In the Indian public education system, block/cluster resource persons are mid-level officials who are responsible for monitoring the performance of teachers. In this note, Vivek et al. discuss their experience of working with the state government of Jharkhand to design a ‘results-based financing’ programme that incentivises these mid-level officials as well as teachers to perform better – with the ultimate objective of improving education outcomes.
The primary objective of school education is to ensure quality learning for all children. Teacher presence and classroom instruction are two of the most important school inputs necessary to improve student learning. Results-based financing (RBF) programmes – that is, programmes that reward favourable results in the delivery of education through monetary or non-monetary incentives – have shown significant promise in improving teacher inputs and student learning. However, not all RBF designs are successful, and existing literature offers several lessons to improve their effectiveness (Lee and Medina Pedreira 2019).
We worked with the state government of Jharkhand on an RBF design and its implementation for the state’s elementary school system. In this note, we present our learnings from the experience. While some of these may be context-specific, many are applicable across education systems stuck in a low-performance equilibrium (Pritchett 2015).
Role of mid-level officials in the public education system
Working within the rich kaleidoscope of India’s education system is humbling. There are about 9 million teachers in the Indian school system, each with varying levels of educational qualifications, pre-service training, and performance. The Government of India recruits block resource persons (BRPs) and cluster resource persons (CRPs) for supporting the teachers in the lower-primary (grades one to five) and upper-primary (grades six to eight) public schools.
BRPs and CRPs constitute the ‘middle-level management layer’ or ‘meso-level players’, and one of their key functions is to provide academic mentorship and on-site support to teachers. They are also responsible for regular on-site monitoring of schools; collecting and analysing data on school-level attributes, teacher performance, and student achievement; and mobilising local community members to participate in improving school management. However, several studies have highlighted that most of their time is spent on collecting data from schools, leaving little time for academic mentoring, and supporting teachers (Ministry of Human Resource Development (MHRD), 2010, 2011).
Job tenure of BRPs/CRPs and teachers are different with implications for asymmetric power dynamics – while most teachers are permanent staff protected by virtually lifelong employment, the BRPs/CRPs are contractual workforce with a short tenure and are easily replaceable. In addition, BRPs’/ CRPs’ remunerations are significantly lower compared to the teachers they monitor (partly because the former are hired as contractual staff), and they often lack formal training. The result is subpar motivation along with misaligned power dynamics between the two groups. Hence designing RBF with an aim to improve the eventual improvement of education delivery is complicated.
We designed an experiment in partnership with the government of Jharkhand where performance-based incentives are provided to BRPs/CRPs and teachers. The treatment also included training of BRPs/CRPs in building their capacities for academic mentoring and monitoring. The expected result is improved accountability and performance of teachers who are mentored by these BRPs/CRPs. In designing the incentive framework for this intervention, we encountered various considerations, especially around: ‘Whom to incentivise?’, ‘What to incentivise?’, ‘What sort of incentives to provide?’, and ‘How to monitor performance?’ These are described here and can serve as operational lessons for design efforts around incentive frameworks in other programmes.
Figure 1. Key players in the context of the study
Figure 1 captures the roles and responsibilities of BRPs/CRPs and teachers who in turn influence student learning outcomes (as measured by performance in math and language tests). BRPs/CRPs are responsible for monitoring and mentoring teachers who are responsible for regular attendance and classroom instruction, which together influences student learning.
Whom to incentivise?
The objective of the government was to incentivise BRPs/CRPs to improve teacher efforts. However, any incentive given to BRPs/CRPs conditioned upon their performance and/or teacher performance (as measured by attendance and or classroom instruction) is likely to fail because teachers as permanent staff have no incentive to coordinate and exert effort in activities that will give rewards to BRPs/CRPs who are short-term staff. The relationship between BRPs/CRPs and teachers is further complicated by the fact that BRPs/CRPs are – at least on paper – in a supervisory role for the teachers. For instance, a CRP that we spoke to said, “The teachers don’t respect us. While a few are openly hostile, most question our qualifications and training. How can we train them when we are ourselves untrained and contractual? Over time, some have stopped questioning us, but they don’t always listen to us”.
It was, therefore, important that the intervention found a way to provide incentives that extend to the teachers as well. This presents the first learning – multiple actors may need to be incentivised to achieve a single outcome, since the results chain flows through all of them.
What to incentivise?
The desired overarching outcome for our experiment is improving teacher presence and classroom instruction. Classroom observations are expensive, and complicated to be conducted repeatedly, at scale, with reliability and validity. Similarly, for BRPs/CRPs, the exact behaviours that might result in effective teaching efforts by the teachers under their supervision have not been analysed and empirically established in the contexts being studied. It is also necessary to ensure that the outcome measures are easy to understand and communicate, and stakeholders trust their reliability and validity.
