MINISTRY OF PLANNING AND INVESTMENT GENERAL STATISTICS OFFICE POPULATION PROJECTIONS FOR VIETNAM

Transcription

1

2 MINISTRY OF PLANNING AND INVESTMENT GENERAL STATISTICS OFFICE POPULATION PROJECTIONS FOR VIETNAM HANOI,

3 ii

4 FOREWORD One of the important input information of policy making and socioeconomic planning work is future population size of the whole country and lower administrative units. Consequently, population projections are normally implemented periodically. However, population projections might also be done extraordinarily due to demands of the information users, particularly at the beginning of a new development period considered a turning-point. It is not an exception in the case of Vietnam that after each Population and Housing Census, population projections are made for many years later, serving the setting-up of long-term, medium and shortterm development programs. The 2009 Population and Housing Census started at 00:00 hours the 1 st April 2009 not only provided valuable and basic data sources for assessing the process of socio-economic plans accomplishment of the years before the Census, but the results of the Census were also the most precise input for population projections. These population projections were deployed right after we had the results of the completed 2009 Population and Housing Census. The projections for the period of used the Census as basic information, including two independent projections: one is for the whole country, urban and rural areas of the whole country; and the other is for centrally governed provinces and cities. On the occasion of releasing the publication of projection results, the General Statistical Office would like to express our sincere thanks to the financial and technical assistance from the UNFPA office in Vietnam in our projection implementation. We also highly appreciate the great effort of the officers in the Department for Population and Labor Statistics who directly or indirectly did related works for this publication to reach its readers the soonest. Despite our greatest effort, this publication cannot avoid mistakes due to the fact that population projection is a difficult job, thus we hope we will receive readers feedback to improve our population projection work in the years coming. GENERAL STATISTICAL OFFICE iii

5 iv

6 TABLES OF CONTENTS Foreword... Table of contents... iii v PART 1: DATA SOURCES, METHODOLOGY AND PROJECTION RESULTS Introduction Main characteristics of the Vietnam 2009 population projections Objectives Scope and period of projections Variants Projection and calculation method Methodology of the Vietnam 2009 population projection Base population Mortality assumptions Fertility assumptions Migration assumptions Projection results for Vietnam Projection results for the whole country Projection results for urban and rural areas Projection results for socio-economic regions Projection results for provinces, cities PART 2: PROJECTION RESULT TABLES ANNEXES Annex 1: Socio economic regions Annex 2: Component method Annex 3: Method of United Nations secretariat v

7 Annex 4: Base population of population projection by sex and administration units Annex 5: Life expectation at birth projection for the whole country, urban areas, rural areas, Annex 6: Life expectation at birth projection for provinces/cities, Annex 7: Total fertility rate (TFR) projection and fertility model for provinces/cities Annex 8: Migration assumptions for provinces/cities REFERENCE Note: In some tables, a total may be not equal to sum of the components due to rounding. vi

8 TABLE IN TEXT Table 3.1: Estimation of overcount and undercount rate of the 2009 census... 7 Table 3.2: Joint score index of the United Nations for the completed data of the 1/4/2009 population and housing census... 9 Table 3.3: Quinquennial gains in life expectancy at birth according to initial level of life expectancy and sex Table 3.4: Total fertility rate (TFR) assumed for projection stages Table 3.5: Fertility model classification, Table 3.6: Sex ratio at birth Table 3.7: Sex ratio at birth projections, Table 4.1: Projected population and average annual growth rate of each stage, four variants, Table 4.2: Sex ratio and median age, four variants, Table 4.3: Comparison between population structure of the base year and the ending year of the report period, Table 4.4: Urban and rural population and annual average annual growth rate projection, four variants, Table 4.5: Projected population and average annual growth rate by region, medium variants, Table 4.6: Projected population by provinces, cities, medium variant, FIGURES Figure 3.1: Compare age specific death rate of Viet Nam with 4 families of North, South, East, West Figure 3.2: Age specific fertility rate (ASFR) by urban and rural areas, Figure 4.1: Average annual growth rate, four variants, Figure 4.2: Viet Nam population pyramid, medium variant, vii

9 viii

10 Part 1 DATA SOURCES, METHODOLOGY AND PROJECTION RESULTS 1

11 2

12 1. Introduction Humans are the objectives and motivations of the development. Thus, population data provided sufficiently, timely and precisely are very necessary for the socio-economic planning. Actual planning work in our country over the recent years reveals that policy makers and administrators need not only population data of the past and present but also those of the future which are provided by population projections. Over the last 3 decades, many population projections have been accomplished. The General Statistical Office, after the 1979, 1989 and 1999 Population Census, has implemented population projections with various detailed level. Vietnam population projections are mainly based on the data of the 2009 Population and Housing Census (from now on will be abbreviated to 2009 Census ) in order to project the population of the next 40 years (in this publication we call it The Vietnam 2009 population projections ). The projections are implemented in 2 independent branches, one is for the whole country, urban and rural areas of the whole country (in this publication from now on will be written as urban and rural projections ); and the other is for centrally governed provinces and cities (from now on will be abbreviated to provinces and cities projections ). 2. Main characteristics of the Vietnam 2009 population projections 2.1 Objectives The projections are to provide data of population by sex and age in future. They also provide data of population changes (fertility, mortality and migration) and some other demographic indicators. Besides, projection result reports would provide some technical aspects of population projections. 2.2 Scope and period of projections - Space: projection results of the whole country, urban and rural areas of the whole country, socio-economic regions and centrally governed provinces and cities (Annex 1: Socio-economic regions). 3

13 - Time: projections are made for a period of 40 years, from 2009 to 2049, including 8 stages of 5 years each: , ,, Variants The projection brings out 4 variants of population change, basing on 4 scenarios of fertility, one scenario of mortality and one scenario of migration. Detail as follows: Mortality: Mortality level: assume that mortality, illustrated by the life expectancy at birth (e 0 ), increases following the models for mortality improvement suggested by the United Nations (revised version 2004), stems from the mortality level estimated from the results of the 2009 Census. Mortality model: assume that mortality models, described by the Age Specific Death Rate (ASDR), based on the North model life tables of Coale- Demeny. Migration: Migration level: assume that migration, described by net migration of the 2009 Census to be used as a basis, with detailed migration assumptions which is stated in Part Migration model: assume that migration models, described by the Age Specific Net Migration Rate (ASNMR), estimated from the results of the 2009 Census, remain constant during the report period. Fertility Fertility level: assume that fertility, described by Total Fertility Rate (TFR), vary with the 4 following scenarios: + Medium variant: assume that the fertility that has varied as observed in the past would continue until it reaches TFR = 1.85 births/woman, that fertility will remain constant at that rate until the end of the projection period. + High variant: formed from medium fertility variant with TFR at each 5-year period higher than those at the medium fertility variant by 0.3 births/woman. 4

14 + Low variant: formed from medium fertility variant with TFR at each 5-year period lower than those at the medium fertility variant by 0.3 births/woman. + Constant variant: Use TFR estimated from the result of the 15% sample of the 2009 Census as a basis for of the period and assumed that this TFR remain unchanged during the period. Fertility model: assumed that fertility models, described by Age Specific Fertility Rate (ASFR), observed from the results of the 15% sample surveys of the 2009 Census, remain unchanged during the period. 2.4 Projection and calculation method In order to get the objectives and demand above, the Vietnam 2009 Projection use the Standard Component Method (refer to Annex 2: Component method) in 5-year age groups with open age of 80 and over (80+). Population distribution by sex and 5-year age groups of the completed 2009 Census is used to be the base population of the projection. Since the census date of the 2009 Census is the 1 st April, so year used in this projection means starting at 1 st April and ending at 31 st March. There are 2 projections implemented independently: one is for the whole country and the urban, rural areas of the whole country; and the other is for the whole country and 63 provinces/cities, where the population projection results by sex and age of the whole country are used as control population, so that population projection by sex and age of urban and rural areas must be totalized equally to those of the whole country; the totality of those of provinces/cities must also be equal to those of the whole country. Projection results of socio-economic regions are the data summed up from such results of the provinces belonging to those regions. The calculation is implemented using PEOPLE 3.01 software (refer to General Statistical Office, Projection results of the whole country, geographiceconomic regions and 61 provinces/cities, Vietnam , Chapter 1, Page 3, Hanoi, 2001). 5

15 3. Methodology of the Vietnam 2009 population projection Population projection is actually to apply mathematical models and normally base on a number of practical data and assumptions. Thus, the precision of projections depends on the precision of the data and the appropriateness of assumptions to the practice. Certainly, projection basing on detailed data sources is expected to be at higher quality than those basing on less detailed data sources. In addition, higher quality data will generate better projection results. Moreover, population projection for a time point in the future must base on assumptions of the changing process of such components that result in the population change including fertility, mortality and migration; and these components also change in some certain trends. Similarly, the less error the initial data used to be the projection s base populations get the better. Normally, demographers who implement population projections will have to generate assumptions and calculate by some various variants. If the projection period gets longer, then the errors found in the projection results might be higher due to the fact that assumptions chosen might not be appropriate for a long time. Reality has shown that, the precision of population projections depends on the appropriateness to the practice of the assumptions given, not on the sophistication of the projection calculating methods. A better technical must firstly point out variables that can be predicted with high probability and can be used for population projection, then can exploit ready data sources for detailed population projections. Simultaneously, the important thing before implementing population projection is that the data used for projection must be evaluated and corrected appropriately the errors, the insufficiency and other appropriateness. In addition, we should notice that population projections for the whole country will be more precise than those for the provinces. So, projections for the provinces need to be adjusted following projections for the whole country independently. The following sections will present each part in details. 3.1 Base population As stated above, population distribution by sex and five-year age groups following the results of the completed 2009 Census is used to be the base 6

16 population for projections. One of the most common types of error in the population censuses is errors in the survey scope due to over-counting or undercounting. The combined effect of the two types of errors in scope of the surveys tends to lead to an over-count or under-count of the population. The level of overcounting or under-counting might differ between males and females, between ages, urban and rural areas and between regions. With age distribution, there are errors due to age reporting, especially age rounding Total population In order to evaluate the level of overcounting or undercounting, the 2009 Census undertook a Post-Enumeration Survey (PES) in 60 enumeration areas (EAs). The samples were designed and EAs were selected randomly in order to represent for the whole country and for every socio-economic region (6 regions). The Central Population and Housing Census Steering Committee has organized survey teams which directly visit the selected EAs to interview each usual resident with the 4 questions: (1) full name, (2) his/her relation with the head of the household, (3) sex, and (4) date of birth. The results of such questions would be matched with the census questionnaires in order to find cases of overcount or undercount. If we assume that the result of the PES is accurate, then those who present in the PES but not in the census were undercounted, and those who present in the census but not in the PES were overcounted. The rate of overcount and undercount was shown in the below Table 3.1. TABLE 3.1: ESTIMATION OF OVERCOUNT AND UNDERCOUNT RATE OF THE 2009 CENSUS Socio-economic region Overcount rate Undercont rate Net error (Undercount rate overcount rate) The whole country Northern Midlands and Mountains Red River Delta North and South Central Coast Central Highlands Southeast Mekong River Delta Unit: % 7

