I was reading Ajay Shah’s blog the other day where he made the point that merely sending money down the pipe is problematic because the pipe leaks and we have no idea about what happens at the other end.

That’s true. Mere increased investments may not translate into improved outcomes in the presence of administrative inefficiencies, high administrative costs, and leakages which make accountability for outcomes nearly impossible. This is quite well known, and yet there is surprisingly little empirical data or analysis on the specific processes by which outlays translate into action on the ground. Not much is known in the public domain about planning processes and mechanisms through which expenditure priorities are determined, particularly at the district level. Information on fund flows is even scarcer.

We launched PAISA project as a step toward tackling these problems head-on, in the specific context of elementary education financing in India. We have tracked school level fund flows and expenditures under the Sarva Shiksha Abhiyan (SSA) with a focus on three types of grants which every government school in the country is expected to receive annually (PAISA Reports 2009, 2010, 2011)¹. These grants constitute 6-8% of total SSA allocation. To develop a holistic understanding of SSA finances, we undertook a more in-depth study in 9 districts spread across 7 states in India². One of the main findings of that exercise was that there is huge variation in fund flows across districts and even within districts (i.e. across the schools in the same district), as shown in the table below. In one of the previous blogs, I had discussed the factors creating variation in fund flow within district (i.e. at the school level). This blog looks at what is happening at the district level- what factors influence district level expenditure and whether district level expenditure has any implication for outputs at school level³.

What can explain variation at district level?

Clearly, State level financial performance (measured on the basis of quantum and timing of fund released by the State SSA society) should matter to the district⁴. Within the Indian federal structure, districts do not raise their own finances and are dependent on State for receiving their allocations. Thus it is plausible that the financial efficacy of a district is dependent on the financial efficacy of the State. The trend in the following graph confirms this hypothesis⁵.

Figure 1: Relation between State and district Financial Performance 

We also found that even though the relation between financial performance of State and district seems positive when yearly numbers are analysed, it is much weaker when one analyses half-yearly numbers. Our interactions with various district and State level officials have thrown up many possible reasons why a district may take much longer to spend what it has received such as, unpredictability in timing and quantum of fund receipt from State; Orders from the State/ Government of India/ Courts to prioritising certain types of expenditures; limited human resources to carry out expenditures;

We also thought that socio-economic characteristics of a district should also matter. However, a close look at the district expenditure and the socio-economic index suggests that there is no clear relationship between the two⁶. Jalpaiguri, Sagar, Kangra and Satara differ widely in their socio-economic indicators. Yet their expenditure performance is quite similar.

Figure 2: Relation between financial performance & socio-economic characteristics (district)

Thus it seems that an efficient State administration has the potential to overcome governance challenges faced by poorer districts. To illustrate, Jaipur and Udaipur have significantly different socio-economic characteristics as seen in the above graph. Yet, when it comes to expenditure efficiency, both districts rank amongst the highest in the State SSA releases and the district expenditures. On the other hand, Satara, Maharashtra, is the highest ranking among the sample districts in the socio-economic index. However, it is ranked below Jaipur and Udaipur in the State and district releases. It could thus be argued that efficient State level financial management can ensure efficient district level financial management.

Is there a relation between district spending and outputs in schools?

We analysed the links between the district expenditure and school level outputs- construction activity being undertaken in a school, and student and teacher attendance on the day of the survey. What did we find? The first graph looks at relation between district expenditure and % of schools undertaking infrastructure activity, while the second graph looks at relation between district expenditure and teacher and student attendance.

Figure 3: Relation between district finances and infrastructure activity in schools

Given that infrastructure deficit in schools (as per the RTE norms) is so high and allocation to infrastructure is second largest item in district budgets, one would expect a strong relation between the two. But that does not seem to be the case in our sample districts. To illustrate, Udaipur ranks higher than Nalanda and Purnea in terms of district expenditure but the percentage of schools initiating infrastructure activities is much lower in Udaipur than in either Nalanda or Purnea. In another example, Jalpaiguri’s expenditure performance is not very different to Sagar or Satara. Yet the percentage of schools starting construction activity is much higher in Jalpaiguri than in Sagar or Satara.

Figure 4: Relation between district finances and student & teacher attendance

Like the district infrastructure budget, here too, we find no clear association between how districts spend their teacher salary and student related budgets, and teacher and student attendance in schools. Jalpaiguri, Sagar, Kangra and Satara show similar levels of expenditure on these items. But student and teacher attendance are much higher in Satara and Kangra, compared with Jalpaiguri or Sagar.

Lack of strong correlation between district expenditure and school level outputs points to breakdown of accountability in public expenditure management systems for elementary education. Districts are driven by perverse incentives to show high levels of expenditure in their account books but have little incentive to actually monitor and facilitate spending in schools. In other words, districts are thus held accountable for ‘expenditures’ but not for the consequences of this expenditure. More worryingly, increased allocations and concomitant increased expenditures in elementary education seem to have little relationship with the day-to-day functionality of schools, measured here by initiation of infrastructure activities, and teacher and student attendance. These findings points to a larger failure in the current planning and budgeting system, as increased allocations seem to be disconnected from any real improvements in the functioning of schools.