Grand theory of state capacity - What determines the implementation of a public policy?

State capacity can be defined as the capacity to make rules or policies and implement them. It is argued that weak state capacity is the critical constraint in delivery of education but our bureaucracy also amuses us with miraculous feats of organising world’s largest democratic elections, which seems like a paradox. Often, we mistake the capability to conduct elections for capability to deliver education. Hence, it is useful to understand this phenomenon in more detail, which we call it elections vs. education paradox from now on.

Kurt Von Hammerstein Equord, is a famous German army general. He served in the army during World War I and is known for his rivalry with Hitler and his plots to overthrow Hitler. He once remarked, “I divide my officers into four groups. There are clever, diligent, stupid, and lazy officers. Usually two characteristics are combined. Some are clever and diligent—their place is the General Staff. The next lot is stupid and lazy—they make up 90% of every army and are suited to routine duties. Anyone who is both clever and lazy is qualified for the highest leadership duties, because he possesses the intellectual clarity and the composure necessary for difficult decisions. One must beware of anyone who is stupid and diligent—he must not be entrusted with any responsibility because he will always cause only mischief.”  His statement can be visualized in a two-dimensional framework as shown in Figure 3.4.

                                                                    Figure 3.4

Our policies can also be analyzed using this framework. Clever is analogous to policies which are good at design, stupid are those which are bad at design. Diligent is analogous to policies which are implemented well and lazy is analogous to policies which are not implemented well. Like people, policies also come in combinations: good in design but bad in implementation (clever and lazy); good in design and good in implementation (clever and diligent); bad in design and bad in implementation (stupid and lazy); and the most dangerous–bad in design and good in implementation (stupid and diligent).

We thus observe that policy design and implementation together determine outcomes and that design of policies is a factor that should be considered in our analysis of election vs. education paradox. However, this analogy doesn’t tell us much about the interaction between policy design and implementation.

Policies interact with implementation through rules.  These rules governing the functionaries who are responsible for implementing the policy can affect their functioning. This can be best understood by observing the daily functioning of frontline employees in education, Cluster Resource Centre Coordinators (CRCCs). CRCCs in the education administration are expected to provide continuous academic support to teachers. This would mean that CRCCs have to regularly interact with teachers, understand their problems and give them appropriate support, which is a job involving adapting to the situations and responding accordingly. Researchers from Accountability Initiative, Centre for Policy Research tracked CRCCs (time-use study) to understand the nature of the jobs that they actually end up doing in reality. This survey finds that, in reality, “CRCCs spend less than 10-20% of their time inside classrooms. For the rest of the time, they are busy checking registers to collect data required by their superiors. Even in their monthly meetings, they received orders only related to administrative duties”. In another set of interviews, the frontline workers shared that they have a limited role in decision making and their job is to only implement orders properly and in the end whatever the government wants will be implemented.[i]

We observe that people who are supposed to perform tasks involve engaging with others are being monitored by rules misaligned to the nature of task and functionaries are being made accountable as per these rules. We must note that the norms governing CRCCs are not mandated by any particular public policy or scheme. Thus, we must distinguish between two types of rules - rules mandated by public policies and rules that are inherent to the functioning of the lower level bureaucracy. Rules discussed above correspond to inherent norms of the bureaucracy. Professor Akshay Mangla of Harvard Business School studied this phenomenon of bureaucratic norms, inherent to bureaucracy, in the context of Himachal Pradesh and Uttarakhand. 

Himachal Pradesh and Uttarakhand are both hill states and have similar socioeconomic characteristics as shown in Table 3.4

Table 3.4

Himachal Pradesh
Population (million)
Number of administrative districts
Sex ratio (females per 1000 males)
Annual per capita income (Rs.)
%Scheduled Tribe
Source: Census of India (2001; 2011). Reserve Bank of India, Handbook of Statistics on Indian Economy, 2010‐11.

Both states have similar education policies too. Policies which are implemented nationwide like mid-day meal scheme are in implementation in both the states. Sarva Shiksha Abhiyan (SSA) provides for schools within 1-3 km range of habitats. The hierarchy in education administration is also similar. The administrators at the top level are from the Indian Administrative Services (IAS), both states have Village Education Councils (VECs). Both states achieved near universal primary enrolment. Though they display similar characteristics, the learning outcomes in these two states vary significantly.