Our intervention therefore used teacher attendance1 as the outcome measure for incentivising both teachers and BRPs/CRPs. These joint incentives are aligned in a way that they can motivate changes in behaviour (as in a standard principal-agent problem). While measuring teacher attendance might seem trivial at first, it does have its own challenges since it requires a systemic commitment to create and ensure compliance to a specific definition. What measures to choose for triggering incentives is, therefore, a strategic question that requires weighing the benefits of sophisticated measures against the constraints of limited resources and challenges in communicating complex measures to all relevant stakeholders.
Type of incentives: Monetary versus non-monetary
The selection of incentives depends on the behavioural responses expected from the incentives, as well as financial sustainability of the incentives in the public system.
In our intervention, we decided not to utilise monetary incentives, even though the BRPs/CRPs were not satisfied with their salaries. This is because closing the enormous gap between their salaries and those of the permanent teachers seemed to be a case for financial reforms in the education system, rather than a case for small incentives. Preliminary discussions appeared to indicate that small incentives would be ineffective. A BRP remarked, “I am paid around Rs. 12,000 per month. Teachers are paid anywhere between Rs. 40,000 to Rs. 100,000 depending on seniority. Even the peon who brings us tea in the office earns Rs. 48,000. We don’t tell anybody our salaries because we are ashamed of it.” In addition, discontinuation of monetary incentives after the intervention due to the government’s budgetary challenges could worsen the relationship between the State, and the teacher, and BRPs/CRPs.
Interviews and focussed group discussions before the intervention revealed that the BRPs/CRPs continuously felt that their work was not acknowledged. When speaking with us, a number of BRPs/CRPs mentioned this in statements such as, “Nobody has appreciated me in the last 14 years of my service”, “I remember once the Block Programme Officer had appreciated me”, “I still remember my first Block Education Officer had appreciated the work I’d done in re-opening some defunct schools”. Hence, this made social recognition in the form of certificates of excellence. To make such social rewards strongly salient and ensure ownership and continuation of incentives beyond the intervention, government officials were chosen for their distribution. This fit well with the local culture of hierarchy wherein recognition from senior government officials was a matter of pride.
Financial sustainability considerations, culture of the government system, and a need for buy-in right from the beginning, were critical in choosing non-monetary incentives distributed by senior government officials.
Enforcing credible verification criteria
Credible verification is one of the most important aspects of RBF since the incentives and/or financing are provided on the basis of achievement of results. Some outcome measures are more easily verifiable than others. Attendance is among the simpler ones: it is objective and can be easily documented without enumerator biases.
Even with suitable, credibly verifiable outcomes being chosen for incentivisation, there remains the question of gaming and cheating by the agents receiving the incentives. Benefits of RBF interventions may get severely compromised if there are possibilities of manipulating the conditions based on which incentives are obtained. For example, teachers could coordinate with the BRPs/CRPs to be present on those days when attendance monitoring is scheduled. Hence, an independent organisation was tasked with conducting random spot-checks of teacher attendance.
To conclude, while designing an incentive framework, it is important to identify all actors that impact the results chain; select outcome measures that are relatively objective, easy to understand, and credibly verifiable; provide effective and sustainable incentives, and have sufficient checks to prevent manipulation of the system.
This note is the second in a series of findings from World Bank’s “Technology-Enabled Strengthening of Elementary School Monitoring Systems in Jharkhand” project in the state of Jharkhand, India and funded by Results in Education for All Children (REACH) Trust Fund. The first note in the series presents the lessons learned from the launch of a mobile application by the Government of Jharkhand, to replace the pen-and-paper-based data collection by education administrators in public schools.
1. Teacher attendance was measured by direct physical verification of teacher presence within the first fifteen minutes of the enumerator’s entry into the school.
- Government of India (2010), ‘Role of Block & Cluster Resource Centres in Providing Academic Support to Elementary School’, Department of School Education and Literacy, Ministry of Human Resource Development, New Delhi.
- Government of India (2011), ‘Approaches to School Support and Improvement’, Department of School Education and Literacy, Ministry of Human Resource Development, New Delhi.
- Lee, Ling Jessica Diana and Octavio Medina Pedreira (2019), ‘Results-Based Financing in Education: Learning from What Works’, Working paper, World Bank.
- Pritchett, L (2015), ‘Creating Education Systems Coherent for Learning Outcomes’, RISE Working Paper 15/005.