17 The results reveal that net errors (the difference between undercount rate and overcount rate) of the 2009 Census is -0.3% (equal to 258,000 people), which means that double counting is greater than missing people. This rate is a very low one 1. The levels of errors are different between regions, but with a small rate (under 1%). The PES of the Census is not designed for the calculation of net errors for provinces/cities. But we suppose that the errors of the provinces/cities are also at the low level like the regions stated above. Thus, this projection does not adjust the total population collected in the 2009 Census due to errors (overcount) Population distribution by sex and five-year age groups For evaluating the level of errors in regard to distributions by sex and fiveyear age groups, we have used the method of the UN Secretariat, or as it is often called United Nations Joint Score Index (refer to Annex 3). This index consists of giving points to sex ratios and age ratios for all five-year age groups from age 0 to 74. The following formula is used to calculate the United Nations Joint Score Index: In which: JS Joint Score; SRS Sex ratio score; ARSM Male age ratio score; ARSF Female age ratio score. JS = 3*SRS + ARSM + ARSF Basing on analyzing experientially the age and sex statement in the Census of developed and developing countries, the United Nations suggested that the age and sex structure of a population will be (a) accurate if the Joint Score is below 20, (b) inaccurate if the Joint Score is between 20 and 40 and (c) very inaccurate if the Joint Score is higher than If the population has the Joint Score higher than 40 2, then the data have problems of errors, not that due to abnormal changes of the components which generated population fluctuations (fertility, mortality and 1 According to the UNFPA. in the 2000 Round of the Population and Housing Censuses, the net errors of some country s censuses are: India: 7.8%; Bangladesh: 3%; Australia: 1.6%; United States: 4%; South Korea: 1.5%; Indonesia: 3.3%; Malaysia: 4.4%; Japan: Urban: 0.4%. Rural: 0.7%; Pakistan: 4.4%. 2 See Population analysis with microcomputers, volume 1, by Eduardo E. Arriaga, November 1994, p. 226,

18 migration) and in practical use, it is necessary to adjust (mainly by smoothening method). With the index of components (sex or age), also experientially, sex index or age index higher than 10 is considered high level, which means that the data regards to sex or age have errors, not that due to abnormal changes of the components which generated population fluctuations (fertility, mortality and migration). Table 3.2 provides the Joint Score of the whole country, urban areas, rural areas and provinces/cities basing on the corporate results of the 2009 Census. From this suggestion, we see that all the Sex ratio score is at high level and the Joint Score of almost all provinces is inaccurate. Comparison to the 1999 Population and Housing Census 3 shows that the accuracy level of sex-age distribution of the whole country in the 2009 Census is lower (JS=20.4 in 2009 compared to JS=21.2 in 1999). Consequently, this projection would use the sex-age structure in the 2009 Census to be base population without any adjustment. The completed results of the 2009 Census population by sex, urban, rural areas and centrally governed provinces/cities are used to be base population for the projection (refer to Annex 4). TABLE 3.2: JOINT SCORE INDEX OF THE UNITED NATIONS (THE COMPLETED DATA) Sex ratio score Age ratio score ARS SRS Male - MARS Female - FARS Joint Score JS The whole country Urban Rural Ha Noi Ha Giang Cao Bang Bac Kan Tuyen Quang Lao Cai Dien Bien Lai Chau Son La Yen Bai Hoa Binh Thai Nguyen Lang Son Refer to: Result of population projections for whole country, geographic regions and 61 provinces/city, Viet Nam , Ha Noi 2001, Chapter 2, Table 2.2, Page 6. 9

19 Sex ratio score Age ratio score ARS SRS Male - MARS Female - FARS Joint Score JS 22. Quang Ninh Bac Giang Phu Thọ Vinh Phuc Bac Ninh Hai Duong Hai Phong Hung Yen Thai Binh Ha Nam Nam Dinh Ninh Binh Thanh Hoa Nghe An Ha Tinh Quang Binh Quang Tri Thua Thien Hue Da Nang Quang Nam Quang Ngai Binh Dinh Phu Yen Khanh Hoa Ninh Thuan Binh Thuan Kon Tum Gia Lai Dak Lak Dak Nong Lam Dong Binh Phuoc Tay Ninh Binh Duong Dong Nai Ba Ria - Vung Tau Ho Chi Minh city Long An Tien Giang Ben Tre Tra Vinh Vinh Long Dong Thap An Giang Kien Giang Can Tho Hau Giang Soc Trang Bac Lieu Ca Mau

20 3.2 Assumptions of mortality Mortality In developed countries that have fully-constituted vital registration system (fertility and mortality) and periodical population surveys, fertility rate (or morality rate) is calculated as dividing the number of births (or deaths) registered each year by the population estimated at the middle of that year. Birth registration in those countries is favorable because almost all women have a hospital birth, and death registration is also convenient thanks to legal regulations regarding to funeral which are strictly executed. In developing countries like our country where do not have such system. Demographic statisticians have to undertake independent sample surveys or combine samples in the censuses, then use the direct or indirect method to estimate fertility, marriage and mortality rates. Direct estimation is to calculate basing on the data collected from the census without any further adjustment. Indirect estimation is to calculate with adjustments basing on some other information by the method of statistical model. The objectives of applying indirect method are to improve the weaknesses of direct method. For example, we estimate the total fertility rate (TFR) from the data of fertility in the year preceding the census and the number of children of women aged (Brass method). Or to estimate the mortality rate of children basing on the total number of children ever born and the number of children surviving (dead children) of women aged (Brass method). Population data collection and processing of the General Statistical Office are undertaken cautiously and seriously in order to guarantee the most reliable data. Mortality is a sensitive matter in every family and every person s life. In practice, to collect mortality data is more difficult than to do that on fertility and migration data. Moreover, with dead people, statisticians cannot approach survey objects to get information but they have to indirectly interview via a head of the household or a representative. Hence, in estimation, mortality rate is always the 11

21 most inaccurate one. Even when using indirect estimation methods, it is still likely that children s mortality rate is low. In the 2009 Population and Housing Census, mortality data were collected from the 15% sample EAs. Basically, designing questions about mortality in the 2009 Census is similar to the 1989 and 1999 Census, including (1) to interview about mortality cases of the household within 12 months preceding the census date and (2) to interview about total number of children ever born and number of children surviving (or dead children) of women aged Particularly in the 1979 Census, only the first type of questions was applied. There are many indicators presenting mortality level. In this projection, assumptions are given following the life expectation at birth (e 0 ) of males and females. In the results of the 2009 Census, life expectation at births of males is 70.2 and of females is In 2004, the United Nations adjusted models for mortality improvement and quinquennial gain in life expectancy at birth to initial level of life expectancy with 5 increasing scenarios: very fast pace, fast pace, medium pace, slow pace and very slow pace. The Table 3.3 below presents models for mortality improvement suggested by the United Nations with only 3 increasing scenarios: medium, low and very low (refer to: World Population Prospect: The 2004 Version). TABLE 3.3: QUINQUENNIAL GAINS IN LIFE EXPECTANCY AT BIRTH ACCORDING TO INITIAL LEVEL OF LIFE ECPECTANCY AND SEX Initial expectancy level Medium pace (years) Low pace (years) Very low pace (years) (years) Male Female Male Female Male Female

22 To every projection units (the whole country, urban areas, rural areas and provinces/cities), we assume that the first stage of has the value of e 0 which was calculated from the 2009 Census. The life expectancy at birth of the next stage is calculated based on the value of e 0 of the previous stage and quinquennial gain in life expectancy at birth suggested by the United Union (equal to e 0 of previous stage plus quinquennial gain in life expectancy). As the mortality in our country has become low, small changes in mortality do not affect the projection results significantly. This is not because changes in mortality are not as important as those in fertility, but because the life expectancy at birth of Vietnam population is now far higher than the upper limit of the childbearing age group. Hence, the improvement in life expectancy in the future only can have impact on the population increase speed due to the number of elder people. From that point, this projection assumes that life expectancy at birth of Vietnam population in the future keeps increasing following the United Nations model as the Table 3.3 and initial level of life expectancy terms of starting e 0 as follows: Increasing scenario applied Medium pace Low pace Very low pace Male If e 0 <65 years If 65 years e 0 < 70 years If e 0 70 Female If e 0 <70 years If 70 years e 0 <75 years If e 0 75 Assumptions of the values of e 0 in various projection stages of the whole country, urban areas, rural areas are presented in the Annex 5, those of provinces/cities are in Annex Mortality model Beside life expectancy at birth, the component method requires a mortality pattern presented by an appropriate Age Specific Death Rate (ASDR) from which to calculate the survival probability by age. When comparing the Vietnam s mortality model according to the data of the 1989 Census with 4 families of Coale- Demeny model life tables, it was concluded that the mortality patterns of the Vietnamese population are most similar to the North model life family the family with low child and elderly mortality level (GSO, 2001). 13

23 The changing mortality level by time results in the changing mortality model. This projection continues to test the mortality model of our country s population by age sex basing on the data of the 2009 Population and Housing Census. Figure 3.1 illustrates a comparison of the Vietnam mortality model with 4 families North, South, East, West of the Coale-Demeny model life tables. FIGURE 3.1: COMPARE AGE SPECIFIC DEATH RATE OF VIETNAM WITH 4 FAMILIES OF NORTH, SOUTH, EAST, WEST Males Age Specific Death Rate (per 1000 persons) Vietnam South model East model North model West model Age group Females Age Specific Death Rate (per 1000 persons) Vietnam South model East model North model East model Age group As can be seen from the figure, ASDR curves drawn for the whole country are quite close to the ones drawn for the families of North, South, East and West. The least square method with 4 families of Coale-Demeny model life tables shows that the results of the North family has the smallest minimum value of square root function among the four families. In detail, we have minimum values when compare among 4 families of the Coale-Demeny model life tables as follows: North: Males = Females =

24 South: Males = Females = East: Males = Females = West: Males = Females = This reveals that mortality model of our country is closer to the North family than other families. In the population projection with component method, what mortality model to be chosen does not significantly affect the projection result. Moreover, child mortality in our country has decreased considerably in the recent years. According to the 1989 Population and Housing Census, infant mortality rate which was 45 in has decreased to 36.7 in 1999 (Central Steering Committee.2000) and to 16 infant deaths per thousand live births in Hence, the projection uses the mortality model of North family in the Coale-Demeny model life tables. 3.3 Fertility assumptions Fertility There are many measures about fertility, such as crude birth rate, average children born, total fertility rate, completed family size. v.v In this projection, fertility assumptions were drawn for Total Fertility Rate (TFR). The trend of fertility change in the future is projected in 3 variants: high, medium and low. Moreover, for comparison, we provide constant fertility variant assumes that TFR as estimated for the 2009 Census keeps unchanged during the projection period. Extrapolation of the past trend was used to set the medium variant. Extrapolation method used is the logistic curve (formula): K P t = 1 e a+bt In which, P t is population at the time point t and a, b and K are the parameters of the formula. 15