Table 3.5

Himachal Pradesh
% Fifth graders who can read a basic paragraph
% Fifth graders who can do basic arithmetic
Source: ASER

Akshay Mangla conducted more than 500 interviews and focus group discussions with policymakers, bureaucrats, teachers and parents to understand the mechanisms behind the difference in learning outcomes of these two states, despite having similar characteristics. He finds that “bureaucratic norms—unwritten rules that guide public officials—influence how well state agencies deliver services for the poor.”[ii] He illustrates many examples that depict this phenomenon.

For instance, in Himachal Pradesh, a small nomadic tribe, the Gujjars migrate to hills in summer to herd their cattle and return to plains in winter. This affected their children’s education. Local officials noticed this and responded to it by creating a mobile primary school, which would travel along with them, with teachers and learning materials. Local officials were encouraged by their seniors to take special time out of their schedule to visit this school and help implement the programme. The officials also hired volunteers outside official recruitment channels. In contrast, in Uttarakhand, a local NGO, Uttarakhand Seva Nidhi (USN) helped community to establish child care centers (balwadis) and union government funded them. This was before the state of Uttar Pradesh was divided. After the formation of Uttarakhand, officials started scrutinizing balwadis forcing them to adhere to rules pertaining to place, timing and schedule of operation of centres. These balwadis were in between residential areas where people felt accessible but officials wanted it to be placed inside a school as per the rules, and so on. Over time, these had to be closed.

In one state, local officials used discretion, went out of their way to help people. While in the other, they used rules to stifle well running community programmes. This interplay between rules and discretion or autonomy shaped by unwritten bureaucratic norms, Professor Mangla argues, influence the quality of service delivery of poor in these two states. We now have disentangled one more layer of the complex knot; rules governing the employees is the second factor that can explain the elections vs. education paradox.  

The lesson from the above story is that inherent norms of bureaucracy influence implementation but it does not necessarily tell us the nature of rules and the contexts in which they can be applied. Francis Fukuyama, noted political scientist elaborates on this in his famous essay ‘What is governance?’[iii].  Fukuyama explains - if one has to procure an advanced strike fighter, one way to do this is to just give a mandate to procure an advanced strike fighter to the bureaucrats and not put any additional constraints. This is the case of a completely autonomous bureaucracy, where just goals are set (procure an advanced fighter) and bureaucracy is responsible for figuring ways to achieve that. The other way could be to mandate “purchasing a strike fighter using contractors that increases employment in districts X and Y, or through minority and women-owned businesses or to achieve Z degree of performance desired by a rival service.” This is the case of a non-autonomous or a subordinate bureaucracy which is micromanaged with a myriad of rules and regulations.

Autonomy, as in the first case, gives scope for innovation and experimentation; at the same time, it can be misused. Subordination, as in second case, can ensure accountability but can lead to inefficiencies. Thus, Fukuyama suggests that the relationship between autonomy and government quality[1] looks like an inverted U as shown in Figure 3.5.

Figure 3.5

Source: Figure 1: Bureaucratic Autonomy and Quality of Government, Fukuyama (2013)

At one extreme, the bureaucracy has little autonomy and is bound by myriad of rigid rules and regulations, making it ineffective. At the other extreme, bureaucracy has high autonomy where it sets not just internal procedures but goals as well and can escape political control. In between these extremes, there is an optimal autonomy point. Fukuyama further argues that the position of optimal autonomy point is decided by the underlying capacity of bureaucrats, bureaucratic. Bureaucratic capacity, which will be referred to as capacity from now on, is different from state capacity. It can be thought of in simple terms as – if given a task, how confident are we that functionaries will implement it without resorting to corruption or evading work. If the bureaucracy is full of political appointees, then one would not want to give too much discretion to them as they may misuse it. Hence, the optimal autonomy point would be towards left in those contexts. In contexts with higher capacity (bureaucratic capacity), the optimal autonomy point would be towards the right as we are confident that functionaries will not misuse autonomy and this autonomy may improve efficiency. In summary, Fukuyama argues that discretion to employees should be reduced in contexts with low capacity to avoid misuse and autonomy should be increased in contexts with higher capacity to improve quality. Figure 3.6 shows what the optimal levels of autonomy for different bureaucratic capacities would look like.

