At a glance
We find concerning levels of closed facilities, especially for anganwadis and – most severely – for health (sub-)centers.
Closed facilities
1
Limits of tech surveillance
Despite close tech-based surveillance through real-time monitoring systems, services are frequently unavailable for citizens.
2
Worryingly, a higher share of closed anganwadis and health (sub-)centers is found in those districts which have a higher share of SC/ST.
Unequal service availability
3
The Government of Bihar may want to undertake steps to ensure the availability of staff at all facilities during opening hours. Instead of relying on tech-based surveillance, peer monitoring paired with improved working conditions (such as female toilets) alongside steady wages might provide pathways to ensure higher staff attendance and motivation. Recognising the limitation of top-down monitoring to enforce attendance appears essential and is supported by international research.
Recommendations
4
OPEN FACILITIES
In many districts, less than 60% of health (sub-)centers were open.
We next consider the first basic step for citizens to receive their services: open facilities. A closed health center does not provide any benefit to citizens. We first consider aggregate shares of open facilities at district-level. Overall, we find that schools are largely all open. In contrast, there are districts where 20% or more of anganwadis were closed when enumerators visited them. Health (sub-)centers are the most concerning with many districts where 60% or less of their health (sub-)centers were open.
We next project these findings onto a map to visualise how closed facilities are distributed geographically. This highlights again the severe issues for health (sub-)centers and, in several districts, anganwadis.
We do, however, find a statistically significant correlation with the share of SC/ST population in a district: districts with a higher share of SC/ST have a lower share of open facilities. As there is little variation in the openness of schools, these results are due to differences in the share of open anganwadis and health (sub-)centers.
Like previous exercises for resources at frontline sites, we next check if citizens in lower literacy-rate districts are more likely to find their facilities closed. We do not find statistically significant correlations between a district’s literacy rate and the share of open anganwadis, health (sub-)centers, or schools (albeit there is a slight upwards trend, indicating that higher literacy districts tend to have a higher share of open facilities, we cannot confidently reject the possibility that this is due to chance as the correlation is not statistically significant at the 5%-level of significance).
​
We find concerning levels of closed facilities, especially for anganwadis and - most severely - for health (sub-)centers. While our data does not allow to disaggregate the reasons for closure (which could include unsanctioned staff absence as well as vacancies or sanctioned leave for training or other purposes), the bottom line remains that from the point of view of citizens, undertaking the journey to a health (sub-)center to find it closed is highly unsatisfactory and could, in the worst case, result, e.g., in the death of a mother and her unborn child. There is therefore a pressing need to ensure continuous staff presence at both anganwadis and health (sub-)centers. Anganwadis and ASHA workers are under close tech-based surveillance through real-time monitoring systems but still frequently unavailable for citizens. This indicates the limits of top-down tech-focused monitoring, a limitation of centralised monitoring also highlighted by international research. A recent World Bank study on Bihar’s health system instead suggested to focus on steady wage structures and improved peer-to-peer monitoring (Khemani et al 2020). Our findings are in line with these recommendations as more than 50% of frontline staff reported to be unsatisfied with their salaries and anganwadi and health workers frequently reported that they are feeling on their own. As noted previously, anganwadi and health staff, consisting practically exclusively of women, is particularly poorly paid. We also highlighted the poor infrastructure in anganwadis and health (sub-)centers. Without working toilets and running water, it seems unjustified to put the onus of the blame for closed facilities on frontline staff. More research is needed to be better able to disaggregate the reasons for staff absence. In addition, closed facilities are more frequent in districts with a higher SC/ST share which appears particularly concerning. There is an urgent need to address these shortfalls in service provision to ensure that every woman, man, and child has access to essential health care in rural Bihar.
STAFF ABSENCE
Staff absenteeism remains a concern across services.
We next consider in more detail patterns of staff absence. In this case, we treat closed facilities as zero staff present. Overall, we find that schools and anganwadis have comparable staff absence above 20% whereas this figure is almost 60% in the case of health (sub-)centers.
​
We can, however, find no clear pattern of staff absenteeism correlating with either the literacy rate or the share of SC/ST in the population.
​
Below we present the geographical pattern of staff presence. It is particularly concerning that for some districts, staff presence in health (sub-)centers is below 20%.
Outputs & outcomes
