Sample composition
At a glance
The findings in this chapter are based on the composition of the survey sample which is not necessarily representative of the public sector employees responsible for the four services included. It should therefore be treated with caution and any conclusions about the actual composition of the public sector should be avoided.
General information
1
Gender
Overall, our sample indicates that managerial posts are dominated by men with the exception of managerial posts in the social welfare department.
Women dominate frontline posts in the social welfare and health department whereas the rural development department across all levels is heavily male dominated.
2
Managerial posts are disproportionately held by those from the General caste category.
Managerial bureaucrats come from more privileged backgrounds whereas frontline workers are more likely to come from more marginalized socio-economic families.
While managerial bureaucrats in the social welfare departments are overwhelmingly female, most of them are coming from privileged socio-economic backgrounds.
Managerial positions
3
Our findings suggest strong sorting by gender, caste, and socio-economic background to administrative levels.
The Government of Bihar may want to consider additional measures to support women, those from marginalized caste backgrounds, and those from lower socio-economic backgrounds to aspire for and succeed in reaching leadership roles at the block- and district-level. It might also want to pay attention to renumeration and job security aspects of frontline staff under a lens of gender equity, especially in the case of the health and social welfare department. Additionally, attracting more women to the rural development department might be considered.
Recommendations
4
GENDER
Women are underrepresented in managerial positions.
Overall, the lowest paid and most fragile positions (in terms of renumeration and protection) are frontline positions in the health and social welfare department. These are almost exclusively dominated by women. Except for managerial positions in the social welfare department, more desirable positions (with better pay or some civil service protection) remain heavily male dominated and there is no apparent closing of the gender gap for managerial positions.
Our sample indicates a sharp difference in gender composition across departments and levels.
For school education, our sample has a higher share of teachers or headmaster who are male. This is likely an overestimate of the real gender imbalance as, given local gender norms, female teachers were less willing to participate in the survey and most schools had more than one staff present. Nonetheless, the data indicates that the share of female staff declines the higher one moves in the administrative level with the post of District Education Officer (DEO) being heavily male dominated.
Rural development is heavily male dominated across all administrative levels and has only a very low number of female staff.
Social welfare and health are both heavily female dominated at the frontline. This is not surprising given that their frontline staff for the positions considered in the survey are almost exclusively female. In the case of Social Welfare, this predominance of women also applies to managerial level. In contrast, block- and district-officials in the health department are predominantly male, despite a frontline workforce that is almost exclusively female.
GENDER COMPOSITION BY LEVEL AND DEPARTMENT
The figure below depicts the sample composition by department and level.
GENDER COMPOSITION OVER TIME
When considering gender composition by recruitment year, we find remarkable stability across recruitment years. Splitting the survey into three recruitment periods (before 2004, 2004-2011, and since 2011), we cannot find a closing of the gender gap for managerial positions although there is an increase for women among our pool of frontline workers. This is likely due to an expansion of health and early childcare services where frontline staff is female dominated.
CASTE
General castes dominate managerial positions.
We also find strong sorting to administrative levels by caste. While Other Backward Classes (OBCs) make up the bulk of frontline workforces across departments, General castes become more dominant among higher levels of the administration. This overrepresentation is particularly stark given the share of General castes being around 15.5% in the population, according to the 2022 Bihar Caste-Based Survey.
Overall, our findings indicate that General castes are dominant in managerial posts while the frontline has a large share of OBCs.
Below we exclude those who refused to provide information on their caste.
CASTE COMPOSITION BY LEVEL AND DEPARTMENT
The figure below depicts the sample composition by department and level.
CASTE COMPOSITION OVER TIME
When splitting caste composition by recruitment year, we can find an increase of the share of OBC and SC block-level officials over time but not for the district-level. For post 2011 recruitments at district-level, more than half comes from General castes.
SOCIO-ECONOMIC
Frontline workers come from less priviledged backgrounds.
We find strong sorting by socio-economic background to administrative levels.
Bureaucrats at managerial levels generally come from more privileged backgrounds.
FIRST GENERATION GOVERNMENT JOB
Only 10% of district officials in our sample report that they are the first generation in their family to get a government job whereas half of frontline respondents report that this is the case.
FIRST GENERATION LEARNERS
Generational educational capital also differs by administrative level. While only around half of frontline respondents report that both their parents are literate, this number is more than two thirds for managerial bureaucrats in our sample.
COMPOSITE INDEX
Taken together, a clear pattern emerges of socio-economic background predicting the administrative level. These aggregate measures by level, however, mask considerable heterogeneity by department. To provide a compact overview, we next consider the first principal component of all these measures combined to create an index that allows us to reduce the above information to a single measure.
LAND OWNERSHIP
While most respondents across all levels come from families who owned land, there are around 13% of the frontline respondents reporting that they come from landless families.
SCHOOLING
Considering the schooling of respondents themselves, similar differences emerge. 90% of frontline respondents report that they went to a government primary school whereas this number is 75% or less for managerial bureaucrats.
For our sample, in particular the social welfare department shows a strong sorting of socio-economic background to administrative levels. At the same time, the social welfare department, as indicated above, is the only department where women are the majority in managerial posts in our sample. Together with the present information, this indicates that women from more privileged backgrounds can secure managerial posts in the department. The education department has a sizeable share of lower socio-economic background managerial bureaucrats in our sample.