Introduction

In this analysis, we explore and compare the birth rates, mortality rates, population dynamics, workforce characteristics, and overall health of four countries: the United States, China, Portugal, and Afghanistan. The United States and China are widely recognized as global powerhouses, marked by high levels of industrialization and technological advancement. Portugal, although also considered a first-world country like the US and China, differs due to its lower levels of industrialization and technological development. In contrast, Afghanistan is a third-world country that faces challenges in these areas. Our aim is to examine how these countries are similar or differ across the selected variables, especially as they relate to their level of economic development. While we do not make causal claims regarding the relationship between a country’s economic development and these factors, this analysis offers valuable insights into how these indicators have evolved over time and highlights emerging trends across diverse national contexts.

The data used in this analysis is a subset of the World Bank’s Global Development Indicators dataset which contains data on 1496 variables across 266 countries from 1960 to 2023. Within this vast dataset, we will specifically explore the variables related to mortality, nutrition, and population across the four previously-mentioned countries from 1960 to 2023. Specifically, the columns in the dataset represent broad identifying information about the country of interest, variable of interest, and the year that the data was taken. Each row represents one data point for some combination of country, variable, and year.

Our analysis aims to answer three main research questions: 1) How do the birth rate, mortality rate, and population of these countries differ by level of economic development? 2) How do nutrition levels compare between these countries and how have they changed over time? 3) How has the age dependency ratio (non-working-age individuals per 100 working-age individuals) changed over time for these countries?

Analysis

To begin, we are interested in the factors that have affected the overall population of each country, namely the birth and mortality rates.

The above graph shows the crude birth rate over time for each of the four countries, with each country measuring between 0 and 60 births per 1000 people per year. Across all plots, we see a general decrease in the birth rate over time. However, while the trends for Portugal and the United States, look very similar, China sees a large jump in the birth rate shortly after 1960, but then declines over time to similar levels as Portugal and United States at about 10 births per 1000 people per year. On the other hand, the birth rate for Afghanistan remains relatively flat at about 50 births from 1960 to 2000 before slowly decreasing to about 35 births in 2023. Overall, we see that the three first-world countries have all generally shown similar trends in birth rates, with a slow decrease towards about 10 births per 1000 people per year. This may align more closely with their status as first-world countries rather than their specific level of industrialization and technological advancement, something that sets Portugal apart from the United States and China. This further suggests that third-world countries may generally have higher birth rates, though Afghanistan’s birth rate has also been decreasing in the past 20 years, a potential indicator that their infrastructure is improving and that they may become a second-world country in the future.

We next examine the crude mortality rates of these four countries, similarly measured as the number of deaths per 1000 people per year. We see that mortality rates have been stagnant for all three first-world countries for almost the entire range of the dataset, hovering also around 10 deaths per 1000 people per year. However, we see that although China initially had high mortality rates in 1960, these rates dropped sharply within a few years and is now lower than that of the US and Portugal, at about 5 deaths. On the other hand, the mortality rates for Afghanistan has shown a continuous decline since 1960 and has actually reached levels below US and Portugal as well. This may suggest that mortality rates decreasing is either a precursor to a country becoming a second or first world country or that mortality rates are not closely linked to the level of economic development of the country.

Furthermore, these birth and death rates together suggest that population growth rate is greater in China and Afghanistan than in the United States and Portugal. We’ll next look at the overall population of these countries over time to compare the magnitude of these differences.

The above graph shows the 10-year moving average of the total population of each of our four countries. As seen, China’s population is currently, and has always been, much greater than the three other countries at about 1.3 billion people. The United States population is second-highest at about 300 million people, while Portugal and Afghanistan both have populations below 50 million. China’s population, as predicted, has been growing quickly, though we see a slight curve, suggesting that their growth is slowing down. The population of the United States and Portugal appear to have fairly linear population growth over time, while Afghanistan’s population has been increasing faster in recent years. This aligns with our previous observation where death rates have decreased slowly to about 5 deaths per 1000 people per year but birth rates have only decreased slightly to about 35 births per 1000 people per year. Similarly, we noted previously that the first-world countries’ death rates have all dropped to similar levels, but China’s mortality rate was lower than that of the US and Portugal. This combined with China’s high population levels has led to the high absolute population growth seen in the graph.