Figure 3.6

Source: Figure 2, Optimal levels of autonomy for differing levels of capacity, Fukuyama (2013)

In Figure 3.6, capacity level 4 is greater than capacity level 3, which is greater than capacity level 2 and so on and the dotted line connects all the sweet spots of different curves. The framework in general would look like the graph in Figure 3.7.

Figure 3.7

Source: Figure 3, Autonomy and Capacity, Fukuyama (2013)

Countries above the line have to be moved to the right to bring them on to the line of sweet spot without reducing capacity. It means that they have excess rules in relation to their capacity and hence should be given more autonomy. Similarly, countries below the line have to be moved to the left to bring them on to the line. It means that these have excessive autonomy and have to be bound by some rules. Fukuyama then hypothetically locates various countries on the graph (Figure 3.8).

Figure 3.8

Source: Figure 4, Reform Paths, Fukuyama (2013)

Fukuyama points to the complexity in locating India on this map because there are cases like spectrum auction where more rules have to be formed to prevent misuse of discretion and there are also cases like local governments where more autonomy has to be given to enable effective functioning. This, he argues, underscores the need for context-specific decisions.

Locating India on Figure 3.8 is probably looking complex because we are considering only the variation of capacity of bureaucracy in our analysis and assuming that nature of all tasks performed by the government is same. Lant Pritchett, Professor at Kennedy School of Government, Harvard University categorizes the tasks performed by government[iv] as in Table 3.6.

Table 3.6
Tasks of all different types in each sector of government engagement
Basic education
Financial Sector
(policy-making, elite)
Setting a curriculum
Tertiary hospitals
Appellate Courts
Central Bank policy
Policy of setting rates, services
Building school buildings
Vaccination of childhood diseases
Notary services
Inter-bank payments
Delivering the post
Implementation-intensive service delivery
Classroom teachers
Ambulatory curative care
Lending to SMEs
Implementation intensive imposition of obligations
Supervision of teachers
Regulation of drug retailing
Regulation of private banks
Wicked hard
Raising performance
Promoting behavioral change (e.g.weight loss)
Dispute resolution
Financing entrepreneurship
Source: Pritchett, L. (2012)[2]

Concentrated tasks are those which require only a few people and can technically perform them by sitting in a closed room anywhere. Tasks of logistics category are those where there is a clear established path to reach the goal, which can then be codified easily into rules. Tasks of implementation-intensive service category are those which require constant engagement with citizens and require their cooperation, where functionaries have to constantly iterate and adapt to the circumstances (iterative adaptation). Implementation intensive imposition of obligations are those which require engagement with public but in forcing them to do something, not necessarily through cooperation. Wicked hard problems are the hard set of abstract problems like raising performance.  One can now place India on Figure 3.8 for each of the above categories of tasks, which then resolves the dilemma posed by Fukuyama.

We now have the third factor, nature of tasks and bureaucratic capacity, which can explain the elections vs. education paradox. Most tasks involved in administering elections belong to logistics category where there are clearly established standard operational procedures. Employees can be administered by codifying standardized rules and enforcing them. On the other hand, crucial part of delivering education falls under implementation intensive service delivery which requires discretion on part of the teachers and hence autonomy. This poses complexity in contexts with weak capacity because the nature of tasks require autonomy to employees but a rule based approach is preferable because employees in weak capacity systems misuse the autonomy, as understood by Fukuyama’s framework.

Additional complexity arises when a functionary has to perform tasks of multiple natures. Consider the example of CRCCs, a part of their work involves visiting classrooms and mentoring teachers which is adaptive in nature (implementation intensive service delivery). Other part of their work also involves supervising teachers (implementation intensive imposition of obligations). They are also required to collect data and implement programmes, which belong to the category of logistics, requiring them to adhere to rules and time bound performing of duties. In such scenarios, when functionaries are hard pressed to perform tasks of multiple nature, they act to prevent the worse from happening rather than excelling at each of them. In this attempt, they tend to prioritise urgent tasks (on-time delivery of data, implementing programmes), those where ramifications for non-performance are visible and affect them in short term, over important tasks (coaching teachers), where ramifications for non- performance are not visible. Supervisors also prefer monitoring them on those tasks which are easy to validate, like data collection, carrying out administrative duties.

