users of the model at the time of simulation. 3. Output variables*** Available food Births Current land Deaths Life expectancy (eo) Literate Life Expectancy (LLE) Fertilizer Food import/exports Food production HIV/AIDS morbidity rates Irrigation Loss in transport/storage Machinery Proportion food insecure Total population Technical education TFR Urban bias factor Water This is the sum of the total amount of food produced in the country in a particular year minus the loss of food in the harvest, transport and storage, +A food imports and exports. This indicator is only available for the country as a whole. Total number of births The combination of the quantity and quality of land for each year of the projection period. Index value summing up the effects of land degradation and regeneration. This indicator is only available for the country as a whole. Total number of deaths When requested for each of the eight sub-groups separately and sex specific, this is the graduated value of the assumptions of the user. At the country level, however, it is also influenced by the relative weights of the subgroups in the population (provided that different life expectancies have been set for the different subgroups). Number of years a person is expected to live in a literate status from the age of 15 onwards.. This is the sum of the total amount of food produced in the country in a particular year. This indicator is similar to food availability, but it does not account for loss in the harvest, transport and storage or food imports or exports. This indicator is only available for the country as a whole. In addition to the population size that can be generated by urban/rural place of residence. literacy status and food security status; the model has an extra output variable that portrays the proportion food insecure in the country for any year of the projection period. This indicator is only available for the country as a whole. Population size. It can be generated for each of the eight subpopulations separately and for both sexes separately. There is also a possibility to extract age specific information from the databases (see ****). When requested for each of the eight sub-groups separately, this is the graduated value of the assumptions of the user. At the country level, however, it is also iniluenced by the relative weights of the subgroups in the population (provided that different fertility rates have been set for the different subgroups) * * use EWM in stead?*" Output variables give simulation results of alternative policy scenarios.