This discussion paper is a working paper, thus newer versions of this working paper may be published in the future before being finalized
The content was changed on the 26th of August 2021, updating the age-group analysis section of this discussion paper to reflect a more sophisticated analysis of the topic of the discussion paper, if you wish to see the previous version, please contact the author.
ABSTRACT:
“This discussion paper looks at the latest employment data published by LMRA. In particular, we look at Bahraini employment ratios derived using working-age population estimates from IGA. We find that overall, Bahraini employment ratios have decreased in recent years, with a particular notice in a decline of male employment ratios.”
MAIN BODY
Introduction
In this discussion paper, we are going to look over the latest labor market data. In particular, this paper will look at Bahrain’s employment ratio. What is the employment ratio? The employment ratio is the percentage of the working-age population (defined as ages 15 to 64) currently active in the labor market. The employment ratio can give us an insight into how the labor market changed over the past ten years; that is, the employment ratio is a good measurement of understanding how much of the applicable population are participating in the labor force. In the long run, it can give us insights into whether we see a job shrinkage or growth (OECD, n.d). However, the downside of the employment ratio is that it includes unemployed people who are not actively looking for jobs.
This discussion paper will look at the employment ratio for Bahrainis as an overall, then by gender, age, and gender & age. After looking at the stats, we will look at potential reasons for movements in the trends. We will not look at expatriate employment ratios. In general, the number of working-age expatriates is linked to the amount of expatriate labor in the country, where such labor requires a work visa to be present in the Kingdom. Thus, an expatriate employment ratio analysis would be more difficult to yield a reasonable hypothesis for such movements in the employment ratios.
We use the Information Government Authority (IGA) data via Bahrain Open Data Portal under the folder population for the working-age population. As for Employment data, we use Table A, A2, Table 8, and Table 9 from Labor Market Regulatory Authority Bahrain Labor Market Indicators (blmi) (Discussions on the limitations of using Table 8 and 9 are discussed further in the discussion paper).
This discussion paper finds that the Bahraini employment ratio has decreased over the past few years. In particular, we see that male employment ratios have decreased in both the public and private sectors. We also see that female employment ratios have slightly increased and remained stable throughout the periods we’ve examined.
An overview of Bahrain’s Employment Ratio
First, let’s start with how the overall employment ratio in Bahrain has been over the past ten years.
click to enlarge chart
In general, up till 2017, the employment ratio for Bahrainis has been relatively stable at around 38%. 2020 is the lowest point of the employment ratio with figures of 34%, respectively. To understand the drivers of the changes in the Employment Ratio, we can further break down the employment ratio into the Public sector and Private Sector Employment.
Page 1
click to enlarge chart
We can see that up till 2014, the employment ratio for the public sector had increased by 1.14%, while the Private sector employment ratio had seen a 0.86% decline. After 2014, we see that the Public sector employment ratio has continuously declined, where the effects were most pronounced in 2019. This is related to the oil price collapse in 2014, where the budget deficit had increased substantially, which had slowed down all public sector hiring. In 2018, the government announced voluntary retirement schemes for government employees, allowing government employees to retire early (BNA, 2018). We see the effects of this in 2019’s employment ratio data (where we saw a total decline of 2% for the public sector employment ratio).
How does the employment ratio overall look for men?
click to enlarge chart
Overall, the male employment ratio has declined substantially over the past ten years, where we see a near 9% decline in the male employment ratio. If we wanted to decompose the movement in such a trend between the public and private sectors, we see that both sectors had seen a decline in their respective male employment ratio.
click to enlarge chart
For males, we see that the public employment ratio has decreased by 5%. For the Private, we also see a decrease by nearly 4%. What is interesting is how the male public employment ratio has moved. Like overall trends in the public employment ratio (both male and female together), we see a similar peak in public male employment ratio movements in 2014 and the subsequent further decline from 2018 onwards due to the retirement package. The private sector employment ratio appears to have continuously declined over the years, with no real indicators of any significant events which had caused any further declines or increases in their employment ratio.
