dcyphr | Fertility, mortality, migration, and population scenarios for 195 countries and territories from 2017 to 2100: a forecasting analysis for the Global Burden of Disease Study


         The patterns for fertility mortality, migration, and population are instrumental to guide investment in infrastructure to accommodate population shifts. This study uses other statistical models and the researchers own propriety model to demonstrate the effect of education and contraception on the completed cohort fertility at age 50 years (CCD50), which is the average number of children born to an female from a specific age (15-49 years). This study uses the length of time to reach universal contraceptive and education access to produce varying models to more accurate represent reality. Generally, this study shows that if the pattern of female education and access to contraceptives continues, this will further the decline in fertility and thus population growth.



       Governments need short and long-term population forecasts to estimate needs for important institutions such as hospitals, business, schools, and public services. The largest population forecasts depend on two key factors: fertility decline in sub-Saharan Africa and countries that have a total fertility rate (TFR) lower than their replacement rate. FTR is the number of children born per woman if she passes through her childbearing years successfully. This study demonstrates limitations in previous forecasting models and demonstrates their own model on fertility, mortality, and migration in relation to changes in educational attainment and if contraceptive needs are met. 


The population from 2018-2100 for 195 countries and territories was projected.


The researchers took a model on mortality established by Foremann and colleagues, made minor changes, and extended it to 2100.


This study looks at the less obvious trends that appear in countries/territories with a lower total fertility rate (TFR) than the replacement rate of the population. This essentially means that when less individuals are being born than are able to replace the portion of the population. The TFR trends in relation to other variables such as education increase in complexity. Due to this complexity the researchers modeled fertility with the use of CCF50. This has already been defined as the average number of children born to an individual female from ages (15-49 years). This rate is more stable than the TFR and permits the researchers a less volatile variable to increase reliability of their model. The researchers also used urbanicity as an additional variable to help explain inaccuracy in the data, but it did not explain their findings.


Migration was measured as a result of SDI. SDI is death due to conflict, natural disaster, differences between death and birth rate, and a random chance variable. 2017 UN data on past migration patterns were used. However, the researchers note that the migration patterns are incredibly subject to enormous changes and are therefore are very uncertain in their quantification.



       The population was calculated by keeping the mortality, fertility, and migration rate constant for each age, sex, and location during a specific calendar year. Two other independent scenarios were created to account for total secondary education and universal coverage in contraceptive needs by the year 2030 (as indicated by the UN’s Sustainable Development Goal (SDG)). Gross domestic product (GPD) was also brought into consideration during the construction of their models. Similarly, other models were referenced for comparison.




         Large inequalities still persist in 2100. With life expectancies ranges from 69.4 years to 88.9 years for both sexes. The difference in life expectancy decreased from 6.9 years in 2017 to 3.6 years in 2100.

Global Fertility Scenarios

       The trajectory of fertility was highly dependent upon if the model was fast or slow with respect to access to contraceptives and educational attainment by the year 2030 as indicated by the SDG. 

Global Population Scenarios

         The researchers combined the models on fertility, mortality, and migration to forecast a population. If the educational and contraceptive needs are met quickly than the population as seen in the SDG model, then the population in 2100 is forecasted to be 6.29 billion. If the educational and contraceptive goals are met much slower, then the global population is estimated to be as much as 13.6 billion. The peak population in the SDG scenario was estimated to occur in 2045. All scenarios noted a large shit in the age structure of the global population from a mean age of 32.6 years in 2017 to 46.2 years in 2100.

Country-level migration scenarios

       The USA, India, and China will have the largest absolute number of immigrations. While Somalia, Philippines, and Afghanistan are predicted to have the largest absolute number of emigrations. Canada, Turkey, and Sweden will have the largest immigration rate while El Salvador, Samoa and Jamaica have the largest emigration rates. The countries with the expected largest life expectancies populations in 2100 are China, Bangladesh, Brazil, Ethiopia, the USA, and Nigeria

Regional and Country-level fertility and population scenarios

       Large variations were observed for the TFR by country and territory across the various scenarios. The top 10 countries reached peak population before 2100, except Nigeria is expected to peak near the year 2100. 

Economic consequences 

In all scenarios population declines caused a reduce in economic growth.



       Based on the researcher’s models on fertility, mortality, and migration the global population will peak in 2064 at 9.73 billion and decline to 8.79 billion in 2100. The access the education and contraception severely influenced the population particularly in high fertility countries like the sub-Sharan. The decline in population in the latter half of the century is good news for the global environment. However, climate change and environment issues will still have serious consequences in the coming years unless action is taken and passionately pursued. The large changes in population will have dramatic effects on the GDP of countries as the ratio of working age adults will shift wildly. This effect could be minimized by an increase in robotic autonomy. However, these effects are difficult to model. Labor force issues can be correct by a variety of methods including increasing the TFR or the age during which adults participates in the labor force. However, only increasing TFR is effective for long-term protection of the labor force. Countries that rely on immigration to maintain and grow their labor force such as the USA may be in danger in the future as the amount of willing emigrants could potentially decrease sharply.

Technical Limitations

         This study uses imperfect data, therefore this will lead to imperfect results. Past trends are not always predicative of future trends. This study does not account for large global changes induced by disease, war, climate change, economic collapse, or changes to migration policies. Future work is required for better quantification to improve estimation in modeling pathways.



       The global population will likely peak before the end of century. Some countries will have rapidly growing or decreasing populations. Meanwhile, others will struggle or succeed to maintain their population with liberal immigration and social policies. Most importantly nothing is set in stone, so predications made in this paper could be obsolete if large global changes occur.