25 This is the method used quite popularly in population study and projection. To be convenient for calculation, we use excel spreadsheet TFRLGST of the PAS (Population Analysis Software) provided by the United State Bureau of the Census in order to match TFR values collected in the past to logistic curve. We have used TFR values estimated from the result of 3% sample survey of the 1999 Population and Housing Census, 15% sample survey of the 2009 Population and Housing Census and annual population change sample surveys from 2001 to 2008 in order to accomplish this matching. The 2 important parameters of the matching curve are the TFR of the year chosen as the starting point and rock-bottom. That is the basic principal of fertility projection, however calculation methods may differ depending on whether the starting fertility is higher or lower than the rock-bottom TFR of 1.85 children/woman. With such projection units that TFR estimated from the 2009 Census higher than 1.85 children/woman, fertility is assumed to continue decreasing with the speed as being observed in the past until it reaches 1.85 children/woman, then keep constant at that rate in the rest of the projection period (means up to 2049). With such projection units that TFR estimated from the 2009 Census lower than 1.85 children/woman, fertility is assumed to continue this trend in 5 or 10 years more. After that transition stage, fertility is assumed to increase with the speed of 0.05 children/woman for each 5-year stage. Therefore, such localities that the current fertility is very low do not necessarily reach 1.85 children/woman in This is considered to be the most feasible variant. TFR in high variant is higher than medium variant by 0.3 children/woman and low variant is lower than the medium variant with the same level. The projection results of TFR for the whole country, urban, rural for 5-year projection stage by medium variant are shown in the Table 3.4. TABLE 3.4: TOTAL FERTILITY RATE (TFR) ASSUMED FOR PROJECTION STAGES 2009 Census The whole country Urban Rural

26 Total fertility rate assumptions for projection periods of provinces/cities are presented in Annex Fertility model Assumptions of fertility model were given for the Age Specific Fertility Rates (ASFR). Figure 3.2 presents ASFRs estimated from the 2009 Population and Housing Census. Through practical observation, demographers have divided fertility models into 2 types 4 : early and late fertility depending on position where the maximum value of ASFR curve is at. In detail, early fertility model has the maximum value at the age group of and late fertility model has the maximum value at the age group of Fertility is changing by time so fertility models can change, too. However, in population projection by component method, fertility model is less important than fertility level. Hence, this projection assumes that fertility model of projection units as observed in the 2009 Population and Housing Census keeps unchanged during the projection period. FIGURE 3.2: AGE SPECIFIC FERTILITY RATE BY URBAN AND RURAL AREAS, The whole country Urban Rural From the curves in figure 3.2, we classify models as follows: 4 It could be divided into 3 or 5 types. In the case of 3 types, the medium type is added, when value of ASFR is maximum and closer in age groups of and Regarding to 5- type division, it is due to that the maximum value of ASFR is existed at age group of in some countries. Regardless how many types it is divided, the main issue is based on the position where the ASFR curve reaches maximum value. 17

27 TABLE 3.5: FERTILITY MODEL CLASSIFICATION, 2009 The whole country Urban Rural Fertility model Late Late Early Assumption of fertility models for provinces/cities is also undertaken similarly and the results are presented in Annex Sex ratio at birth projection After having the assumption for TFR and fertility model, then calculating number of children born in each single 5-year period is simple. As we do not estimate TFR by sex, we have to separate projected total fertility by males and females basing on sex ratio at birth (refer to Annex 2, Step 2c). This rate is very stable by time and space, and normally fluctuated within /100 (UNFPA, 2009). As the estimation of the United Nations, Vietnam is not among the Asian countries that bear the unbalanced child gender. However, the recent statistic evidence has proved that Sex Ratio at Birth (SRB) is on the trend of increasing. The results of annual population change survey, the 1999 and 2009 Census have shown that trend. TABLE 3.6: SEX RATIO AT BIRTH, Census/Sample survey Sex ratio at birth The 1999 population census Population change survey 1/7/2000(*) Population change survey 1/4/ Population change survey 1/4/ Population change survey 1/4/ Population change survey 1/4/ Population change survey 1/4/ Population change survey 1/4/ Population change survey 1/4/2007(**) The 2009 population census Source: From , according to Table 3, page 20, UNFPA, 2009, Data source 2009, according to Trend of bearing males in Vietnam as the data of sample surveys of the 2009 Census preliminary (Report of Mr CZ Guilmoto at the Seminar of Sex ratio at birth. Hanoi May 2010). Note: (*) For such cases of birth in 12 months preceding the survey. (**) estimation of the experts. 18

28 From this data sequence by time, the projection has undertaken line recurrence with an independent variable of time (investigation time) and dependent variable of SRB. The result shows that the annual increase level of SRB in the period of is almost 0.5 percentage point. UNFPA has projected the trend of changing sex ratio at birth with 3 variants of high, medium and low (UNFPA, 2010), in which we incline to the medium variant. Combining the results of recurrence and SRB projection medium variant of UNFPA, we suggest the assumption: SRB as in the 2009 Census (110.5 male births/100 female births) continues with the same speed in the past until it reaches the SRB of 115 male births /100 female births, after that, SRB will decrease by about 1 point per cent per year until it gets back to the standard level of 105/100 then it will keep constant at that level until the end of the projection. In detail, the result of SRB is as follows: TABLE 3.7: SEX RATIO AT BIRTH PROJECTIONS, Period/year Medium of the period/year SRB projection (male births/100 female) In this projection, SRB of our country will reach 115 in 2020 and get back to the standard level of 105 in This projection is applied for the projection of the whole country, urban areas, rural areas and provinces/cities. 3.4 Migration assumptions International migration Until now, there is still no reliable data source of international migration of our country. Therefore, this projection assumes that net international migration is none. 19

29 3.4.2 Internal migration There are 2 assumptions about migration, one is for urban/rural projection and the other is for provinces/cities projection. Detail is as follows: i) Regards to urban/rural projection: Urban-Rural Growth Difference (URGD) is the most common method to project the population urbanization level. If we assume that URGD observed in the past will continue to many years coming, it seems still reasonable. However, urban population is generally growing faster and faster. The Vietnam data also prove this trend. In detail, that difference of Vietnam in the period is 2.57% and of is 3.05%. Urban/rural projection uses URGD to project urban population in the future. This method assumes that urbanization (presented by the proportion of urban population on total population) matches with the logistic curve. At the beginning, level and speed of urbanization is low, after that they will increase. Urbanization speed increased continuously until it reached the proportion of urban population of 50% (according to the result of the 2009 Census, the proportion of urban population in our country is 29.6%). After reaching this level, the proportion of urban population continued to increase, but its speed would decrease until the speed gets 0 when urbanization is asymptotic to the highest level. The logistic curve to match the proportion of urban population is presented in the following formula: 100 U T t t = 100e dt 1 + e In which: U t /T t is the proportion of urban population at the time point t, means urbanization level at the time point t; and d is the difference of the growth in urban population and rural population. This means that if urban population s grows with the rate of u, then U t =U 0 e ut and if rural population grows with the rate of r, then R t =R 0 e rt since u-r=d. dt Basing on the data of our country s urbanization above in 20 years and the calculation suggested by the United Nations (refer to: United Nations 1974, Manual VIII Methods for Projections of Urban and Rural Population. New 20

30 York), the proportion of urban population is calculated for each year of the projection period. Using the urban population increase of the period to be base, basing on the difference between the proportion of urban population in 2014 (as the method stated above) and that in 2009, we can calculate the urban population increase in the period of This calculation is repeated for the remaining periods of the projection period. The increased urban population of projection periods also includes the people who migrate from rural areas to urban areas and the people at rural localities which have been conversed to urban localities due to the changes in administrative division from rural communes to district capital or urban wards, and due to the changes in administrative border (to expand urban areas). This increase is assumed to be the immigration of urban areas and simultaneously be the emigration of rural areas. ii) Regards to provinces/cities projection: external migration which is presented by net migration calculated from the result of the 2009 Census would be lower in each 5-year stage of the projection periods (refer to Annex 8: Migration assumptions for provinces/cities) Migration of the 0-4 aged group Because it is not possible to calculate the number of migrants among people aged 0-4 years of age from the 2009 Population and Housing Census, therefore we will have to estimate it. In previous population projections, the following formula was used to estimate net migration for the 0-4 aged group: NM(0-4) = 0.3*[NMMW(15-49)] In which, NM(0-4) Net migration aged 0-4; NMMW(15-49) Net migration among currently married women aged

31 This formula based on the formula of 0-4 age group migration estimation of the United Nations with the adjustment of multiple coefficient towards married women aged who have migrated. When issuing this formula, the General Fertility Rate (GFR), which was defined as the average fertility within the year of 1,000 or 1 woman in the childbearing age of 15-49, in almost every country it is around 100, thus the multiple coefficient is 0.5 (100 x 5 years = 0.5). The 2009 Census estimated GFR of the whole country to be 61, correlative with the multiple coefficient of 0.3 in the formula above. By using this formula, one assumes that 1/3 of married women in childbearing age who have migrated in the 5-year period will give birth one time during those 5 years Migration model Because migration level is the net migration, migration model would be the distribution by sex and age of net migrants of the projection units. Study of net migration of the whole country, urban and rural areas and provinces/cities by sex and age shows a common mode that migration is low at the low ages, then increases steadily and reach the top at a youth age; after the utmost point, migration decreases gradually when the age increases. From the views above, we assume that migration model by age in the 15% sample survey of the 2009 Population and Housing Census keeps continue during the whole projection period. In addition, in projection by component method, between the two factors of migration level and migration model, migration model has the weaker impact on projection results than migration level does. 4. Projection results for Vietnam 2009 This part summarizes projection results for Vietnam in the period of The projection consists of two main data sets: one is the population projection results for the whole country and urban/rural areas of the whole countries; the other is projection for provinces/cities divided into socio-economic regions. In each set, there would be four different variants and four assumptions about the trend of changes in fertility in the future. Among the four variants, the medium variant is considered the most feasible. 22

32 Assuming the trends of components changes that caused population fluctuation, as mentioned above, mainly bases on the trend of that component changes in the past to extrapolate for the future. Therefore, we do not make any subjective imposition. Simultaneously, when issuing assumptions, we assumed that the socio-economic situation of our country would keep stable and disasters such as wars, famines, epidemics would not happen. 4.1 Projection results for the whole country Population size The population of our country at the time of the 2009 Population and Housing Census was 85.8 million people, and by the end of the projection period (2049), the population will be million people according to the medium variant, people according to high variant, 98.3 million people according to low variant and million people according to constant variant (refer to Table 4.1). Therefore, in the next 40 years ( ), our country s population increases by 26.6%, 39.6%, 14.5% and 30.3% respectively according to the medium, high, low and constant variant. TABLE 4.1: PROJECTED POPULATION AND AVERAGE ANNUAL GROWTH RATE OF EACH STAGE, FOUR VARIANTS, Variant of fertility medium level Population (000 ) Growth rate (%) Variant of fertility high level Population (000 ) Growth rate (%) Variant of fertility low level Population (000 ) Growth rate (%) Variant of fertility constant level Population (000 ) Growth rate (%) ,847-85,847-85,847-85, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , According to medium variant, in the first 5 years of the projection period , our country s population is projected to increase at the rate of 1.09% 23

