Correlation between Life Expectancy and Fertility

Demographic Economics - Migration
Education - Health

The World of Science is surrounded by correlations [1] between its variables. This is why the growing importance of Data Scientists, who devote much of their time in the analysis and development of new techniques that can find new relationships between variables. It is important to emphasize that this task is not only reduced to just find a correlation between variables, but also it is equal or more contributive to find an explanation and/or possible consequence for these relations to occur.

Under this precept, the article presents a correlation analysis for the period of time (1962-2012) between life expectancy (defined as the average number of years a person is expected to live in given a certain social context) and fertility rate (average number of children per woman), that is generally presented in the study by Cutler, Deaton and Muney (2005), with the main objective of contributing in the analysis of these variables, through a more deeper review that shows if this correlation is maintained throughout of time, and if this relationship remains between the different countries of the world which have different economic and social characteristics. The results of the article affirm that this relationship does indeed hold as much in time as between developed and developing countries, as is the case of Bolivia, which showed a notable advance in the improvement of the variables of analysis.

The general idea of the analyzed correlation holds in general terms that a person with a high level of life expectancy is associated with a lower number of children compared to a person with a lower life expectancy, however this relationship does not imply that there is a causal relationship [2], since this relation can also be interpreted from the point of view that a person with a lower number of children, could be associated with a longer life expectancy.

Given this correlation, it is important to understand what are the possible channels or reasons for this particular phenomenon to occur [3]. One of the possible reasons is expressed in Figure 1, which supports the idea that globally: i) the notorious trend of having a longer life expectancy as well as a smaller number of children, is maintained somehow over time (if we compare the data for the year 1962 and 2012), which is equivalent to say over 50 years of comparison over time as the study of György & Nemeskéri (1990) called History of Human Life Span and Mortality also concludes, ii) On the other hand, countries with developed economies (Europe and Asia) tend to have a higher level of life expectancy and a lower fertility rate compared to countries with developing economies (Africa and part of America), both for data registered in 1962 as for the year 2012, bearing out what is expressed in the Office of National Statistics study (2005) called Trends in Life-expectancy by Social Class.

Following the analysis, Figure 2 shows the evolution of the relationship between the selected variables over time, for all the countries from American during the period (1962-2012).

The fertility rate between the period 1962-1970, presents a similar behavior that ranges from a value of 4 to 7 children on average. Accordingly, during the period 1980-2012 the average fertility rate gradually decreases until it reaches an average value of 1 to 3 respectively. In the case of Bolivia, the fertility rate, although it follows a downward trend over time like the rest of the countries in the region, it ends up among the 3 countries with the highest fertility rate in the continent for the year 2012.

Regarding the level of life expectancy, this variable reduced its oscillation over time, registering in 1962 a level between 50 to 70 years, while in 2012 registering a level between 70 and 80 years respectively. Contrary to the explanation of the fertility rate, Bolivia is among the countries in the region with the lowest life expectancy for almost all periods, except for the year 2012, when the country considerably managed to raise its level of life expectancy, being approximately among the average of the continent. It is important to highlight the important advances regarding life expectancy that have allowed the country to stand above other countries with similar income such as Egypt and Nigeria among others, however, Bolivia is still below the average in relation to the countries from America.

Another issue to be highlighted is how the correlation between the analysis variables loses strength over time, this due to the reduced dispersion of data in 2012, compared to the widely dispersed data recorded in 1962. This event can be explained by the notorious increase in the life quality of people, as cited in Hofman (2011) which indicates that a person with a longer life expectancy tends to have a smaller number of children, because gives a greater importance to their own integral self-care, and thus get more personal time to carry out projects that are very important for him/her, in other words, he/she gives greater importance to their own quality of life, since having a children represents a significant cost.

One of the main problems in a correlation analysis apart from the issue of causality already described above, is to demonstrate that the relationship is not spurious. In this regard, Doblhammer, Gabriele and Vaupel (2001) argues that one way to reduce the intensity of the mentioned problem, is to analyze these variables from other fields or branches of science. In that regard, I can highlight the study in medicine by Kuningas (2011) which concludes that evolutionary theories of aging predict a trade-off between fertility and lifespan, where increased lifespan comes at the cost of reduced fertility. Likewise, the study in Biology of Kirkwood (1977), concludes that energetic and metabolic costs associated with reproduction may lead to a deterioration in the maternal condition, increasing the risk of disease, and thus leading to a higher mortality. Finally, the study in genetics by Penn and Smith (2007), holds that there is a genetic trade-off, where genes that increase reproductive potential early in life increase risk of disease and mortality later in life.

It is important to note that the study by Cutler, Deaton and Muney (2005), also demonstrates its own reasons for clarifying that the relationship between the variables of analysis doesn`t correspond to a spurious relationship, however there is no doubt that this is a topic of discussion for the different branches of science that have analyzed these variables according to their own criteria.

1. Correlation: Measurement of the level of movement or variation between two random variables.
2. A causal relationship between two variables exists if the occurrence of the first causes the other (cause and effect). A correlation between two variables does not imply causality.
3. For the correlation analysis presented in the article, I considered the following control variables: income, age, sex, health improvement and population.

Share this