World Population Prospects, the 2010 Revision

The 2010 Revision is the most recent revision of World Population Prospects, the official United Nations population estimates and projections. It was released on 3 May 2011.

2. When will the next Revision of the World Population Prospects be released?

The Population Division of the Department of Economic and Social Affairs of the United Nations issues a new Revision every two years. The next one is due in the first half of 2013.

3. When has the world population reached or is expected to reach each successive billion?

3 Billion: 20 October 1959

4 Billion: 27 June 1974

5 Billion: 21 January 1987

6 Billion: 5 December 1998

7 Billion: 31 October 2011

8 Billion: 15 June 2025

9 Billion: 18 February 2043

10 Billion: 18 June 2083

These dates are derived from the annual series of world population estimates obtained by interpolating the results of the 2010 Revision of World Population Prospects by assuming exponential growth within each year. Because the estimated dates depend on the population estimates and projections used to derive them and those estimates and projections change from one Revision to another, they constitute the best approximations at any given time.

4. How do we know that the world population reaches 7 billion on 31 October 2011?

We don’t know that the world population will reach exactly 7 billion on 31 October 2011. No one can know this.

The 31^st October is a symbolic date, which is based on interpolated data from the original 5-year period estimates prepared by the Population Division. These population estimates, which are published every second year as part of the World Population Prospects, are based on the most recently available censuses, surveys, vital and population registers, and other data sources. The Population Division is preparing such estimates since the early 1950s. They have been recognized as the “gold standard” of population estimates and projections for all countries of the world. The estimates for each country are always prepared separately. Only once these data are uploaded into a database, projected, aggregated and interpolated is it possible to calculate the date when the world population reaches a certain number of people.

The reason, for the high uncertainty in this date is the fact that statistical data are not perfect. Even the best censuses in the world have enumeration errors in the range of at least 1 to 2 percent. If we assume an enumeration error of 2 percent for the census of China, India and Indonesia, the world population could be 56 million larger or smaller. Of course, the Population Division tries to compensate these errors by using information from post-enumeration surveys and other sources if available. It is also likely that overestimation in some countries is partially compensated by underestimation in other countries. Yet, it is very likely that we have uncertainty in total population estimates of at least 1 percent at the global level. If we assume an error margin of only 1 percent at the global level the 7 billion world population could be reached 6 months earlier or later (in 20 March 2011 or 12 April 2012 as compared to our calculated date of 31 October 2011).

In other words, no one can determine the exact date of a 7 billion world population with an error margin smaller than about 12 months because of the inevitable inaccuracies in all demographic statistics – including even the most professional censuses. In fact, due to very poor demographic statistics in a significant number of developing countries the uncertainty may be even larger.

5. What data have been published from the 2010 Revision of the World Population Prospects?

The results of the 2010 Revision of the World Population Prospects are available at different levels of detail, for various kind of aggregations and in several electronic formats. The table below provides an overview:

	Name	Location	Level of detail	Format
1.	On-line Database	Web Site (Free)	Selected results available in 5-year age groups and five year estimation and projection intervals; only the four standard variants (medium, high, low, constant fertility); only 5 countries at a time	Web site tables and comma-separated files (csv-files)
2.	Tables in EXCEL-Format	Web Site (Free)	Selected results for all countries and major regions; only in 5-year age groups and five year estimation and projection intervals; only the four standard variants (medium, high, low, constant fertility)	Files in EXCEL format
3.	Key Indicators: Tables	Web Site (Free)	Selected demographic indicators for all countries of the world with a population of more than 100,000 inhabitants - available in tables that can be sorted by clicking on the column-header	html- tables
4.	Figures / Maps	Web Site (Free)	Selected demographic indicators - available in figures that can be retrieved for one period and all countries of the world with a population of more than 100,000 inhabitants or for 5 countries and all periods (1950-2100).	gif-Format
5.	Comprehensive Dataset	CD-ROM	All available data by 5-year period: Main results for all countries and major regions; in 5-year age groups and five year estimation and projection intervals. In addition: all eight projection variants, including estimates and projections for AIDS countries; annually interpolated total population; fertility and mortality data by age and sex. See detailed description	Files in EXCEL format
6.	Extended Dataset	DVD	All available data including annually interpolated data: Detailed results for all countries and major regions; all data in 5-year age groups and five year estimation and projection intervals, most demographic indicators annually interpolated; all variants, including estimates and projections for AIDS countries; annually interpolated total population and annually interpolated population by single years of age and sex ("double interpolation"). See detailed description	Files in EXCEL format and in ASCII format for database input
7.	Special Aggregates	DVD	Medium: All major demographic indicators - aggregated for a large number of special country groupings, such as World Bank income groups. For details see the description of Special Aggregates.	Files in EXCEL format

6. Is the 2010 Revision using a target value of 2.1 children per woman for projecting fertility?

Fertility projection in the 2010 Revision is based on two probabilistic models. Only one of these models is using the fertility level of 2.1 children per women as an asymptote.