How does the employment ratio look for women? Overall we see that Bahraini women employment ratios have more or less been stable and that the ratio has increased by nearly 2% in the past ten years (graph on next page)
Page 2
click to enlarge chart
However, let’s look a little deeper at how the female employment ratio has moved regarding the public and private sectors.
click to enlarge chart
Overall, we see quite an interesting relationship between female public and private sector employment ratios. It appears that the two have some sort of substitution effect (although it would be premature to conclude such an effect is happening). It seems like as the female public employment ratio increases, we see a decline in the private employment ratio, albeit at less than a 1 to 1 rate (and vice versa).
It could be entirely possible that given public wages are generally higher than private-sector wages. When the fiscal position of the public sector was able to expand employment opportunities, women were more likely to enter the public sector than the private sector. However, the issue with this conclusion is that wage data is
based on Table B from LMRA, where the last update was Q1 of 2011, nor do they segregate public wage data by gender. Nevertheless, we see the effects of the oil price shock in 2014. It appears that public employment hiring rates had declined, while private sector employment hiring rates had increased, thus seeing this less than perfect substitution effect between the two sectors.
Overall, we see some interesting dynamics towards the Bahraini employment ratio. It appears that what kept employment ratio rates stable pre-2016 was that the public sector employment ratio had essentially increased. In contrast, private-sector employment ratios had generally remained in negative territory in terms of changes. Post-2016, we see that overall employment ratios had decreased, which can be attributed to the decline in public sector employment ratio from 2014.
The most interesting (and worrying) trend is that the male employment ratio has decreased, where the decrease is evident in both public and private employment ratios. What is also interesting is the appearance of a slight substitution effect for female employment ratios in the public and private sectors. This may explain why overall female employment ratios have not moved substantially and have remained relatively stable over the past ten years.
How has the employment ratio moved in terms of age groups?
Here we discuss how has each employment ratio moved for each age group. We use table 8 to discuss our analysis. The surrounding issue with our analysis below is that it only relies on private-sector data, as data for the public sector regarding the combination of age groups and gender is not available. Therefore we advise caution to apply the analysis towards the general Bahraini labor market. With this being said, the analysis of the private sector employment ratio for each age group and gender yields some interesting insights.
Page 3
Before we begin the analysis, we first need to note that we are not comparing the absolute changes in employment ratios in this section but the population-weighted employment ratios and their changes. This is because absolute changes can be misleading, where if there is a significant increase in employment figures for a minority portion of the population, then this would reflect a significantly large increase. Our method does the following for each year, we take the employment ratio for each age group in the private sector and multiply it by their respected population representation (% of overall working-age population), which would give us the population-weighted employment ratio for that particular year (and so on).
First, let’s look at the overall age-related employment ratios and their ten-year change. For simplicity, we’ve grouped our age groups as 18 to 24-year-olds (youth), 25 to 44-year-olds (middle-aged), and 45 to 64-year-olds (older-aged).
click to enlarge table
Here, we see that, as expected, the youth group has seen a decrease in their employment ratios, given that the youth are likely to prioritize education over the workforce. Interestingly, male employment ratios in this age group decreased further than that of females. It’s important to note that females within this age group were already unlikely to be employed at the beginning of our dataset. Thus the data reflects the slight decrease in our population-weighted employment ratio over the past ten years.
What is particularly interesting is the middle-aged group. We see that the female employment ratio had increased by 2.20% and that the male employment ratio had increased by only 0.80%. In this category, it appears that females are able to find employment more readily than their male counterparts. However, both genders reflect growth in their employment ratios.
Last, we look at the older age group. We see quite strikingly that male employment ratios had decreased significantly by 2%. In contrast, female employment ratios had more or less remained the same. This may be due to two reasons:
These age groups are struggling to find employment (i.e., the growth of total jobs is lower than the population growth of men from these age groups)
Men in these age groups are not actively seeking employment.