33 (refer to Figure 4.1). In the future, population increase rate is projected to decrease and get 0.11% in the period of According to medium variant, in the stage of , projected population of the whole country increases 961,000 people per year. This figures continues to decrease until 642,000 people in period, then decrease gradually, and get 121,000 in the period of FIGURE 4.1: AVERAGE ANNUAL GROWTH RATE, FOUR VARIANTS, ,5 Observed 1,5 Projected 0,5 Medium variant High variant Low variant Constant variant -0, Age and sex structure Table 4.2 presents data on the sex ratio (males/100 females) and median age of the whole country s population following four variants. First, one sees that there is almost no difference in the sex ratios between the four variants, and there is a trend of decreasing by year but it is not significant. We know that sex ratio depends on sex ratio at birth, that is, it depends on fertility, differential mortality between men and women and sex differential in net migration. For the nation projection, non-significant international net migration was assumed, so the observed immaterial increase in the sex ratios shows that the declines in fertility and mortality in the projection substantially affect the sex structure of the population. The fertility and mortality differences have contributed to change the population s age structure. All the four variants show that our country s population will be aged considerably. In medium variant, the median age increased speedily from 27.9 years at the first year of the report period (2009) to 40.5 years at the 24

34 ending year of the report period (2049), that is, increase by 12.6 years within 40 years. Low variant shows more speedy increase; high variant and constant variant show slower increase than medium variant. The differential median age between the four variants increase gradually and at the ending year of the report period. Figure 4.2 presenting our country s population pyramid at the starting year and ending year of the report period has shown clearly the changes of age structure following the trend of aging. TABLE 4.2: SEX RATIO AND MEDIAN AGE, FOUR VARIANTS, Variant of fertility medium level Sex ratio (male/100 female) Median age (years) Variant of fertility high level Sex ratio (male/100 female) Median age (years) Variant of fertility low level Sex ratio (male/100 female) Median age (years) Variant of fertility constant level Sex ratio (male/100 female) Median age (years) FIGURE 4.2: VIETNAM POPULATION PYRAMID, MEDIUM VARIANT, Male 2009 Female Male 2049 Female Percentage in total population Percentage in total population

35 All four variants show the proportion of the population under 15 will shrink. According to the medium variant, the decrease in 40 years of the report period is nearly 7 percentage point (from 24.5% in 2009 to 17.6% in 2049). The decrease of high, low and constant variant is 3.9, 12.0 and 5.8 percentage point respectively (refer to Table 4.3). TABLE 4.3: COMPARISON BETWEEN POPULATION STRUCTURES OF THE BASE YEAR AND THAT OF THE ENDING YEAR OF THE REPORT PERIOD, Variant of fertility medium level Variant of fertility high level Variant of fertility low level Variant of fertility constant level Total population (mil people) Under 15 years old (%) years old (%) years old and above (%) Median age (years) According to the medium variant, the percentage of the group of 65 years old and above increased from 6.4% in 2009 to 18% in The 3 remaining variants also reflect that this percentage has increased but not significantly different among the variants. The 2009 Census results show that, our country s labor force is in the advantageous period which is sometimes called demographic bonus. The United Unions defines this as a period when proportion of population aged 15 years and under is below 30% and proportion of elder population aged 65 years and over is also below 15% of the total population. According to medium variant and the definition above, Vietnam demographic bonus will last until 2040 because the proportion of population aged 65 years and over begins to exceed 15% as of that time. 4.2 Projection results for urban and rural areas As mentioned above, in the four variants, the medium variant is considered the most feasible. Therefore, the projection result analysis below only mention about this variant. Simultaneously, because the total urban and rural population is equal to the nation population, while the analysis of nation population projection result has been presented in detail in the section 4.1, so we only summary the projection result for urban areas. 26

36 The data summarizing projection results for urban and rural areas are presented in Table 4.4 with four variants of medium, high, low and constant. As can be seen in the following table, according to the medium variant, urban population of our country increased from 25.4 million people in 2009 to 63.9 million people in Therefore, after 40 years, the urban population of our country is projected to increase by 38.5 million people, on average increase by 962,000 people per year, reaching the urban population rate of 58.8% in TABLE 4.4: URBAN AND RURAL POPULATION AND ANNUAL AVERAGE ANNUAL GROWTH RATE PROJECTION, FOUR VARIANTS, Variant of fertility medium level Population ( 000) Average annual growthe rate (%) Variant of fertility high level Population ( 000) Average annual growthe rate (%) Variant of fertility low level Population ( 000) Average annual growthe rate (%) Variant of fertility constant level Population ( 000) Average annual growthe rate (%) Urban , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Rural , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Projection results for socio-economic regions This section presents summarily the results of medium variant. As mentioned above, data of regions are added up from provinces/cities in the region (refer to Table 4.5) 27

37 TABLE 4.5: PROJECTED POPULATION AND AVERAGE ANNUAL GROWTH RATE BY REGION, MEDIUM VARIANTS, Population ( 000) The whole country 85, , , , , ,091.8 Northern Midlands and Mountains 11, , , , , ,492.9 Red River Delta 19, , , , , ,325.1 North and South Central Coast 18, , , , , ,740.2 Central Highlands 5, , , , , ,030.4 Southeast 14, , , , , ,853.2 Mekong River Delta 17, , , , , , Average annual population growth rate (%) The whole country Northern Midlands and Mountains Red River Delta North and South Central Coast Central Highlands Southeast Mekong River Delta Data in the Table 4.5 show that, population growth rate of regions after 2009 decreases until the end of the projection period. In the period of , the nation average growth rate is 0.46%. That of Central Highlands is the highest (0.9%), followed by the Southeast (0.71%). The growth rate of Mekong River Delta is the lowest (0.3%). 4.3 Projection results for provinces/cities Even though projection is made for 8 stages from 2009 to 2049, the more detailed the projection is and the longer the period is, the less accurate the projection results are, hence projections for provinces/cities are only made for the period of (means 25 years). Table 4.6 presents projected population scale according to the medium variant for provinces/cities. The data show that population increase of projection periods decreased gradually compared to the 10-year period of In the period of there are only 5 provinces that have average annual growth rate of 1% and above (twice of the nation population growth rate). Those are Kon Tum (1.4%), Binh Duong (1.21%), Lai Chau (1.05%) and Dak Nong (1.03%). This 28

38 can be explained by the fairly high starting fertility of Kon Tum, Lai Chau and Dak Nong; and nation-wide highest net immigration rate of Binh Duong. The projection data also show that, in the period of there are 10 provinces that have very low population growth rate, at about half of the nation rate. Among these provinces, it is noticeable that annual growth rate of Thai Binh is nearly zero (0.02%), or Ben Tre (0.09%), Thanh Hoa (0.11%) and Vinh Long (0.11%). TABLE 4.6: PROJECTED POPULATION BY PROVINCES/CITIES, MEDIUM VARIANT, Population ( 000) Hanoi 6, , , , , ,382.8 Ha Giang Cao Bang Bac Kan Tuyen Quang Lao Cai Dien Bien Lai Chau Son La 1, , , , , ,469.9 Yen Bai Hoa Binh Thai Nguyen 1, , , , , ,336.7 Lang Son Quang Ninh 1, , , , , ,391.4 Bac Giang 1, , , , , ,757.4 Phu Tho 1, , , , , ,494.3 Vinh Phuc , , , , ,196.4 Bac Ninh 1, , , , , ,268.1 Hai Duong 1, , , , , ,898.9 Hai Phong 1, , , , , ,199.8 Hung Yen 1, , , , , ,299.5 Thai Binh 1, , , , , ,854.4 Ha Nam Nam Dinh 1, , , , , ,028.2 Ninh Binh Thanh Hoa 3, , , , , ,

39 Population ( 000) Nghe An 2, , , , , ,439.0 Ha Tinh 1, , , , , ,332.3 Quang Binh Quang Tri Thua Thien Hue 1, , , , , ,287.8 Da Nang , , , ,238.6 Quang Nam 1, , , , , ,585.0 Quang Ngai 1, , , , , ,338.1 Binh Dinh 1, , , , , ,695.3 Phu Yen Khanh Hoa 1, , , , , ,404.0 Ninh Thuan Binh Thuan 1, , , , , ,398.9 Kon Tum Gia Lai 1, , , , , ,766.0 Dak Lak 1, , , , , ,288.8 Dak Nong Lam Dong 1, , , , , ,580.7 Binh Phuong , , ,105.3 Tay Ninh 1, , , , , ,256.8 Binh Duong 1, , , , , ,777.0 Dong Nai 2, , , , , ,457.2 B.Ria-V.Tau , , , , ,281.8 Ho Chi Minh city 7, , , , , ,975.2 Long An 1, , , , , ,655.4 Tien Giang 1, , , , , ,835.3 Ben Tre 1, , , , , ,333.9 Tra Vinh 1, , , , , ,121.2 Vinh Long 1, , , , , ,117.1 Dong Thap 1, , , , , ,862.0 An Giang 2, , , , , ,430.5 Kien Giang 1, , , , , ,996.0 Can Tho 1, , , , , ,576.8 Hau Giang Soc Trang 1, , , , , ,461.4 Bac Lieu Ca Mau 1, , , , , ,

40 Part 2 PROJECTION RESULT TABLES 31

41 32

42 Year Table A.1 POPULATION PROJECTION BY URBAN/RURAL RESIDENCE, 4 VARIANTS, Unit: thousand persons Medium variant High variant Low variant Constant variant Total Urban Rural Total Urban Rural Total Urban Rural Total Urban Rural

43 Table A.2 POPULATION PROJECTION FOR ENTIRE COUNTRY BY AGE GROUPS AND SEX, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

44 Table A.2 (continued) Unit: thousand persons TOTAL MALE FEMALE

45 Table A.2 (continued) Unit: thousand persons TOTAL MALE FEMALE

46 Table A.2 (continued) Unit: thousand persons TOTAL MALE FEMALE

47 Table A.3 POPULATION PROJECTION FOR URBAN BY AGE GROUPS AND SEX, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

48 Table A.3 (continued) Unit: thousand persons TOTAL MALE FEMALE

49 Table A.3 (continued) Unit: thousand persons TOTAL MALE FEMALE

50 Table A.3 (continued) Unit: thousand persons TOTAL MALE FEMALE

51 Table A.4 POPULATION PROJECTION FOR RURAL BY AGE GROUPS AND SEX, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

52 Table A.4 (continued) Unit: thousand persons TOTAL MALE FEMALE

53 Table A.4 (continued) Unit: thousand persons TOTAL MALE FEMALE

54 Table A.4 (continued) Unit: thousand persons TOTAL MALE FEMALE

55 Table A.5 PROJECTION OF SOME MAJOR INDICATORS FOR THE WHOLE COUNTRY, URBAN AND RURAL, MEDIUM VARIANT, Year Total Population (thousand persons) Age Rate* ( ) Median depend Aging Sex ratio Females (males/ age ency index 100 Crude Crude Natural years years years + (years) ratio (%) females) birth death growth years (%) WHOLE COUNTRY URBAN RURAL (*): The rate is caculated on average for each phase of the projection period, in which the year recorded in the first column is the ending year of each phase 46