1. The first model is a Bayesian Hierarchical Model that is used for countries with high and medium fertility in 2005-2010. This model is not applying any kind of target value or asymptote. It only uses information about the distributions of the rates of decline from all countries of the world that have already experienced fertility declines. This model is projecting fertility declines for the high and medium fertility countries which reach their lowest levels often far below 2.1 children per women at some time in the future. Only after these countries have reached their lowest fertility level, the second model is applied.

2. The second model is a first-order autoregressive time series model (AR1) which is used for countries that have already reached a turning point in the rates of the fertility decline. For these countries the distributions in the rates of change are calculated from all those low fertility countries that have already experienced slight increases in fertility. In the long-term, the total fertility is assumed to converge toward and fluctuate around replacement-level fertility of 2.1 children per woman.

There is no empirical evidence to suggest that countries would have sub-replacement fertility for very long periods. We consider it more plausible that low fertility countries are “stabilized” with a fertility fluctuating around 2.1 in the long run. See assumptions for further details.

7. Are all countries projected to reach a total fertility of 2.1 children per woman in 2100?

No. The fertility projections in the 2010 Revision are not converging to a total fertility rate of 2.1 children in 2100 for all countries. The following figure illustrates the relationship between our total fertility estimates for the 2005-2010 period and our total fertility projections for the 2095-2100 period:

A. There are 13 countries with fertility above 2.1 in 2005-2010, which are projected to have fertility levels still above 2.1 in 2100 (right-upper quadrant).

B. However, most of the high and medium fertility countries in 2005-2010 are projected to have declined below a fertility level of 2.1 children per woman in 2100 (right-lower quadrant).

C. Almost all low fertility countries (below 2.1 children per woman) in 2005-2010 will still have a fertility of below 2.1 children per woman in 2095-2100 (left-lower quadrant)

D. There is not a single fertility country with below-replacement fertility in 2005-2010, for which we project a fertility level of more than 2.1 children per woman in 2095-2100 (left-higher quadrant).

8. Does the 2010 Revision project the decline of fertility in currently high or medium fertility countries to stop at 1.85 children per woman?

No. This was done in previous revisions. In the 2010 Revision of the World Population Prospects we have not set an arbitrary lower bound for the fertility decline. Instead we follow the results of the Bayesian Hierarchical Model, which predicts fertility for several high and medium fertility countries to fall below 1.85 children per woman - at least temporarily. The following figure illustrates the relationship between current levels of fertility and the projected lowest level of total fertility, which can be at different periods in the future.

A. There is only one (very high fertility country) country in the world for which we project the median fertility to not fall below 2.1 children in the projection period between 2010 and 2100 (upper-right quadrant).

B. In most countries we project that the fertility will fall below 2.1 children per woman at some time between 2010 and 2100 (lower-right quadrant). In many countries we project a decline of fertility to below 1.85 children per woman.

C. For a significant number of countries we have estimated that the fertility has already fallen below the reproductive level of 2.1 children per woman in the periods between 1975 and 2010 (lower-left quadrant).
All these relations refer to the median of 100,000 stochastic projections of total fertility.

9. Why does the 2010 Revision assume a slight fertility increase for countries with below-replacement fertility?

Based on the Estimates of the 2010 Revision there is empirical evidence that at least 21 countries with sub-replacement fertility have experienced slight increases in total fertility - after they had reached their lowest fertility level. Some of these countries have now slightly increasing fertility for several years. It should be noted that countries had been excluded from this list where total fertility rates have fluctuated without a clear trend. Only countries are listed that have shown at least two subsequent data points of (slight) increase in total fertility. Based on the (positive) rates of change in total fertility from these countries the time-series model (AR1) for the below-replacement countries was developed. Below is the list of these countries, which includes the five-year interval when the lowest level of total fertility was reported:

Belgium	2000-2005	Western Europe
Bulgaria	2000-2005	Eastern Europe
Channel Islands	1985-1990	Northern Europe
Czech Republic	2000-2005	Eastern Europe
Denmark	1985-1990	Northern Europe
Estonia	2000-2005	Northern Europe
Finland	1975-1980	Northern Europe
France	1995-2000	Western Europe
Germany	1995-2000	Western Europe
Ireland	1995-2000	Northern Europe
Italy	2000-2005	Southern Europe
Latvia	2000-2005	Northern Europe
Luxembourg	1985-1990	Western Europe
Netherlands	1985-1990	Western Europe
Norway	1985-1990	Northern Europe
Russian Federation	2000-2005	Eastern Europe
Singapore	1985-1990	South-Eastern Asia
Spain	2000-2005	Southern Europe
Sweden	2000-2005	Northern Europe
United Kingdom	1980-1985	Northern Europe
United States of America	1980-1985	Northern America

10. What are the main differences between the 2008 and the 2010 Revisions of World Population Prospects?

The medium variant of the 2010 Revision produces a world population in 2050 that is 156 million larger than that produced by the 2008 Revision (9.31 billion vs. 9.15 billion). Most of that difference can be traced to a higher number of births projected by the 2010 Revision (147 million more than the 2008 Revision over the period 2010-2050) and fewer number of deaths (22 million fewer than the 2008 Revision over the same period). In addition, the starting population in 2010 is lower in the 2010 Revision than it was in the 2008 Revision (by 13 million).

The difference in the number of births is mainly attributable to differences in projected fertility between the two Revisions. At the world level, total fertility in the 2010 Revision is slightly lower than in the 2008 Revision at the start of the projection period (2.45 vs. 2.49 children per woman in 2010-2015) but after 2020 total fertility is higher in the 2010 Revision and the difference between the two increases over time so that, by 2045-2050, the 2010 Revision projects a global fertility of 2.17 children per woman whereas the 2008 Revision projects 2.02 children per woman.

These small differences at the world level mask the important changes made at the country level. Whereas the projection of fertility in the 2008 Revision was done using four deterministic models of change which incorporated the strong assumption that the fertility of all countries would converge toward 1.85 children per woman, the projection of fertility in the 2010 Revision is carried out using two probabilistic models according to which fertility tends toward 2.1 children per woman in the long run and the future fertility paths for a given country are generated taking into account the specific past fertility trends in that country, those of countries within the same region and the trends in all countries.

The probabilistic approach permits the simulation of many future fertility paths for each country (100,000) so that their median can summarize a central tendency. The median path is used to generate the medium variant in the 2010 Revision. The results show that the 2010 Revision produces a more varied distribution of fertility levels in 2045-2050 than the 2008 Revision did. In particular, whereas in the 2008 Revision, 93 countries had a fertility of 1.85 children per woman in 2045-2050 and 111 countries had a total fertility ranging from 1.85 to 2.1 children per woman, in the 2010 Revision, just 51 countries have a total fertility in 2045-2050 within that range. In addition, whereas in the 2008 Revision just 32 countries had fertility below 1.85 children per woman and 18 had fertility above 2.5 children per woman, in the 2010 Revision the equivalent numbers are 86 and 39. That is, the 2010 Revision keeps more countries at both at the low and high ends of the fertility spectrum than the 2008 Revision did.

11. What is the estimation period for the 2010 Revision of the World Population Prospects?

The 2010 Revision provides population estimates for the period 1950-2010. Estimates of stocks are presented for every year that is a multiple of five from 1950 to 2100. Estimates of the components of population change are presented for five-year periods, starting with 1950-1955 and ending with 2095-2100. Stock estimates refer to 1 July of the year in question. Period estimates may be assumed to refer to the mid-point of the period concerned (e.g. the mid-point of the period 1 July 1970 to 1 July 1975 is the 1 January 1973). Data presented by single calendar years or single groups of age are derived by interpolation.

12. What is the projection period for the 2010 Revision of the World Population Prospects?

The 2010 Revision provides population projections for the period 2010-2100. There is only one series for the population estimates, but there are several series for the population projections, because different projection variants are calculated. Calculations are carried out by five-year periods using data classified by five-year age group.

13. Which empirical data have been used to prepare total population estimates?

For each revision of the World Population Prospects the Population Division is collecting, reviewing and, if necessary, adjusting population statistics from many different sources. The table below reports the latest data sources that were used to estimate total population.