Page 4
Looking at our two hypotheses, we should look at the accumulated growth rates (adjusted for population weights) for each gender and their respective age groups. First, we look at men; below is the table of population growth and employment growth in the private sector:
click to enlarge table
From the table, we see that for youth, our expectations align where we see that their employment had declined by 4%. In comparison, the population in this age group had increased by 3%. For the middle-aged group, we see that their employment and population growth rates more or less align. However, we see that for the older group, their employment had declined by 4%, while their population total had increased by 9%. Again, this indicates that there seems to be an absolute reduction in the total employed older males within the private sector and that their movements in employment ratios are not necessarily due to potentially faster increases in population relative to their employment changes.
Next, let us look at females:
click to enlarge table
For females, we see that for the middle age group, their employment growth had outpaced their population growth, thus explaining the increase in their employment ratios over the past ten years. As for the youth, we see that little change was recorded with their employment figures, only seeing a 1% decline. As for the older age group, we see that employment figures increased by 8%, with the population increasing by 11%. Nevertheless, it appears that the middle and older age group are finding employment opportunities more readily or are more willing to work.
What is the main takeaway from this? First, we see that overall, females in the middle age group and older age group have seen higher growth rates in their employment figures than males. This is reflected in at least the middle-aged group employment ratio changes. Furthermore, the younger age groups' employment ratios appear to behave as expected; the youth prioritize education over employment. Last, we see that men in the older age group appear to be finding either less employment opportunities or are dropping out of the labor force altogether. However, it is essential to note that this analysis is incomplete. It only encompasses the private sector, and that more data is needed before applying it to the economy as a whole. In our next section, we will discuss the potential drivers.
Page 5
What could be the possible drivers?
From the data, some clear trends can be explained by events, some of which need expansion and others that need a deeper analysis on why they’re happening. Primarily, the movements in public employment ratios for both sexes are quite explanatory. When oil prices have collapsed, employment has halted, and therefore public sector employment grows slowly or even reduces. As the total working-age population is still increasing, the public employment ratio decreases as a result.
Next, let’s expand on why we have seen a general increase in women’s employment ratio over the past ten years and the potential drivers. First, it is essential to acknowledge that women in Bahrain have high access to education in Bahrain, which initiatives have started since 1928, with the opening of the first public school for girls (Al Gharaibeh, 2011). Furthermore, initiatives taken by the government to continuously increase female participation in education have had noticeable and profound effects, where illiteracy rates had fallen from 76% in 1971 to 11.76% in 2006 (Al Gharaibeh, 2011). If we look at University of Bahrain statistics data in 2016/2017, we see that females represent 65%, while males represent 35% (Alnaser, 2017). Furthermore, when we look at the tertiary enrollment by gender, we see that over-time female participation has increased compared to males (right-side).
click to enlarge chart
Although limited, with the data above, it seems that women in the country are more active in participating in higher education than men. In turn, when we looked at the percentage of “skilled” and “unskilled” women and men (skilled being those with post-secondary education, and unskilled with secondary education and below), we see that the trend in private sector employment mimics the tertiary, gross enrolment ratio above.
click to enlarge chart
Page 6
From the data, we see that the percentage of skilled workers in each gender mimics the movement in the gross enrolment ratio above, notably stronger in men than women. These indications give us a rough idea of what is happening in the labor force by gender. It appears that female employment ratios have remained stable and are increasing in the private sector for the middle age group as it may be that females entering the labor force are skilled compared to males, which see a decline and a smaller increase in their employment ratios in the private sector.
However, what about the older male age group? Why do we see significant declines in their employment ratios and their total employment numbers? Two potential reasons may explain this. The first explanation is that firms may potentially replace older men for younger and potentially higher-skilled females. From a firm's perspective, hiring women to replace older men would make sense. Such employees would bring higher productivity (due to potentially having higher education levels) and potentially lower costs. This is indicated by the gender wage gap. The wage gap is approximately 44%, meaning that skilled men earn 44% higher than skilled women (as indicated by data in table 24 of LMRA) as of Q4 2020. Thus from a firm's perspective, the entrance of women into the workforce who are higher skilled and have lower labor costs would make sense to replace an older force.
