Human development is a measure for general wellbeing, by focusing on equality, fair opportunity and availability of choice. The human development index that is developed by the United Nations Development Programme (UNDP), takes into account several dimensions of human wellbeing. To use UNDP’s own words: “Three foundations for human development are to live a healthy and creative life, to be knowledgeable, and to have access to resources needed for a decent standard of living. Many other aspects are important too, especially in helping to create the right conditions for human development, such as environmental sustainability or equality between men and women”.
A key condition for human development is participation in labour. The more jobs available and the better the conditions for accessing jobs would generated income and greater life satisfaction. However, there is an evident inequality in labour participation between males and females, that may be due to suppressed opportunities or lifestyle choice or a combination of the two.
I have always wondered whether there is ‘right’ mix between male and female labour participation. Surely not everyone should work, and traditionally there have been more men than female that worked at any given point in time. But I wonder whether this imbalance and the mixture has been changing over time.
This is an exploratory graph objective. I employ human development as a proxy for general wellbeing and examine whether there is some sort of convergence to a preferred balance between male and female labour participation in achieving comparative high levels of well-being.
Clearly, there are so many confounding factors that I do not take into account so you should take in the following analysis with a grain of salt.
Given the exploration of relation between labour participation and the Human Development Index (HDI), it seems only fair to exclude troubled countries and jurisdictions suffering from ongoing war and other sociopolitical extreme events. Human development is directly related to not only the development of capability but also to the opportunity to put to use these capabilities.
To determine which countries should be excluded from the analysis, I consult the Australia Department of Foreign Affairs and Trade’s classification of high risk jurisdictions for travel. I exclude those countries of which the majority of their landmass is classified as “Do not travel” or “Reconsider your need to travel”. This filter excludes 31 countries, as disclosed in the Stata do-file that is provided at the end of this page.
The data management management protocol involves merging the two sources of data and expressing the labour participation rates into proportions. I focus my attention on the HDI for 2017 and therefore I restrict the data to the period 1990-2017; before 1990 there is no reliable labour participation data.
I focus on the top 25 countries with the highest HDI and the bottom 25 countries with the lowest HDI. This is because the HDI index is designed to to be a relative type of measure that ranks countries from first to last and
and rank transition is gradual. It would be much clearer, therefore, to focus on the extremes of this distribution, whereby 25 countries account for about 15% of the total.
I use point implantations to encode the 2017 coordinates of the portion of female labour participation against the portion of male labour participation, as in a scatter plot.
I also use line implantations to encode the evolution of the relation between the female and male labour participation during 1990 to 2017, thus describing how the mix has changed over time.
The point and line implantations will be encoded using the colour retinal variable, thus contrasting the two categories of top 25 with the bottom 25 countries in two different colours.
I increase the size of the point implantation to place emphasis on the final destination as at 2017, and decrease the size of the line thickness otherwise the information would be too overwhelming.
The point implantations and the retinal variable of colour is internally identified through a legend placed at the top of the graph, just below the title. The purpose of the line implantation is identified through a note in the graph.
External identification includes a graph title stating the graph objective, a note acknowledging the data sources and the purpose of the line implantation, axes titles and regularly spaces axes labels from 0 to 1 n increments of 0.1 that enable table look-up.
No direct identification was deemed necessary, however I can see the value of directly identifying some of the countries in the graph. You can do so by adapting the Stata do-file provided at the end of this page.
The two important graph enhancement steps are the imposition of a 1:1 aspect ratio and the specification of a regular grid on both axis to enable quick and accurate table look-up.
The scale range shows the entire permissive range of variation for both proportion variables, from 0 to 1. Although there is not much data in the lower range of male labour participation this is an important piece of information by itself, i.e. that during 1990-2017 the minimum male participation in labour for these 50 countries was about 50%, yet the minimum in female for as low as 35%. That is to say, the empty white space is an important piece of information.
Here is my proposed solution:
Those countries with high levels of human development (orange colour) appears to be clustered together around a desirable labour mix in 2017, comprising of about 50-65% of females and about 60-75% of males. There is one exception – Iceland has more than 70% female participation and more than 80% male participation. Importantly, the trail during 1990-2017 suggests a covering path towards this mix.
Many countries with low levels of human development (blue colour) seem to be overworking with exceptionally high levels of labour participation, and most simply do not appear to be party to this desirable mix.
Download the Stata code for reproducing this analysis: labour_participation_hdi.do