Responsible researcher: Eduarda Miller de Figueiredo
Article title: LOSING PROSOCIALITY IN THE QUEST FOR TALENT? SORTING, SELECTION, AND PRODUCTIVITY IN THE DELIVERY OF PUBLIC SERVICES
Article authors: Nava Ashraf, Oriana Bandiera, Edward Davenport and Scott S. Lee
Sample size: 1,585 individuals
Location of intervention : Zambia
Sector : Health
Type of intervention : Position for formal public health employee
Variable of main interest : Number of home visits
Evaluation method : Others - Experiment
Policy Problem
Providing health services in remote rural areas is a challenge for all countries at different levels of development, as medical teams trained for the region have high turnover rates (World Health Organization, 2006).
In Zambia, in 2010, the health post had an average of 1.5 Ministry of Health employees, including those who were not permanent employees. As a result, the government created the position of Community Health Assistant (CHA), which is a position for formally trained government health workers rather than informal community health workers.
Furthermore, given the shortage of skilled jobs, the Zambian government program may have the added benefit of creating employment opportunities in these communities.
Assessment Context
Through the Community Health Assistant , 2 health professionals have been deployed in each of the 167 communities in Zambia's 48 districts, where the CHAs' main task is to visit families and refer them to health facilities as needed. Therefore, such a position requires medical skills – vital signs, diagnosis, screening for common illnesses – and social skills – counseling, support, patient education –.
In this program adopted by Zambia, health workers join the ranks of civil servants, enabling advancement, through additional training, to higher and better paid positions. Thus, the monthly starting salary is $290 for CHAs and $530 for entry-level nurses. As CHAs are part of the public service, they are eligible for in-service training, that is, they attend school (medical or nursing school) while continuing to receive a salary or some type of tuition sponsorship.
The Ministry of Health periodically requests district doctors to nominate candidates based on merit, with no connection to performance measures. Therefore, promotions to senior management are not automatic, but have a high expected return, as employment opportunities that allow a career in government are rare in the remote communities from which health professionals are recruited.
This model was chosen by the country's government in the hope of attracting agents with strong technical skills to do community work, however, it ran the risk of excluding candidates motivated to help the community. And it was this trade-off that encouraged the experiment carried out by the authors.
Policy Details
Aiming to assess whether a career in the public service attracts talents towards pro-sociality and whether this affects who is hired and their performance, the authors designed an experiment, given the relevance of evaluating the role of selection in the provision of public services. In this experiment, the main challenge was to separate the selection effect from the effect of incentives at work. To achieve this, the study approached it in two stages: (i) opening the selection channel; (ii) closing the incentive channel.
For this purpose, recruitment posters and informational materials were used and distributed, with detailed instructions, to health agents.
The treatment poster emphasizes the public service identity in the new CHA position, relating it to the benefit of the opportunity to move up the public service career ladder. Furthermore, the material also explicitly encourages a feeling of belonging to public service.
The control poster uses the standard community health recruitment approach, emphasizing the social identity of the position, highlighting the impact of acquiring skills to promote the health of your family and neighbors.
Therefore, what changes between treatment and control posters is the wording of the benefits.
All successful candidates are eligible for career opportunities once hired. After recruitment, agents train together for a year, receiving the same information about career opportunities that public servants are entitled to.
Therefore, treatment and control have the same incentives when contracted, differing only in the form of selection.
Methodology Details
Because recruitment was organized by district officials, the authors randomized treatment at the district level, thereby maximizing compliance along with experimental assignment. The 48 districts were divided into two groups, implying that each employee is exposed to just one treatment.
The recruitment campaign yielded 1,585 applications from candidates who were eligible for the position, as they were required to provide information on skills, career ambition and pro-sociality. This information was used to measure the effects of treatment on candidate pool composition. To measure prosociality, the authors combine the candidate's self-reported willingness to remain in the community long-term with the Inclusion of Others in Self (IOS) scale that measures alignment of interests (Aron et al., 2004).
Results
Evidence suggests that career opportunities attract candidates who have different skills, career motivation and prosociality, with a greater propensity to choose candidates who rank highly on all three points.
The results also demonstrate that agents attracted by career opportunities are more effective, providing more inputs at the same cost. They also noticed that there was an increase in the number of children undergoing health examinations, which became 24% higher in the treatment group. Furthermore, the presence of CHAs showed an improvement in a series of family health practices.
Regarding home visits, there was a large effect, in which health professionals recruited through career opportunities carried out 94 more visits, that is, 29% more than the control group over 18 months. Given that for the majority of these families, these recruited health professionals are the only health service providers, the intervention could have positive implications for the health of the community.
Another point is that the program leads to a substantial increase in the number of health professionals operating in communities, an increase from 1.5 to 3.5 on average. Given the size of the increase and the magnitude of the effect, the authors expected the intervention to increase the utilization of healthcare facilities and practices, through increased demand, behavioral changes, provision of information, etc. In turn, better use of healthcare facilities and practices leads to better outcomes.
The results indeed suggest that career opportunities improve the utilization of healthcare facilities. And when it comes to the number of women giving birth, it was observed that there was a 30% increase compared to the average in the control areas.
On the other hand, it was observed that agents in the career incentive treatment group may leave their positions before agents in the control group. So whether this incurs a welfare cost depends on whether they can be easily replaced and whether the government can use their skills in other roles.
Public Policy Lessons
The results indicate that offering a position in a public service with career opportunities attracts agents who provide health services, causing a significant impact on the health of the community. And, as health work generates more social value, a position with the possibility of growth in the public career, it improves social well-being.
Reference
ASHRAF, Nava et al. Losing prosociality in the quest for talent? Sorting, selection, and productivity in the delivery of public services. American Economic Review, vol. 110, no. 5, p. 1355-94, 2020.