	Source of latest total population data					Source of latest total population data
	Number of countries or areas					Total population in mid-2010 in million
Year	Census	Official Estimates	Other	Total	Year	Census	Official Estimates	Other	Total
<1980	3	0	0	3	<1980	29.7	0.0	0.0	29.7
1980-1989	6	0	0	6	1980-1989	230.0	0.0	0.0	230.0
1990-1999	7	0	0	7	1990-1999	222.1	0.0	0.0	222.1
2000	12	0	0	12	2000	360.0	0.0	0.0	360.0
2001	25	0	0	25	2001	490.0	0.0	0.0	490.0
2002	15	1	0	16	2002	210.0	0.0	0.0	210.0
2003	6	0	0	6	2003	22.0	0.0	0.0	22.0
2004	7	1	2	10	2004	18.0	0.1	15.0	33.1
2005	9	6	3	18	2005	280.0	27.0	18.0	325.0
2006	15	5	0	20	2006	180.0	25.0	0.0	205.0
2007	11	5	0	16	2007	210.0	210.0	0.0	420.0
2008	9	23	0	32	2008	1,600.0	1,500.0	0.0	3,100.0
2009	6	37	0	43	2009	30.0	390.0	0.0	420.0
2010	9	6	0	15	2010	130.0	170.0	0.0	300.0
2011	1	0	0	1	2011	6.2	0.0	0.0	6.2
Total	141	84	5	230	Total	4,018.0	2,322.1	33.0	6,373.1
Distribution	61%	37%	2%	100%	Distribution	63%	36%	1%	100%

2010 census round	60	82	3	145	2010 census round	2436.2	2322	18	4776.2
% since 2005	26%	36%	1%	63%	% since 2005	38%	36%	0%	75%

Source: Meta-information from United Nations, Department of Economic and Social Affairs, Population Division (2011). World Population Prospects: The 2010 Revision. CD-ROM Edition - Extended Dataset in Excel and ASCII formats (United Nations publication, ST/ESA/SER.A/306). For details see also: Meta-Information on CD-ROM

For the estimation of fertility, mortality and migration many additional data sources, such as surveys are used in the World Population Prospects. For details see: DataSources

14. Why are successive Revisions of World Population Prospects issued every two years?

Governments have instructed the Population Division to follow a biennial cycle for the preparation of population estimates and projections. All Revisions of World Population Prospects provide a historical time series of population indicators starting in 1950. The 2010 Revision provides projections up to 2100. Every new Revision takes into account newly released national data to revise estimates of past trends in fertility, mortality or international migration and to update population age distributions on the basis of newly available census data. For more information on the data used in preparing the 2010 Revision, please refer to Data Sources.

15. What country aggregates are available?

The 2010 Revision of World Population Prospects presents aggregated data and indicators for 4 development groups, 6 major areas and 21 geographical regions. In addition, data for special country groups are available on demand, including for sub-Saharan Africa, the land-locked developing countries and the income groups as defined by the World Bank income groups. These full set of special groups will be included among the data sets available for purchase on CD-ROM (to be released by August 2011).

16. Where can I find population estimates for periods before 1950?

For a series of world population estimates starting in year zero, please refer to Table 1: World Population From Year 0 to Stabilization in The World at Six Billion, United Nations, 1999, p. 5.

For a review of world population estimates prior to 1950 prepared by the United Nations see: John C. Caldwell and Thomas Schindlmayr (2002). Historical Population Estimates: Unraveling the Consensus, Population and Development Review, Vol. 28, No. 2 (Jun., 2002), pp. 183-204.

17. Why does the Population Division produce estimates of past population trends instead of using official national data?

World Population Prospects presents estimates for 230 countries and areas. About half of those countries or areas do not report official demographic statistics with the detail necessary for the preparation of cohort-component population projections. The Population Division undertakes its estimation work in order to close those gaps. The availability of data gathered by major survey programmes, such as the Demographic and Health Surveys or the Multiple-Indicator Cluster Surveys, are useful in generating some of the data that is not currently being produced by official statistics.

18. Why do the estimates in World Population Prospects sometimes differ from official statistics?

Official statistics are not perfect. All data have deficiencies. Official demographic statistics are affected by incompleteness of coverage, lack of timeliness and errors in the reporting or coding of the basic information. The analysis carried out by the Population Division takes into account those deficiencies and seeks to establish past population trends by resolving the inconsistencies affecting the basic data. Use of the cohort-component method to reconstruct populations is the major tool to ensure that the population trends estimated by the Population Division are internally consistent.