When bureaucracy takes up programmes on mission mode, limited set of relevant aspects are prioritized by the top officials giving a strong justification for functionaries to keep other tasks aside for some time and just focus on one aspect of their job.  This re-adjustment of focus can yield results in short term. The other tasks cannot be put on hold for long time and hence once the project is over or when there is a change in leadership, functionaries are back to their multi-tasking mode, into a state of comfortable yet undesirable equilibrium where an average supervisor wishes to monitor based on rule based approach, setting expectations for CRCCs and an average CRCC responds by prioritising such tasks over important tasks like coaching teachers.

This gives us the fourth and fifth and sixth factors that can explain election vs. education paradox – consequences for non-performance, focus and time frame of the task. In case of elections, the non-performance can result in serious visible consequences, while in education, not mentoring teachers doesn’t result in any such. In case of elections, special focus is created by putting other tasks of functionaries on hold since it is only for a short period of time and hence manageable. In case of education, it is a long term process and functionaries have to multi task as it is not possible to create that focus by putting other works of functionaries on hold.

In summary, we are able to conduct elections because there is a clear goal, an established path to perform the tasks which can be easily codified into rules (nature of task - logistics), special focus is created by putting other works of functionaries on hold (focus), visible consequence for non-performance of duty, and it is for a short period of time. On the other hand, in education, there is no ‘the’ established procedure, as the nature of work needs adapting to dynamic circumstances which can’t be codified into rigid rules (nature of work – implementation intensive service delivery) which creates complexities in cases of weak capacity where rule based monitoring is preferable since employees can misuse autonomy, this creates complexities. In education, unlike in elections, special focus cannot be created by putting other tasks of functionaries on hold as it is a long term process, and there are no visible serious consequences for non-performance. Thus, organising elections and education service delivery require capacities of completely different nature. Ability to do one of these necessarily doesn’t mean that we can do the other.

Table 3.7: Why can we organise elections but not deliver education?

Organising elections
Education delivery
Nature of task (frontline functionaries)
Enforcement. Does not involve engaging with public. Does much involve adapting to contexts.
Involves iterative adaptation - adapting to the context and needs engagement with public.
Amenable to rules
Amenable to rules. Clear guidelines can be framed on ‘what to do – how to go from X to Y’.
Not completely amenable to rules. There is no ‘one’ path to go from X to Y.
Interaction of norms of eco-system and nature of task
Rule based monitoring for tasks which are amenable to rules
Generally we tend to do rule based monitoring for tasks which are not amenable to rules.
Addressing weak capacity
Contexts of weak capacity can be handled by strict rule based monitoring.
Weak capacity cannot be totally addressed by rule based monitoring as many tasks are not amenable to rules.
Each functionary generally performs tasks of only one nature, which are mostly amenable to rules. This creates focus.
Functionaries may perform multiple tasks of diverse natures. Even in case where they perform tasks of only a particular nature, they are not amenable to rules. In the absence of focus, people tend to prevent worst from happening rather than excelling and also tasks with rule based monitoring get priority.
Consequences for non-performance
Visible consequences of non-performance.
No strong visible consequences of non-performance.
Time frame of task
Short term task. It is like running a 100 meters sprint. You can use all your energies for this short time.
Long term task and have to be done on a continuous basis. These are marathons. You can occasionally use energy bursts to boost up performance but it cannot be done continuously.

In general, some of the major failures in implementation are regarding the tasks which have the following characteristics on above metrics.
Table 3.8
Task and eco-system
Nature of task (frontline functionaries)
Involves iterative adaption.
Amenable to rules
Not amenable to rules.
Interaction of norms of eco-system and nature of task
Involves customising norms to tasks of different nature.
Weak capacity.
Involves multiple tasks of diverse natures.
Consequences for non-performance
No major consequences for non-performance.
Time frame of task
Long term tasks which have to be done on a daily basis.

Understanding components of state capacity – What determines the implementation of a public policy?