The second explanation is that potentially, older men in the workforce are simply retiring early. Reasons for this are because, as previous to the current entrance of women into the labor force, little competition may have existed for older men, especially
those who may be part of a managerial team. Thus, given that little competition existed in the past, for the given small labor supply that existed in the past (primarily men), older men in the private workforce would earn substantially larger wages and thus allow them to achieve higher savings. With higher savings, this may enable the older male group to retire early from the labor force, especially when faced with potential higher competition from our first hypothesis.
It is essential to note that both of these explanations are simply pure speculation and that without the relevant data, it is nearly impossible to measure or test. In reality, we have learned that men, primarily older males in the private sector, have dropped out in both absolute terms and when measuring their employment ratios. We also know that middle-aged males have had less of an increase in their total employment and employment ratios than females. Finally, it is important to note that a primary bulk of the age and skill-dependent analysis relies primarily on private-sector data. Public sector data availability is unknown to the author. Thus caution is urged not to apply the analysis towards the economy as a whole.
To test our hypothesis out, first, we need to have more data on a micro-level. The data should be panel in nature (where we can track the changes in job statuses and other characteristics of individuals across time in the labor force). This allows us to plot and perform the necessary statistical tests to determine the correlated factors towards changes in women’s and men’s employment statuses across the island across individuals and across time.
Page 7
Potential Macroeconomic implications of the trends found.
From the trends, it’s clear that underemployment may be a potential issue, especially in male employment ratios. First, it is important to note a link between employment and output via Okun’s Law. In Okun’s law, the relationship between output and employment is positively related. When output increases above its potential output, employment rates will rise above its natural employment rate (and vice versa). Where there may be potential underemployment due to structural issues surrounding the labor market (primarily indicated by the lack of skilled male labor), output will also be lower than its potential output, leading to potential lost output.
The second potential macroeconomic implication is that of long-term welfare support associated with underemployment. With long-term underemployment, potential job seekers who do not find employment may require additional welfare support. This, as a result, will have direct burdens on public finances directly, in addition to the opportunity cost associated with funds that may be used towards welfare support as a result of underemployment. Therefore, the need to rectify and further discover why male employment ratios are declining is essential, particularly for those in the “older” age group.
Potential Policy implications
Although limited in data, we can take some insights away. We can form some potential initial ideas on policy implications to curb the current trend of male employment ratios and education participation rates. First, we need to understand why male participation
rates in tertiary education are lower than women. We also need to explore why UNESCO data shows that male education participation had declined from 2011 to 2013. For example, in the UK, there are indications that boys do not hold education and schooling as high as young women do (Richardson, 2016). The difference in attitudes may explain the difference in educational achievements and participation rates in higher education. Studying attitudes and behaviors amongst young men in Bahrain towards tertiary education and what barriers they may face towards higher education is the potential first step towards solving male education stock and employment rates.
Assuming that if similar trends are found, then we may implement several policies to try and increase male tertiary education participation rates. Within the US, several papers have found that higher education participation increases by 3 to 5 percentage points for every 1000 US dollars spent on student grant aid (Dearden, Fitzsimons, and Wyness 2014). From the same study, using a difference-in-difference model, the authors have found that a 1000 GBP increase in grants leads towards a 3.95 percentage point increase in participation in the UK. Potentially, education aid or reduction of education rates for male students may be an effective policy towards increasing male participation rates in higher education. The most significant barrier to education participation is cost worldwide.
Another key policy is potentially one of challenging stereotypical roles of boys and girls, primarily in education. As highlighted in Richardson, 2016, there appears to be a difference between boys and girls and their attitudes towards education. A discussion
Page 8
summary by the Education hub and UNGEI in 2016 titled “Changing patterns in boy’s education” discusses the key issue of how boys are underperforming and drop out rates (The commonwealth education hub discussion summary, 2016). Their main findings are that education systems and parental figures should look at challenging stereotypes found in education systems and increase the amount of opportunity and support boys receive. Their discussion findings also point out that a set policy solution should not be homogeneous, given that a multitude of factors also affect how boys (and some which affect girls) affect their educational attainments and performances. These factors are socioeconomic statuses and other demographic characteristics. Thus a set policy should address these multitudes of factors in terms of gender, demographic characteristics, and socioeconomic status.