Based on this analysis, we see that first-world countries typically have low birth rates and mortality rates leading to generally lower population growth rates. While China’s absolute population has grown significantly since 1960, the growth rate as a percentage is lower than that of Afghanistan’s, a third-world country. Finally, these population growth rates do not appear to depend as much on the specific level of industrialization and technological advancement as compared to being a first or third world country, as Portugal has seen slow population growth, similar to that of the United States and China.

We next explore the nutrition levels in these countries, a variable closely related to the mortality rates and population levels which we’ve initially explored. The above graph shows the undernourishment levels of our four countries, measured as the percentage of the total population that experienced mild or more severe levels of undernourishment. Although the World Bank only has data on undernourishment levels since 2001, we can identify clear differences between these countries, especially with Afghanistan as a third-world country. The United States and Portugal have historically always had an undernourishment level of 2.5%, while China initially started at about 10% before slowly decreasing to 2.5% by about 2010. On the other hand, Afghanistan had undernourishment levels above 45% in 2001, which decreased to about 20% in 2011 and then has since increased slowly to about 30% in 2023. Once again, we see that undernourishment levels are similar between the three first-world countries, but different from Afghanistan, our third-world country, and that the specific level of industrialization and technology does not appear to show a difference in this aspect.

In the previous graph, we observed that all four countries either experienced generally decreasing undernourishment levels or generally had low levels already. We want to put this into perspective of how undernourishment levels changed across all countries around the world. The two histograms above show the distribution of undernourishment levels across all 217 countries in the World Bank dataset with data in 2001, the first year that undernourishment levels were recorded. We then compare this to the distribution of undernourishment levels in 2023. Upon visual inspection, we see that the two distributions look fairly similar, with distinct right skew. However, as seen for the histogram for 2023, the density for the first bin, representing the least undernourishment, is greater than 0.15, while this same bin has a density less than 0.12 in 2001, indicating that more countries decreased their undernourishment levels in the past 20 years. To quantify these specific differences, we conducted a Kolmogorov-Smirnov (KS) test to determine whether the distribution of undernourishment levels has changed from 2001 to 2023 as well as a paired t-test to see if there is a significant difference in undernourishment levels between 2001 and 2023. Based on our analysis, both of these tests show significant differences in undernourishment levels between 2001 and 2023. The distribution of undernourishment level has changed significantly since 2001 (p-value 0.0026) while the mean observed difference of 3.80 is statistically significant (p-value \(7 \times 10^{-7}\)), indicating that undernourishment levels decreased over time. Together, this suggests that many countries saw improvements with general nutrition. Afghanistan’s undernourishment levels dropped by about 15% during this time period, much greater than the average decrease of 3.8% across all countries, which may indicate that Afghanistan is experiencing fast growth and may advance out of being a third-world country in the future.

Now that we better understand the main drivers of population growth in these four countries, we next explore how their age dependency ratio has changed over time. The age dependency ratio is defined as the number of non-working-age individuals per 100 working-age individuals, with working age defined as those between 15 and 64 years of age, inclusive. The above graph shows this ratio for China and the United States from 1960 to 2023, representing the trends for highly industrialized and technologically-advanced countries. Overall, we see that the age dependency ratio has decreased since 1960, indicating that, there are proportionally less non-working-age individuals and more working-age individuals compared to before. Framed differently, each working-age individual has less people that they need to support, on average. This could explain why these countries are such economic powerhouses if the increase in working individuals has helped them to accelerate their growth in industries.

On another note, we see that the age dependency ratio has increased slightly in both countries since about 2010. Given that birth rates have remained largely stagnant during this time period for these countries, this likely indicates that there are more individuals turning 65, and thus no longer being of working-age, than individuals turning 15, and entering working-age.