56 Year Table B.1 POPULATION PROJECTION FOR REGIONS AND PROVINCES, 4 VARIANTS, Unit: thousand persons Variant Variant Year Medium High Low Constant Medium High Low Constant R1. Northern Midlands and Mountains R2. Red River Delta R3. North and South Central Coast

57 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant R4. Central Highlands R5. Southeast R6. Mekong River Delta

58 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 01. Ha Noi Ha Giang Cao Bang

59 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 06. Bac Kan Tuyen Quang Lao Cai

60 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 11. Dien Bien Lai Chau Son La

61 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 15. Yen Bai Hoa Binh Thai Nguyen

62 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 20. Lang Son Quang Ninh Bac Giang

63 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 25. Phu Tho Vinh Phuc Bac Ninh

64 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 30. Hai Duong Hai Phong Hung Yen

65 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 34. Thai Binh Ha Nam Nam Dinh

66 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 37. Ninh Binh Thanh Hoa Nghe An

67 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 42. Ha Tinh Quang Binh Quang Tri

68 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 46. Thua Thien Hue Da Nang Quang Nam

69 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 51. Quang Ngai Binh Dinh Phu Yen

70 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 56. Khanh Hoa Ninh Thuan Binh Thuan

71 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 62. Kon Tum Gia Lai Dak Lak

72 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 67. Dak Nong Lam Dong Binh Phuoc

73 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 72. Tay Ninh Binh Duong Dong Nai

74 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 77. Ba Ria Vung Tau Ho Chi Minh city Long An

75 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 82. Tien Giang Ben Tre Tra Vinh

76 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 86. Vinh Long Dong Thap An Giang

77 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 91. Kien Giang Can Tho Hau Giang

78 Table B.1 (continued) Unit: thousand persons Year Variant Variant Year Medium High Low Constant Medium High Low Constant 94. Soc Trang Bac Lieu Ca Mau

79 Table B.2 POPULATION PROJECTION BY AGE GROUPS AND SEX, REGION 1, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

80 Table B.2 (continued) Unit: thousand persons TOTAL MALE FEMALE

81 Table B.2 (continued) Unit: thousand persons TOTAL MALE FEMALE

82 Table B.3 POPULATION PROJECTION BY AGE GROUPS AND SEX, REGION 2, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

83 Table B.3 (continued) Unit: thousand persons TOTAL MALE FEMALE

84 Table B.3 (continued) Unit: thousand persons TOTAL MALE FEMALE

85 Table B.4 POPULATION PROJECTION BY AGE GROUPS AND SEX, REGION 3, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

86 Table B.4 (continued) Unit: thousand persons TOTAL MALE FEMALE

87 Table B.4 (continued) Unit: thousand persons TOTAL MALE FEMALE

88 Table B.5 POPULATION PROJECTION BY AGE GROUPS AND SEX, REGION 4, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

89 Table B.4 (continued) Unit: thousand persons TOTAL MALE FEMALE

90 Table B.4 (continued) Unit: thousand persons TOTAL MALE FEMALE

91 Table B.6 POPULATION PROJECTION BY AGE GROUPS AND SEX, REGION 5, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

92 Table B.6 (continued) Unit: thousand persons TOTAL MALE FEMALE

93 Table B.6 (continued) Unit: thousand persons TOTAL MALE FEMALE

94 Table B.7 POPULATION PROJECTION BY AGE GROUPS AND SEX, REGION 6, MEDIUM VARIANT, Unit: thousand persons TOTAL MALE FEMALE