National Statistical Offices are well aware of the inconsistencies among data generated by different sources. Even in countries with advanced statistical systems, it is common for official statistical series to be revised retrospectively as new data become available and inconsistencies are corrected.

19. Why is only one variant of future mortality used in producing the United Nations projections?

Over the medium-term future, changes in fertility are more likely to have sizable impacts on future population size, growth and age structure than changes in mortality. Furthermore, whereas there is universal agreement that reducing mortality is a worthy goal, there are varied perspectives on what are the fertility trends best suited to satisfy the goals of different societies. Hence, there is more demand for exploring the effects of different trends in future fertility than there is for exploring those of possible future variations in mortality.

20. Who is using the results of World Population Prospects?

The users of World Population Prospects are many and varied. All entities of the United Nations system use the Population Division’s population estimates for the calculation of indicators that require population as an input. The Division’s population projections are also used in projecting other population-related variables or in modeling complex systems that use population is an exogenous variable. In addition, several entities and organizations of the United Nations distribute the results of World Population Prospects through their own databases and websites (e.g., the World Bank, the Statistics Division/DESA, the Food and Agriculture Organization). In addition, the data are used by many groups belonging to civil society, from school-children learning about population to journalists to academics. Being the official United Nations population estimates and projections, the results of World Population Prospects are considered to embody the authoritative view of population levels, trends and characteristics.

21. Why will volumes 1 and 2 of World Population Prospects no longer be available in print?

In an effort to "go green" and reduce costs, the United Nations Population Division will issue the two volumes summarizing the results of the 2010 Revision of World Population Prospects only in pdf format so that they can be downloaded via the internet. Volume 1 will contain the set of Comprehensive Tables showing the values of each demographic indicator over time and for all major areas, regions and countries. Volume 2 will contain Country Profiles displaying time series of relevant demographic indicators for each of the major areas, regional groups and countries at a time. A volume containing printouts of the sex and age distributions of populations will no longer be produced. The data by age and sex will be available only on Excel spreadsheets or through the interactive database accessible via the internet. The two volumes are expected to be released by the third quarter of 2011.

22. Which data sources have been used for estimating mortality in the World Population Prospects?

The following table gives the number of countries for which selected types of information are available for estimating mortality by type of information and period based on the 2008 Revision of the World Population Prospects.

Data sources as used in the 2008 Revision of the World Population Prospects

Notes: Includes only countries with a population of 100,000 or more in 2010. For smaller countries data sources for mortality estimation are even more scarce or lacking completely.
CEB/CS: Children ever born / children surviving. Refers to countries that collected summary birth history data in a census or in a survey that did not include a full birth history.
Source: Tabulation based on the United Nations, Department of Economic and Social Affairs, Population Division (2010): World Mortality Report 2009 (CD-ROM Edition, POP/DB/WMR/Rev.2009/2)

23. Which data sources are used for estimating fertility?

The fertility estimates for the World Population Prospects are based on the following sources:

Vital Registers
Most developed and some developing countries have vital registration systems, which record the number of births and the age of the mother. This information can be combined with estimates of the female population by age to calculate age-specific fertility rates and total fertility. Countries either report the (age-specific) rates or the number of births by age of mother.

Surveys
Most developing countries do not have vital registration systems, or they have only vital registration in urban areas. In those countries surveys are typically used to estimate fertility. The surveys are carried out by various institutions, from National Statistical Offices to international organizations. The most widely used surveys are the Demographic and Health Surveys (DHS) and the Multi-Indicator Cluster Surveys (MICS). In some countries, there are also Demographic Surveillance Sites (DSS), which provide in-depth fertility and health information for particular areas within these countries. Data from surveys are typically available in the form of birth histories (either full histories or histories truncated for the last 5 years) or as the number of "children ever born". Various indirect methods can be used to analyze and adjust fertility estimates from these surveys - particularly for historical time periods.

Censuses
The third major source of fertility information are the censuses, which typically provide information on the number of children ever born, or on the number of children born during the past 12 or 24 months for each woman age 15 to 49 enumerated. Census data, which provide full records for everyone living in a particular household, also allow application of various indirect measures to estimate fertility, such as the "own children method", which is based on the children of the women living in the household.

Adjustments
Fertility information from all data sources, including censuses, may require adjustments - particularly for historical time series. There are also various demographic techniques to adjust under-reporting in particular age-groups. One particular method attempts to reconstruct the fertility of cohorts across different censuses. For details of the methods used in the World Population Prospects see the section on Data Sources.