We then compare this with the age dependency ratios in Portugal and Afghanistan. Portugal’s age dependency ratio falls in a similar range as China and US’s, between 35 and 85 non-working-age individuals per 100 working-age individuals. It has also generally decreased over this time period, with a similar increase in recent years, though this increase started earlier, around 2000 compared to 2010 for China and US. On the other hand, Afghanistan’s age dependency ratio has historically been higher than these three countries, at about 80 to 110 non-working-age individuals per 100 working-age individuals. Furthermore, Afghanistan’s age dependency ratio shows an inverse relationship with that of China, US, and Portugal. Initially in 1960, China and Afghanistan actually have very similar age dependency ratios, at about 80 non-working-age individuals per 100 working-age individuals. However, while China’s age dependency ratio generally decreased before the slight uptick in recent years as mentioned earlier, Afghanistan’s age dependency ratio experienced the opposite. It increased steadily to about 110 non-working-age individuals per 100 working-age individuals, meaning that there were less working-age individuals and each one had to support more than one non-working-age individual, on average. Finally, when the age dependency ratios of China, US, and Portugal started to increase slowly, Afghanistan’s instead decreased dramatically, back to around 80 non-working-age individuals per 100 working-age individuals in 2023. This analysis suggests that a country’s age dependency ratio has high association with their level of economic development. However, a first-world country with less development like Portugal shows similar trends in age dependency as world powerhouses like China and the US.

Similar to our analysis of countries’ undernourishment levels, we want to put the age dependency ratios of our four main countries of interest into perspective by looking at how these ratios have changed across all countries in our dataset. As seen in the two histograms above, the general shape of the distributions look similar, though the distribution for 2023 is centered around a lower age dependency ratio. In addition, many of the countries with an age dependency ratio above 100, more non-working-age individuals than working-age ones, in 1960 have reduced them below 100 by 2023. Furthermore, while age dependency ratios in 1960 were all above 40, with a mean of about 80, age dependency ratios in 2023 reached as low as 20, with a mean below 60. We similarly conducted a KS test and paired t-test to quantify these differences. We similarly find that the distribution of age dependency ratios of countries has significantly changed since 1960 (p-value \(2.2 \times 10^{-16}\)) and that the mean observed difference of 21.76 is statistically significant (p-value \(2.2 \times 10^{-16}\)). In comparison, Afghanistan’s age dependency ratio during this time period actually increased by a few percentage points, while the average across all countries decreased by more than 20 percentage points. This may suggest that third-world countries like Afghanistan generally have higher age dependency ratios which they are unable to decrease while other countries like China, the United States, and Portugal have seen large decreases over time.

Conclusion

Our above analysis suggests that third-world countries tend to have higher birth rates, leading to faster population growth compared to first-world countries. Furthermore, they also have higher levels of undernourishment within their population, though this does not appear to have a large effect on their mortality rates. We also made an interesting observation that first and third world countries exhibit inverse relationships in their age dependency ratios. From 1960 to about 2010, the age dependency ratios of the US, China, and Portugal decreased while Afghanistan’s increased. Then, from 2010 to the present, this ratio for the US, China, and Portugal increased slightly while Afghanistan’s decreased. Across the different levels of economic development that we explored, we saw that a country’s status as a first or third world country affected these variables, while their specific level of economic development did not. Specifically, Portugal exhibited similar trends in all variables as China and the US, despite having overall lower industrialization and technological development.

We see a few areas of improvement in our analysis for future research. First off, we have only examined these variables, including birth rates, mortality rates, population levels, undernourishment levels, and age dependency ratios for four countries: the United States, China, Portugal, and Afghanistan. We chose these countries because we believe they represent a diverse set of levels of economic development, industrialization, and technological advancement, with the US and China representing the highest levels, Portugal representing a middle level, and Afghanistan representing a low level. For simplicity and clarity of our analysis, we only used data from 1 to 2 countries for each of the three levels of economic development that we were interested in. For future research, we would like to conduct similar analysis for other countries to see if these associations generally hold for other countries with similar levels of economic development as well. Furthermore while identifying reasons for these trends is outside the scope of our analysis, further research, for example on why Afghanistan showed an inverted age dependency ratio trend compared to the three other countries, could help us to draw stronger claims about how these variables are related to a country’s level of economic development. Finally, some of our variables like undernourishment levels had limited data as the World Bank only started collecting this data in 2001. If we had access to earlier measures or estimates of this data, we could further examine how this has changed over a longer time horizon.