95 Table B.7 (continued) Unit: thousand persons TOTAL MALE FEMALE

96 Table B.7 (continued) Unit: thousand persons TOTAL MALE FEMALE

97 Table B.8 POPULATION PROJECTION BY AGE GROUPS AND SEX, HA NOI, MEDIUM VARIANT, TOTAL MALE FEMALE

98 Table B.8 (continued) TOTAL MALE FEMALE

99 Table B.8 (continued) TOTAL MALE FEMALE

100 Table B.9 POPULATION PROJECTION BY AGE GROUPS AND SEX, HA GIANG, MEDIUM VARIANT, TOTAL MALE FEMALE

101 Table B.9 (continued) TOTAL MALE FEMALE

102 Table B.9 (continued) TOTAL MALE FEMALE

103 Table B.10 POPULATION PROJECTION BY AGE GROUPS AND SEX, CAO BANG, MEDIUM VARIANT, TOTAL MALE FEMALE

104 Table B.10 (continued) TOTAL MALE FEMALE

105 Table B.10 (continued) TOTAL MALE FEMALE

106 Table B.11 POPULATION PROJECTION BY AGE GROUPS AND SEX, BAC KAN, MEDIUM VARIANT, TOTAL MALE FEMALE

107 Table B.11 (continued) TOTAL MALE FEMALE

108 Table B.11 (continued) TOTAL MALE FEMALE

109 Table B.12 POPULATION PROJECTION BY AGE GROUPS AND SEX, TUYEN QUANG, MEDIUM VARIANT, TOTAL MALE FEMALE

110 Table B.12 (continued) TOTAL MALE FEMALE

111 Table B.12 (continued) TOTAL MALE FEMALE

112 Table B.13 POPULATION PROJECTION BY AGE GROUPS AND SEX, LAO CAI, MEDIUM VARIANT, TOTAL MALE FEMALE

113 Table B.13 (continued) TOTAL MALE FEMALE

114 Table B.13 (continued) TOTAL MALE FEMALE

115 Table B.14 POPULATION PROJECTION BY AGE GROUPS AND SEX, DIEN BIEN, MEDIUM VARIANT, TOTAL MALE FEMALE

116 Table B.14 (continued) TOTAL MALE FEMALE

117 Table B.14 (continued) TOTAL MALE FEMALE

118 Table B.15 POPULATION PROJECTION BY AGE GROUPS AND SEX, LAI CHAU, MEDIUM VARIANT, TOTAL MALE FEMALE

119 Table B.15 (continued) TOTAL MALE FEMALE

120 Table B.15 (continued) TOTAL MALE FEMALE

121 Table B.16 POPULATION PROJECTION BY AGE GROUPS AND SEX, SON LA, MEDIUM VARIANT, TOTAL MALE FEMALE

122 Table B.16 (continued) TOTAL MALE FEMALE

123 Table B.16 (continued) TOTAL MALE FEMALE

124 Table B.17 POPULATION PROJECTION BY AGE GROUPS AND SEX, YEN BAI, MEDIUM VARIANT, TOTAL MALE FEMALE

125 Table B.17 (continued) TOTAL MALE FEMALE

126 Table B.17 (continued) TOTAL MALE FEMALE

127 Table B.18 POPULATION PROJECTION BY AGE GROUPS AND SEX, HOA BINH, MEDIUM VARIANT, TOTAL MALE FEMALE

128 Table B.18 (continued) TOTAL MALE FEMALE

129 Table B.18 (continued) TOTAL MALE FEMALE

130 Table B.19 POPULATION PROJECTION BY AGE GROUPS AND SEX, THAI NGUYEN, MEDIUM VARIANT, TOTAL MALE FEMALE

131 Table B.19 (continued) TOTAL MALE FEMALE

132 Table B.19 (continued) TOTAL MALE FEMALE

133 Table B.20 POPULATION PROJECTION BY AGE GROUPS AND SEX, LANG SON, MEDIUM VARIANT, TOTAL MALE FEMALE

134 Table B.20 (continued) TOTAL MALE FEMALE

135 Table B.20 (continued) TOTAL MALE FEMALE

136 Table B.21 POPULATION PROJECTION BY AGE GROUPS AND SEX, QUANG NINH, MEDIUM VARIANT, TOTAL MALE FEMALE

137 Table B.21 (continued) TOTAL MALE FEMALE

138 Table B.21 (continued) TOTAL MALE FEMALE

139 Table B.22 POPULATION PROJECTION BY AGE GROUPS AND SEX, BAC GIANG, MEDIUM VARIANT, TOTAL MALE FEMALE

140 Table B.22 (continued) TOTAL MALE FEMALE

141 Table B.22 (continued) TOTAL MALE FEMALE

142 Table B.23 POPULATION PROJECTION BY AGE GROUPS AND SEX, PHU THO, MEDIUM VARIANT, TOTAL MALE FEMALE

143 Table B.23 (continued) TOTAL MALE FEMALE

144 Table B.23 (continued) TOTAL MALE FEMALE

145 Table B.24 POPULATION PROJECTION BY AGE GROUPS AND SEX, VINH PHUC, MEDIUM VARIANT, TOTAL MALE FEMALE

146 Table B.24 (continued) TOTAL MALE FEMALE

147 Table B.24 (continued) TOTAL MALE FEMALE

148 Table B.25 POPULATION PROJECTION BY AGE GROUPS AND SEX, BAC NINH, MEDIUM VARIANT, TOTAL MALE FEMALE

149 Table B.25 (continued) TOTAL MALE FEMALE

150 Table B.25 (continued) TOTAL MALE FEMALE

151 Table B.26 POPULATION PROJECTION BY AGE GROUPS AND SEX, HAI DUONG, MEDIUM VARIANT, TOTAL MALE FEMALE

152 Table B.26 (continued) TOTAL MALE FEMALE

153 Table B.26 (continued) TOTAL MALE FEMALE

154 Table B.27 POPULATION PROJECTION BY AGE GROUPS AND SEX, HAI PHONG, MEDIUM VARIANT, TOTAL MALE FEMALE

155 Table B.27 (continued) TOTAL MALE FEMALE

156 Table B.27 (continued) TOTAL MALE FEMALE

157 Table B.28 POPULATION PROJECTION BY AGE GROUPS AND SEX, HUNG YEN, MEDIUM VARIANT, TOTAL MALE FEMALE

158 Table B.28 (continued) TOTAL MALE FEMALE

159 Table B.28 (continued) TOTAL MALE FEMALE

160 Table B.29 POPULATION PROJECTION BY AGE GROUPS AND SEX, THAI BINH, MEDIUM VARIANT, TOTAL MALE FEMALE

161 Table B.29 (continued) TOTAL MALE FEMALE

162 Table B.29 (continued) TOTAL MALE FEMALE

163 Table B.30 POPULATION PROJECTION BY AGE GROUPS AND SEX, HA NAM, MEDIUM VARIANT, TOTAL MALE FEMALE

164 Table B.30 (continued) TOTAL MALE FEMALE

165 Table B.30 (continued) TOTAL MALE FEMALE

166 Table B.31 POPULATION PROJECTION BY AGE GROUPS AND SEX, NAM DINH, MEDIUM VARIANT, TOTAL MALE FEMALE

167 Table B.31 (continued) TOTAL MALE FEMALE

168 Table B.31 (continued) TOTAL MALE FEMALE

169 Table B.32 POPULATION PROJECTION BY AGE GROUPS AND SEX, NINH BINH, MEDIUM VARIANT, TOTAL MALE FEMALE

170 Table B.32 (continued) TOTAL MALE FEMALE

171 Table B.32 (continued) TOTAL MALE FEMALE

172 Table B.33 POPULATION PROJECTION BY AGE GROUPS AND SEX, THANH HOA, MEDIUM VARIANT, TOTAL MALE FEMALE

173 Table B.33 (continued) TOTAL MALE FEMALE

174 Table B.33 (continued) TOTAL MALE FEMALE

175 Table B.34 POPULATION PROJECTION BY AGE GROUPS AND SEX, NGHE AN, MEDIUM VARIANT, TOTAL MALE FEMALE

176 Table B.34 (continued) TOTAL MALE FEMALE

177 Table B.34 (continued) TOTAL MALE FEMALE

178 Table B.35 POPULATION PROJECTION BY AGE GROUPS AND SEX, HA TINH, MEDIUM VARIANT, TOTAL MALE FEMALE

179 Table B.35 (continued) TOTAL MALE FEMALE

180 Table B.35 (continued) TOTAL MALE FEMALE

181 Table B.36 POPULATION PROJECTION BY AGE GROUPS AND SEX, QUANG BINH, MEDIUM VARIANT, TOTAL MALE FEMALE

182 Table B.36 (continued) TOTAL MALE FEMALE

183 Table B.36 (continued) TOTAL MALE FEMALE

184 Table B.37 POPULATION PROJECTION BY AGE GROUPS AND SEX, QUANG TRI, MEDIUM VARIANT, TOTAL MALE FEMALE

185 Table B.37 (continued) TOTAL MALE FEMALE

186 Table B.37 (continued) TOTAL MALE FEMALE

187 Table B.38 POPULATION PROJECTION BY AGE GROUPS AND SEX, THUA THIEN HUE, MEDIUM VARIANT, TOTAL MALE FEMALE

188 Table B.38 (continued) TOTAL MALE FEMALE

189 Table B.38 (continued) TOTAL MALE FEMALE

190 Table B.39 POPULATION PROJECTION BY AGE GROUPS AND SEX, DA NANG, MEDIUM VARIANT, TOTAL MALE FEMALE

191 Table B.39 (continued) TOTAL MALE FEMALE

192 Table B.39 (continued) TOTAL MALE FEMALE

193 Table B.40 POPULATION PROJECTION BY AGE GROUPS AND SEX, QUANG NAM, MEDIUM VARIANT, TOTAL MALE FEMALE

194 Table B.40 (continued) TOTAL MALE FEMALE

195 Table B.40 (continued) TOTAL MALE FEMALE

196 Table B.41 POPULATION PROJECTION BY AGE GROUPS AND SEX, QUANG NGAI, MEDIUM VARIANT, TOTAL MALE FEMALE

197 Table B.41 (continued) TOTAL MALE FEMALE

198 Table B.41 (continued) TOTAL MALE FEMALE

199 Table B.42 POPULATION PROJECTION BY AGE GROUPS AND SEX, BINH DINH, MEDIUM VARIANT, TOTAL MALE FEMALE

200 Table B.42 (continued) TOTAL MALE FEMALE

201 Table B.42 (continued) TOTAL MALE FEMALE

202 Table B.43 POPULATION PROJECTION BY AGE GROUPS AND SEX, PHU YEN, MEDIUM VARIANT, TOTAL MALE FEMALE

203 Table B.43 (continued) TOTAL MALE FEMALE

204 Table B.43 (continued) TOTAL MALE FEMALE

205 Table B.44 POPULATION PROJECTION BY AGE GROUPS AND SEX, KHANH HOA, MEDIUM VARIANT, TOTAL MALE FEMALE

206 Table B.44 (continued) TOTAL MALE FEMALE

207 Table B.44 (continued) TOTAL MALE FEMALE

208 Table B.45 POPULATION PROJECTION BY AGE GROUPS AND SEX, NINH THUAN, MEDIUM VARIANT, TOTAL MALE FEMALE

209 Table B.45 (continued) TOTAL MALE FEMALE

210 Table B.45 (continued) TOTAL MALE FEMALE

211 Table B.46 POPULATION PROJECTION BY AGE GROUPS AND SEX, BINH THUAN, MEDIUM VARIANT, TOTAL MALE FEMALE

212 Table B.46 (continued) TOTAL MALE FEMALE

213 Table B.46 (continued) TOTAL MALE FEMALE

214 Table B.47 POPULATION PROJECTION BY AGE GROUPS AND SEX, KON TUM, MEDIUM VARIANT, TOTAL MALE FEMALE

215 Table B.47 (continued) TOTAL MALE FEMALE

216 Table B.47 (continued) TOTAL MALE FEMALE

217 Table B.48 POPULATION PROJECTION BY AGE GROUPS AND SEX, GIA LAI, MEDIUM VARIANT, TOTAL MALE FEMALE

218 Table B.48 (continued) TOTAL MALE FEMALE

219 Table B.48 (continued) TOTAL MALE FEMALE

220 Table B.49 POPULATION PROJECTION BY AGE GROUPS AND SEX, DAK LAK, MEDIUM VARIANT, TOTAL MALE FEMALE

221 Table B.49 (continued) TOTAL MALE FEMALE

222 Table B.49 (continued) TOTAL MALE FEMALE

223 Table B.50 POPULATION PROJECTION BY AGE GROUPS AND SEX, DAK NONG, MEDIUM VARIANT, TOTAL MALE FEMALE

224 Table B.50 (continued) TOTAL MALE FEMALE

225 Table B.50 (continued) TOTAL MALE FEMALE

226 Table B.51 POPULATION PROJECTION BY AGE GROUPS AND SEX, LAM DONG, MEDIUM VARIANT, TOTAL MALE FEMALE

227 Table B.51 (continued) TOTAL MALE FEMALE

228 Table B.51 (continued) TOTAL MALE FEMALE

229 Table B.52 POPULATION PROJECTION BY AGE GROUPS AND SEX, BINH PHUOC, MEDIUM VARIANT, TOTAL MALE FEMALE

230 Table B.52 (continued) TOTAL MALE FEMALE

231 Table B.52 (continued) TOTAL MALE FEMALE

232 Table B.53 POPULATION PROJECTION BY AGE GROUPS AND SEX, TAY NINH, MEDIUM VARIANT, TOTAL MALE FEMALE

233 Table B.53 (continued) TOTAL MALE FEMALE

234 Table B.53 (continued) TOTAL MALE FEMALE

235 Table B.54 POPULATION PROJECTION BY AGE GROUPS AND SEX, BINH DUONG, MEDIUM VARIANT, TOTAL MALE FEMALE

236 Table B.54 (continued) TOTAL MALE FEMALE

237 Table B.54 (continued) TOTAL MALE FEMALE

238 Table B.55 POPULATION PROJECTION BY AGE GROUPS AND SEX, DONG NAI, MEDIUM VARIANT, TOTAL MALE FEMALE

239 Table B.55 (continued) TOTAL MALE FEMALE

240 Table B.55 (continued) TOTAL MALE FEMALE

241 Table B.56 POPULATION PROJECTION BY AGE GROUPS AND SEX, BA RIA - VUNG TAU, MEDIUM VARIANT, TOTAL MALE FEMALE

242 Table B.56 (continued) TOTAL MALE FEMALE

243 Table B.56 (continued) TOTAL MALE FEMALE

244 Table B.57 POPULATION PROJECTION BY AGE GROUPS AND SEX, HO CHI MINH CITY, MEDIUM VARIANT, TOTAL MALE FEMALE

245 Table B.57 (continued) TOTAL MALE FEMALE

246 Table B.57 (continued) TOTAL MALE FEMALE

247 Table B.58 POPULATION PROJECTION BY AGE GROUPS AND SEX, LONG AN, MEDIUM VARIANT, TOTAL MALE FEMALE

248 Table B.58 (continued) TOTAL MALE FEMALE

249 Table B.58 (continued) TOTAL MALE FEMALE

250 Table B.59 POPULATION PROJECTION BY AGE GROUPS AND SEX, TIEN GIANG, MEDIUM VARIANT, TOTAL MALE FEMALE

251 Table B.59 (continued) TOTAL MALE FEMALE

252 Table B.59 (continued) TOTAL MALE FEMALE

253 Table B.60 POPULATION PROJECTION BY AGE GROUPS AND SEX, BEN TRE, MEDIUM VARIANT, TOTAL MALE FEMALE

254 Table B.60 (continued) TOTAL MALE FEMALE

255 Table B.60 (continued) TOTAL MALE FEMALE

256 Table B.61 POPULATION PROJECTION BY AGE GROUPS AND SEX, TRA VINH, MEDIUM VARIANT, TOTAL MALE FEMALE

257 Table B.61 (continued) TOTAL MALE FEMALE

258 Table B.61 (continued) TOTAL MALE FEMALE

259 Table B.62 POPULATION PROJECTION BY AGE GROUPS AND SEX, VINH LONG, MEDIUM VARIANT, TOTAL MALE FEMALE

260 Table B.62 (continued) TOTAL MALE FEMALE

261 Table B.62 (continued) TOTAL MALE FEMALE

262 Table B.63 POPULATION PROJECTION BY AGE GROUPS AND SEX, DONG THAP, MEDIUM VARIANT, TOTAL MALE FEMALE

263 Table B.63 (continued) TOTAL MALE FEMALE

264 Table B.63 (continued) TOTAL MALE FEMALE

265 Table B.64 POPULATION PROJECTION BY AGE GROUPS AND SEX, AN GIANG, MEDIUM VARIANT, TOTAL MALE FEMALE

266 Table B.64 (continued) TOTAL MALE FEMALE

267 Table B.64 (continued) TOTAL MALE FEMALE

268 Table B.65 POPULATION PROJECTION BY AGE GROUPS AND SEX, KIEN GIANG, MEDIUM VARIANT, TOTAL MALE FEMALE

269 Table B.65 (continued) TOTAL MALE FEMALE

270 Table B.65 (continued) TOTAL MALE FEMALE

271 Table B.66 POPULATION PROJECTION BY AGE GROUPS AND SEX, CAN THO, MEDIUM VARIANT, TOTAL MALE FEMALE

272 Table B.66 (continued) TOTAL MALE FEMALE

273 Table B.66 (continued) TOTAL MALE FEMALE

274 Table B.67 POPULATION PROJECTION BY AGE GROUPS AND SEX, HAU GIANG, MEDIUM VARIANT, TOTAL MALE FEMALE

275 Table B.67 (continued) TOTAL MALE FEMALE

276 Table B.67 (continued) TOTAL MALE FEMALE

277 Table B.68 POPULATION PROJECTION BY AGE GROUPS AND SEX, SOC TRANG, MEDIUM VARIANT, TOTAL MALE FEMALE

278 Table B.68 (continued) TOTAL MALE FEMALE

279 Table B.68 (continued) TOTAL MALE FEMALE

280 Table B.69 POPULATION PROJECTION BY AGE GROUPS AND SEX, BAC LIEU, MEDIUM VARIANT, TOTAL MALE FEMALE

281 Table B.69 (continued) TOTAL MALE FEMALE

282 Table B.69 (continued) TOTAL MALE FEMALE

283 Table B.70 POPULATION PROJECTION BY AGE GROUPS AND SEX, CA MAU, MEDIUM VARIANT, TOTAL MALE FEMALE

284 Table B.70 (continued) TOTAL MALE FEMALE

285 Table B.70 (continued) TOTAL MALE FEMALE

286 Table B.71 PROJECTION OF SOME MAJOR INDICATORS FOR PROVINCES, MEDIUM VARIANT, Year Total Population (thousand persons) Agedepend Ageing Rate* ( ) Median Sex ratio Females (males/ age ency index 0 Crude Crude Natural years years years + (years) ratio (%) females) birth death growth years (%) 01. Ha Noi Ha Giang Cao Bang Bac Kan Tuyen Quang