Overall, while these are reasonable first steps to look at how to address the education gap between males and females, more regional and local studies are needed to find more specific factors about the education gap between males and females. With more detailed studies, We can make a policy orientation to provide financial and other support to try and bridge the education gap. This may reduce the potential macroeconomic implications discussed above of underemployment and unlock potential productivity gains, which will positively affect individuals and the economy as a whole.
Conclusion
Overall, we see that Bahraini employment ratios have decreased in the past few years, in particular in later years of the 2010 era. We see that this is primarily around male employment ratios, wherein both the public and private sectors, we see a decline in employment ratios throughout the measured period. In the private sector, we see a significant and predominant decline in older-aged males regarding absolute employment growth and their employment ratios. We do not know if such declines are associated with the large entrance of women into the workforce, which may be associated with a substitution effect, or that these males are simply retiring early as hypothesized. Last, some of the differences in at least “middle-aged” employment growths and ratios can be explained by differences in education outcomes between males and females, where indications show that females outnumber males in tertiary education, and thus the percentage of skilled labor relative to the total amount of females and males respectively.
However, it is essential to note that this analysis is incomplete. More data is needed to precisely understand how employment ratios have changed throughout time for different genders and age groups. Notably, this requires public sector employment data. Other helpful data points would be those of the total number of individuals by education level and age. Combining the total number of employed personnel by age and education level can be particularly useful to measure the employment ratio for both genders by age and education level. Last, data on individuals across time would be beneficial to conduct panel data regressions that can open towards a more sophisticated understanding of characteristics of the labor market regarding Bahraini’s in particular. We can conclude that we know what has happened with male employment ratios. Still, we do not particularly know why such trends are happening other than indications of differences between education levels of males and females may partly explain the trends we see.
Page 9
SOURCES
Al Gharaibeh, F., 2011. [online] Vc.bridgew.edu. Available at: <https://vc.bridgew.edu/cgi/viewcontent.cgi?article=1115&context=jiws#:~:text=Women%27s%20education%20in%20Bahrain%20has,of%20primary%20education%20for%20girls.>
Alnaser, W., 2017. University of Bahrain Fast Facts. [online] University of Bahrain. Available at: <http://www.uob.edu.bh/en/images/About_UOB/fast_facts_25-1-2017_En.pdf>
BNA, B., 2018. Civil Service to highlight the newly announced voluntary retirement scheme for government employees. [online] https://www.bna.bh/en/. Available at: <https://www.bna.bh/en/CivilServicetohighlightthenewlyannouncedvoluntaryretirementschemeforgovernmentemployees.aspx?cms=q8FmFJgiscL2fwIzON1%2BDu%2BrXkOXmCNxKT296qiA8LI%3D>
Dearden, L., Fitzsimons, E. and Wyness, G., 2014. Money for nothing: Estimating the impact of student aid on participation in higher education. [online] ScienceDirect. Available at: <https://www.sciencedirect.com/science/article/pii/S0272775714000910>
IGA, I., 2020. Bahrain Open Data Portal, Documents. [online] Data.gov.bh. Available at: <https://data.gov.bh/en/ResourceCenter>
LMRA, L., 2020. By sex, citizenship. [online] Blmi.lmra.bh. Available at: <http://blmi.lmra.bh/2020/12/mi_data.xml>
Richardson, H., 2016. Why do more women than men go to university?. [online] BBC. Available at: <https://www.bbc.com/news/education-37107208>
The commonwealth education hub discussion summary, T., 2016. The commonwealth education hub discussion summary 2016. [online] Available at: <https://www.thecommonwealth-educationhub.net/wp-content/uploads/2016/04/Boys-Education-Discussion-Summary.pdf>
ABOUT THE AUTHOR