Let us try to put together the lessons learnt in the process of solving the elections vs. education paradox, and prepare a framework to understand state capacity.

Figure 3.10

The first component of state capacity is the nature of bureaucrats – if given a work, can we be assured that they put maximum efforts to work on it and also do not commit any irregularities. We called it bureaucratic capacity earlier and can also call it as work ethic. There are two aspects to it – (i) inherent nature of a person and; (ii) part of a person's nature that is shaped by the system (s)he is present in. A person may start out being completely honest but over time, the constraints of the system may turn that person dishonest. Thus, the proportion of inherent nature and the nature affected by the system in which bureaucrats work in, decide the state of bureaucratic capacity. Fukuyama argues that bureaucratic capacity can be increased by bringing professionalism into public service, especially through appointment of educated professionals, as education consists of socialisation to certain professional norms that seek to preclude certain types of self-seeking behavior.[3] Professionalisation operates through influencing the inherent nature of bureaucrats but it does not explain manifestations of weak capacity observed in case of  of teachers and doctors who are educated. The second aspect of nature of a person, the part that is shaped by the system may, be the reason.

The second component of state capacity is the skill of employees to perform the tasks they are given. Employees may not be performing some tasks despite having the skill to do it or they may be not performing some other tasks because they do not have adequate skills to perform those. Doctors may belong to the former category while the issue with teachers is a mixture of both.

We observe that first and second components are related to the individual characteristics of a functionary. The third, fourth and fifth components to be discussed are external aspects that interact with functionaries shaping their performance.

The third component is autonomy to take decisions and actively being part of feedback loop. We discussed that frontline functionaries need autonomy to implement tasks which need adapting to contexts. Lack of this autonomy and implementation through centrally stipulated rules can lead to inefficiencies and rough edges. For instance, the Gujjar community example in Himachal Pradesh needed special action from functionaries. Without that it is not possible to devise customised solutions to such contexts.

The front line functionary need not always be the decision making authority to be able to adapt to policies. They can also function as ears on the ground sending active feedback on functioning of policies. Such feedback helps in broadly two ways -

(i) it helps improve the policy customising it to conditions. For instance, Glewwe and Karthik Muralidharan summarise four different observed effects of providing learning resources to school[v] “four different studies found that distributing textbooks and resources to schools had no impact on learning – but for four completely different reasons (and so with different implications about what might work and hence what should be done differently next time). In one case the books were just locked up in a storage cupboard and never made it to the kids, in one the textbooks were just too hard, so that only the few brightest kids could even read them, in another case they were effective the first year, but the second year parents stopped spending their own money on books, so the overall impact dropped back to zero in the second year, and in another there was only a positive impact where teachers were also being paid based on their performance.”[4] An active functionary on ground would report these realities and similarly an active local authority would devise ways to address this depending on the context.

(ii) it helps realise the necessities of other pre requisites to realise policy goals. A deficiency need not be always in the design of a particular policy but it can also be the realisation of necessity of other pre requisites to be able achieve the goal of initial policy. The evolution of Bangladesh Rehabilitation Assistance Committee, a Bangladesh NGO, as narrated by The Economist illustrates this. “BRAC is a sort of chaebol (South Korean conglomerate) for social development. It began with microcredit, but found its poor clients could not sell the milk and eggs produced by the animals they had bought. So BRAC got into food processing. When it found the most destitute were too poor for micro-loans, it set up a programme which gave them animals. Now it runs dairies, a packaging business, a hybrid-seed producer, textile plants and its own shops—as well as schools for dropouts, clinics and sanitation plants.”[vi]

Even if someone starts with a micro credit programme, with the aim of eradicating poverty, it is essential to recognise other necessities, in addition to micro credit, to be able to realise the policy goal of micro credit, eradicating poverty. An active functionary would report these deficiencies. Frontline functionaries would only do this if they believe that they are part of the decision making process and trust that their feedback is going to be acted upon. If not, they can “perceive themselves as cogs in the administrative machine, which inevitably leads them to internalise and interpret the challenges they face in their jobs as something that remains outside of their control.  This legitimises a culture of apathy and lack of responsiveness toward understanding and directly addressing the learning deficit.”[vii]

The fourth component is the nature of task. We discussed typology of different natures of task and implementation of each of these tasks requires different approach and skills.