287 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 10. Lao Cai Dien Bien Lai Chau Son La Yen Bai

288 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 17. Hoa Binh Thai Nguyen Lang Son Quang Ninh Bac Giang

289 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 25. Phu Tho Vinh Phuc Bac Ninh Hai Duong Hai Phong

290 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 33. Hung Yen Thai Binh Ha Nam Nam Dinh Ninh Binh

291 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 38. Thanh Hoa Nghe An Ha Tinh Quang Binh Quang Tri

292 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 46. Thua Thien Hue Da Nang Quang Nam Quang Ngai Binh Dinh

293 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 54. Phu Yen Khanh Hoa Ninh Thuan Binh Thuan Kon Tum

294 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 64. Gia Lai Dak Lak Dak Nong Lam Dong Binh Phuoc

295 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 72. Tay Ninh Binh Duong Dong Nai Ba Ria Vung Tau Ho Chi Minh city

296 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 80. Long An Tien Giang Ben Tre Tra Vinh Vinh Long

297 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 87. Dong Thap An Giang Kien Giang Can Tho Hau Giang

298 Table B.71 (continued) Year Total Population (thousand persons) 0-14 years years 65 years + Females years Median age (years) Agedepend ency ratio (%) Ageing index (%) Sex ratio (males/10 0 females) Crude birth Rate* ( ) Crude death Natural growth 94. Soc Trang Bac Lieu Ca Mau (*): The rate is caculated on average for each phase of the projection period, in which the year recorded in the first column is the ending year of each phase 289

299 290

300 ANNEXES 291

301 Annex 1: Socio-economic regions R1. Northern Midlands and Mountain 1. Ha Giang 2. Cao Bang 3. Bac Kan 4. Tuyen Quang 5. Lao Cai 6. Dien Bien 7. Lai Chau 8. Son La 9. Yen Bai 10. Hoa Binh 11. Thai Nguyen 12. Lang Son 13. Bac Giang 14. Phu Thọ R2. Red River Delta 15. Ha Noi 16. Quang Ninh 17. Vinh Phuc 18. Bac Ninh 19. Hai Duong 20. Hai Phong 21. Hung Yen 22. Thai Binh 23. Ha Nam 24. Nam Dinh 25. Ninh Binh R3. North and South Central Coast 26. Thanh Hoa 27. Nghe An 28. Ha Tinh 29. Quang Binh 30. Quang Tri 31. Thua Thien Hue 32. Da Nang 33. Quang Nam 34. Quang Ngai 35. Binh Dinh 36. Phu Yen 37. Khanh Hoa 38. Ninh Thuan 39. Binh Thuan R4.Central Highlands 40. Kon Tum 41. Gia Lai 42. Dak Lak 43. Dak Nong 44. Lam Dong R5. South East 45. Binh Phuoc 46. Tay Ninh 47. Binh Duong 48. Dong Nai 49. Ba Ria-Vung Tau 50. Ho Chi Minh city R6. Mekong River Delta 51. Long An 58. Kien Giang 52. Tien Giang 59. Can Tho 53. Ben Tre 60. Hau Giang 54. Tra Vinh 61. Soc Trang 55. Vinh Long 62. Bac Lieu 56. Dong Thap 63. Ca Mau 57. An Giang 292

302 Annex 2: Component method The nature of component method is to base on the level and the changing trend of three factors: fertility, mortality and migration, as well as size and structure of the population by sex and age at a certain point of time in order to project the scale and structure of population in various points of time in the future. Therefore, component method has the following advantages: (1) Produce data basing on age and sex. (2) Determine clearly the trend of population develop components that is fertility, mortality and migration. By this component method, one can see the affection of a certain trend. (3) Issue more accurate results than other methods due to the use of certain models about age and sex structure as well as fertility, mortality and migration models by age. (4) More simple in the definition, calculation formulas and calculation method than many other methods. (5) Exploit efficiently ready data sources including such types of information like expectation and wishes about family scale of the spouses collected in many sample surveys at nation and province level. Population projection according to component method needs the following input information: (i) Initial population by sex and age structure (Base population) (ii) Mortality assumptions; (iii) Fertility assumptions; (iv) Migration assumptions. When projecting the population of the whole country, urban areas, rural areas or provinces, input information mentioned above from (i) to (iv) have to be presented for the whole country, urban/rural areas and provinces. The standard component method mentions population project in each 5 year period. Hence assumptions about population fluctuation components must present 293

303 for each 5 year time point or each 5 year period. These assumptions must be presented in various forms of fertility, morality and migration measures. When using the component method to project the population, the results may include: (i) Population by sex and age; (ii) Population change rate due to fertility, morality and migration. The projection process according to the component method includes certain calculation steps that are repeated for each projection period, normally in 5 years. Those steps of the component method are: Step 1: Calculate number of people aged five and above and are still alive at the end of the projection period. In this step one will use mortality assumptions in the future, which are life coefficient by age, to determine the population structure by sex and age at the end of projection periods according to this formula: In which: 5P 1 x+5 = 5 P 0 x. 5 S x (1) 5P 1 x+5 is the population aged (x+5, x+10) at the end of the projection period; 5P 0 x 5S x is the population aged (x, x+5) at the start of the projection period; is the life coefficient of the age group (x, x+5) in the projection period x = 0, 5, In the formula (1), 5 S x life coefficient of the age group (x, x+5) is calculated from the life table as follows: In which: 5S x = 5 L x+5 / 5 L x (2) 5L x is number of people who age (x, x+5) of the life table. 294

304 In principal, a population projection is prepared in the combination with development programs built for a year circle. According to this concept, implementing a projection according to the component method will be repeated for several 5 year circles. Step 2: Calculate number of people aged below five at the end of the projection period. The number of people below 5 years old (0-4) at the end of the projection period is the number of children born within the projection period who are still alive at the end of the projection period. Therefore, in order to calculate this number, one firstly have to calculate the number of children born within the projection period, then multiple this number of children by appropriate coefficient. The number of children born within the projection period is calculated from fertility rate according to fertility assumptions and the average number of women in child-bearing age in the projection period and the sex ratio at birth. a) The average number of women in child-bearing age in the projection period The average number of women of the age group (x, x+5) in the projection period is calculated by summing the number of that age group at the start and end of the projection period then divide by 2, as the following formula: 5P f,x = 1/2( 5 P 0 f,x + 5 P t f,x) (3) In which: 5P 0 f,x Number of women aged (x, x+5) at the start of projection period; 5P t f,x Number of women aged (x, x+5) at the end of projection period. b) Totalize fertilities in the projection period The number of children born in the projection period is calculated basing on the average women in child-bearing age and Age specific birth rate of the projection period. Total of children born is calculated according to the formula: 295

305 B = 5 5 P f,x * 5 ASFR x (4) In which: 5ASFR x is Specific birth rate of women aged (x, x+5) in the projection period. c) Calculate the specific births for each sex Since the number of female children aged under five (0-4 years old) at the end of the projection period is the number of female children born in the projection period still alive until the end of the projection period, hence in the formula (4), if the age specific fertility rate of women presents for each single sex, then when multiplying the average number of women with ASFR of which sex one will have the fertility of that sex. If the ASFR of women is calculated for both sexes, then after calculating the fertility total, one must separate by each single sex. This can be accomplished using the sex ratio at birth. Assume that: B Bm Bf Total of children born; Males born; Females born; SRB Sex ratio at birth, defined by number of males born on a 100 females born. Then, B m + B f = B and B m /B f = SRB. Thus: Bm = B*SRB/(1+SRB) Bf = B*1/ (1+SRB) (5). d) Determine population aged under 5 years by sex After determine the fertility in the projection period by each single sex, we can easily calculate population aged below 5 at the end of the projection period using life coefficient from the time of birth to 0-4 years old (life ability from one s birth to 0-4 years old). Life coefficient from one s birth to 0-4 years old is calculated by dividing number of alive people in the group of 0-4 years olds of the life tables ( 5 L 0 ) by 5 times of the initial fertility of that life tables (l 0 ): 296

306 5P m,0 = B m * 5 L m,0 / 5*l 0 5P f,0 = B m * 5 L f,0 / 5*l 0 (6) In which: 5P m.0 Males aged 0-4 years old at the end of the projection period; 5P f.0 Females aged 0-4 years old at the end of the projection period; 5L m.0 Alive people aged 0-4 of male life table; 5L f.0 Alive people aged 0-4 of female life table; l 0 Initial fertility of the life table (base of the life table); Step 3: Project the net migration Net migration (equal to immigration minus emigration) by age can be estimated by multiplying age specific net migration rate with population by age at the end of the projection period estimated in the step 1 and 2. Net migration for each sex is determined as follows: 5M m,x = 5 P m,x * 5 ASNMR m,x 5M f,x = 5 P f,x * 5 ASNMR f,x (7) In which: 5M m,x 5M f,x Net migration of males aged x in the projection period; Net migration of females aged x in the projection period; 5ASNMR f,x Age and sex specific net migration rate; Finally, total population at the end of the projection period will be calculated by summing the population aged 5 and above by each sex calculated in the step 1 with the population aged 5 and below by each sex calculated in the step 2 and net migration in the period calculated in the step

307 Annex 3: Method of united nations secrecteriat The method of United Nations Secretariat (also called United Nation Joint Score Index) is used to evaluate the level of errors in regard to population distribution by sex and five-year age groups. This index consists of giving points to sex ratios and age ratios for all five-year age groups from age 0 to 74. In order to analyze the sex ratio point, one has calculated the result of sex ratio of a age group minus that of a higher consecutive age group and calculate the average number which does not mention about whether the number is negative or positive), to be the base Sex Ratio Score. Age ratio is determined as the rate of people in a certain age group with the average of people in the two consecutive age groups as follows: P x 2P x Age ratio = * 100 = * 100 (3.1) (P x-1 + P x+1 )/2 P x-1 + P x+1 In which, P x : the population of a certain age group, for example P x-1 : the population of the right previous age group, which is here P x+1 : the population or the right next age group, which is here With a normal age distribution, age ratio, when presented by percentage number, will be very close to 100. Therefore one will calculate the difference with 100 and get the average of them regardless negative or positive sign, to be the base Age ratio scores for each sex. In detail, ARSM is the Age ratio score for males and ARSF is the age ratio score for females. As the sex ratio is more stable, it has greater weight in the Joint Score (JS). In detail, the JS is calculated by summing the 3 times of ARSM and ARSF. The formula is as follows: JS = 3*SRS + ARSM + ARSF (3.2) 298

308 Based on analyzing experientially the age and sex statements in the Census of developed and developing countries, the United Nations suggested that the age and sex structure of a population will be (a) accurate if the Joint Score is below 20. (b) inaccurate if the Joint Score is between 20 and 40 and (c) very inaccurate if the Joint Score is higher than If the population has the Joint Score higher than 40, then the data have problems of errors, not that due to abnormal changes of the components which generated population fluctuations (fertility, mortality and migration) and in practical use, it is necessary to adjust (mainly by smoothening method). With the index of components (sex or age), also experientially, sex index or age index higher than 10 is considered high level, which means that the data regards to sex or age have errors, not that due to abnormal changes of the components which generated population fluctuations (fertility. mortality and migration). 1 Refer to Population analysis with microcomputers, volume 1, by Eduardo E. Arriaga, November 1994, p. 226,

309 Annex 4: Base population of population projection by sex and administration units Administration units Population Total Males Females The whole country Urban Rural Ha Noi Ha Giang Cao Bang Bac Kan Tuyen Quang Lao Cai Dien Bien Lai Chau Son La Yen Bai Hoa Binh Thai Nguyen Lang Son Quang Ninh Bac Giang Phu Tho Vinh Phuc Bac Ninh Hai Duong Hai Phong Hung Yen Thai Binh Ha Nam Nam Dinh Ninh Binh Thanh Hoa Nghe An Ha Tinh Quang Binh Quang Tri

310 Administration units Population Total Males Females 46. Thua Thien Hue Da Nang Quang Nam Quang Ngai Binh Dinh Phu Yen Khanh Hoa Ninh Thuan Binh Thuan Kon Tum Gia Lai Dak Lak Dak Nong Lam Dong Binh Phuoc Tay Ninh Binh Duong Dong Nai Ba Ria - Vung Tau Ho Chi Minh City Long An Tien Giang Ben Tre Tra Vinh Vinh Long Dong Thap An Giang Kien Giang Can Tho Hau Giang Soc Trang Bac Lieu Ca Mau

311 Annex 5: Life-expectation at birth projection for the whole country, urban areas, rural areas, Average life-expectation at birth Males The whole country Urban Rural Females The whole country Urban Rural

312 Annex 6: Life-expectation at birth projection for provinces, cities, Life-expectation at birth of males Life-expectation at birth of females Administrative units Ha Noi Ha Giang Cao Bang Bac Kan Tuyen Quang Lao Cai Dien Bien Lai Chau Son La Yen Bai Hoa Binh Thai Nguyen Lang Son Quang Ninh Bac Giang Phu Tho Vinh Phuc Bac Ninh Hai Duong Hai Phong Hung Yen Thai Binh Ha Nam Nam Dinh Ninh Binh Thanh Hoa Nghe An Ha Tinh Quang Binh Quang Tri Thua Thien Hue

313 Life-expectation at birth of males Life-expectation at birth of females Administrative units Da Nang Quang Nam Quang Ngai Binh Dinh Phu Yen Khanh Hoa Ninh Thuan Binh Thuan Kon Tum Gia Lai Dak Lak Dak Nong Lam Dong Binh Phuoc Tay Ninh Binh Duong Dong Nai Ba Ria - Vung Tau Ho Chi Minh City Long An Tien Giang Ben Tre Tra Vinh Vinh Long Dong Thap An Giang Kien Giang Can Tho Hau Giang Soc Trang Bac Lieu Ca Mau

314 Annex 7: Total fertility rate projection and fertility model for provinces/cities Administrative units Total fertility rate year 2009 Total fertility rate projection for periods Fertility model 01. Ha Noi Late 02. Ha Giang Early 04. Cao Bang Early 06. Bac Kan Early 08. Tuyen Quang Early 10. Lao Cai Early 11. Dien Bien Early 12. Lai Chau Early 14. Son La Early 15. Yen Bai Early 17. Hoa Binh Early 19. Thai Nguyen Early 20. Lang Son Early 22. Quang Ninh Early 24. Bac Giang Early 25. Phu Tho Early 26. Vinh Phuc Early 27. Bac Ninh Early 30. Hai Duong Late 31. Hai Phong Late 33. Hung Yen Late 34. Thai Binh Late 35. Ha Nam Early 36. Nam Dinh Early 37. Ninh Binh Late 38. Thanh Hoa Late 40. Nghe An Late 42. Ha Tinh Late 44. Quang Binh Late 45. Quang Tri Late 46. Thua Thien Hue Late 305

315 Administrative units Total fertility rate year 2009 Total fertility rate projection for periods Fertility model 48. Da Nang Late 49. Quang Nam Late 51. Quang Ngai Late 52. Binh Dinh Late 54. Phu Yen Late 56. Khanh Hoa Late 58. Ninh Thuan Late 60. Binh Thuan Late 62. Kon Tum Late 64. Gia Lai Early 66. Dak Lak Late 67. Dak Nong Early 68. Lam Dong Late 70. Binh Phuoc Early 72. Tay Ninh Early 74. Binh Duong Late 75. Dong Nai Late 77. Ba Ria - Vung Tau Late 79. Ho Chi Minh City Late 80. Long An Late 82. Tien Giang Early 83. Ben Tre Early 84. Tra Vinh Late 86. Vinh Long Late 87. Dong Thap Early 89. An Giang Early 91. Kien Giang Early 92. Can Tho Late 93. Hau Giang Late 94. Soc Trang Late 95. Bac Lieu Early 96. Ca Mau Early 306

316 Annex 8: Migration assumptions for provinces/cities The data of the 2009 Population and Housing Census shows that in the 5- year period of , there are 3,397 thousands people aged 5 and above participating inter-province migration. This is the period when migration blooms due to the fact that many industrial zones and plants take into operation in the main-point economic areas, attracting labors from other provinces. Since the beginning of 2009, due to the adjustment in the investment and construction of industrial zones in all the provinces throughout the country and due to the effect of global economic downward, the province migration has decreased. Basing on net migration of provinces/cities, we divide provinces/cities (called provinces) in the following 5 groups: 1. Group 1: Provinces that emigrates at high rate, including: Thai Binh, Ha Nam, Nam Dinh, Thanh Hoa, Ha Tinh, Ben Tre, Tra Vinh, Vinh Long. 2. Group 2: Provinces that emigrates at medium level, including: Bac Giang, Phu Tho, Vinh Phuc, Bac Ninh, Hung Yen, Ninh Binh, Nghe An, Quang Binh, Quang Tri, Thua Thien Hue, Quang Nam, Quang Ngai, Binh Dinh, Ninh Thuan, Binh Phuoc, Long An, Tien Giang, Dong Thap, An Giang, Kien Giang, Hau Giang, Soc Trang, Bac Lieu and Ca Mau. 3. Group 3: Provinces that emigates at low level, including: Ha Giang, Cao Bang, Bac Kan, Tuyen Quang, Lao Cai, Yen Bai, Hoa Binh, Thai Nguyen, Lang Son, Hai Duong, Phu Yen, Khanh Hoa, Binh Thuan, Tay Ninh. 4. Group 4: Provinces that has abnormal emigrations, including: Dien Bien (emigrates), Lai Chau (immigration), Son La (immigration) and Dak Lak (emigration). 5. Group 5: Provinces that has immigrations, including: Ha Noi, Quang Ninh, Hai Phong, Da Nang, Kon Tum, Gia Lai, Dak Nong, Lam Dong, Binh Duong, Dong Nai, Ba Ria-Vung Tau, Ho Chi Minh city and Can Tho. In the coming years we suppose that migration is still at high level, mainly internal migration. Hence, migration assumptions are made for each group. 307

317 Group 1, 2 and 3 are corresponding with highly, medium and low decreased migration. The province s net migration is assumed to decrease for each 5-year period of the whole projection period. Group 4: in the period of , the provinces of this group have abnormal migration due to the population restructure because of separating provinces (particularly for Son La, due to building up the Son La hydroelectric plant). In the 10-year period of , the old Lai Chau (consists of new Dien Bien and new Lai Chau) and Son La is the emigration province; and Dak Lak is the immigration province. Moreover, since the net migration is very small, with the reasons above, migration assumptions to the provinces belonging to group 4 are not mentioned. Although migration assumptions are not made for Dak Lak, migration assumptions of the whole Central Highlands are balanced to equal to that of the period Group 5: Total emigrants of Group 1, 2 and 3 will be distributed for immigration provinces of Group 5 by the proportion of immigrants in the period of each province gained from the 2009 Census. Particularly the migrants of Can Tho city have been increased 10 times 2, since this city has been lately established and has the very low number of migrants that will be the central city of Mekong River Delta in the future. 2 Number of immigrant in the period of to Can Tho is 3,458 people, being very small in comparison to that of other provinces in the same group. 308

318 REFERENCES 1. Steering Committee for the Central population Census, 1990, Vietnam Population Census-1989, Sample survey result, Hanoi. 2. Steering Committee for the Central population Census, 1991, Vietnam Population Census-1989, Comprehensive survey results, part 1, Hanoi 3. Steering Committee of the Central Population and Housing Census, 1999, Vietnam Population and Housing Census-1999, Preliminary results, Hanoi. 4. Steering Committee of the Central Population and Housing Census, Vietnam Population and Housing Census 1999, Sample survey results, Hanoi. 5. General Statistical Office 1991, Vietnam Population Census Sample survey result analysis, Hanoi. 6. General Statistical Office 1991, Vietnam Population Projection Report, , Hanoi 7. General Statistical Office and the National Committee of Population and Family Planning, 1995, Survey results of population and family planning changes 1/4/1993, Hanoi. 8. UN, 1983, Handbook number X: Indirect techniques of demographic estimation (Vietnamese translation, Science and Technology Publishing House, Hanoi-1996) 9. UN, 1995, World Population Prospects. The 1994 Revision. New York 10. UN World Urbanization Prospects: The 1996 Revision. New York 11. UN. 1999a. World Population Prospects: The 1998 Revision. Volume II. New York 12. UN. 1999b. Long-range World Population Projections: Based on the 1998 Revision. New York 13. Steering Committee of Central Population and Housing Census, the 1/4/2009 Population and Housing Census: Sample extrapolation results, Hanoi, 12/ General Statistical Office, Vietnam population projection result report, Hanoi

319 15. General Statistical Office, Population projection result for the whole country, geographic regions and 61 provinces/cities, Vietnam, , Hanoi, UNFPA. The current changes in sex at birth in Vietnam, Hanoi, 8/ UNFPA. Gender imbalances at birth in Vietnam: first lessons from the 2009 census. Hanoi World Population Prospects. The 2004 Revision. Volume III. Analytical Report. United Nations Population Division. New York. 19. World Population Prospects. The 2008 Revision. Population Database. UN Population Division. New York. 20. United Nations 1974, Manual VIII Methods for Projections of Urban and Rural Population. 310

320