The fifth component is the norms and ways of functioning inherent to the organisation, especially the monitoring rules. The norms and ways of functioning of organisation affect the functionaries in three ways – (i) they set expectations to the functionaries, as to what will be valued by the organisation and what is not. When there is misalignment between the inherent norms of the organisation which set the expectations and the nature of tasks that the functionaries have to perform, we see ineffective implementation. This explains the difference between Himachal Pradesh and Uttarakhand case study discussed above. In Himachal Pradesh, the inherent nature of the administration allowed functionaries to go outside their way, and try innovative aspects where as in Uttarakhand the rigid rule based approach stifled the implementation; (ii) they create incentive and performance pressures to boost and sustain performance of functionaries.

These five components - nature of bureaucrats, skill to perform tasks, autonomy, nature of tasks, and inherent monitoring norms, form the eco-system at the lower level. 

The sixth component is the implementation design of public policy policies which interacts with eco-system at lower level through its demands. The interaction of these demands and rules with the inherent norms of the lower level bureaucracy and their functions can affect them in two ways -

(a) Interaction between public policy design and functioning (inherent norms) of lower level bureaucracy - Consider a lower level bureaucracy where the inherent norms provide space for innovation and the monitoring procedures aren’t strictly rule based. If a new public policy is designed, the implementation design of which requires the lower bureaucracy to strictly follow a rule based approach, it interferes with the inherent unwritten norms of the lower level bureaucracy, disturbing its equilibrium, negatively in this case. On the other hand, if the design of a public policy gives complete autonomy to the lower level bureaucracy which is accustomed to run by rule based mechanisms, it leads to inefficient implementation too.  Thus, a new policy can also either increase or decrease the autonomy of lower level functionaries, the effect of which again depends on the norms of the organisation.

(b) Interaction between public policy decisions and functions of lower level bureaucracy – A new policy affects functions of lower bureaucracy in two ways – (i) it is often the case that there is only one contact point on the field for a department, an Auxilliary Nurse Mid-wife (ANM) in case of health or a CRCC in case of education. Any new policy is thus an increase in work load on them. It is peculiar that private organisations hire new teams whenever there is a new project in place but often in public bureaucracies, the workload of new schemes is piled upon lower level functionaries; (ii) it can disrupt decision making structures. Consider the insights from PAISA study discussed earlier regarding the overlap of decision making structures which observes that implementation decisions are taken at district level but when state government mandates certain aspects, like construction of toilets and boundary walls, it distorts the implementation procedures and priorities at the local level. Of course, this need not always result in negative results. This mechanism can also be used by central authorities to nudge the local bureaucracy to ensure certain aspects but it has to be done in a way accommodating the local priorities.

A well-designed policy thus needn’t just be technically sound but it should also account for the existing norms and nature of the lower level bureaucracy (who implements it), taking into account the effects of interaction of this new policy with the existing framework at lower level. Some of these can be iterated for improvement, the way tech-products are built but one must note that iteration of policies can sometimes have significant costs associated with it, as each change in a policy has potential to affect large population. Hence, one needs to take care anticipating and accounting for all possible aspects while designing rules and regulations of implementation design. This requires specialists with strong domain knowledge and understanding of the nuances of implementation.

                         Cartoon depicting limitations of adaptive approach

  Source: Internet

These six components shape the internal dynamics of the organisation and hence its capacity.

The seventh component is the environment in which policies are being executed. It is possible the same result of the interaction of the six components may yield different outcomes in different contexts. As Joel Midgal argues, the strength of organisation of society resisting state penetration can be a reason for different outcomes of a state with same capacity yielding different outcomes across contexts.[viii] For instance, consider two police departments which have same capacity as defined by the interaction of above six components. If one is assigned a task to implement a policing policy (say, mandating helmets) in a society with little regard to rules and other is assigned to implement the same policy but in a different context where people usually obey laws; the outcomes may differ, though they have same capacity as defined by the six components. In such cases, organisations may have to adjust the six components to be suit the new contexts, which is also a defining feature of high capacity organisations.

These seven factors together decide the effectiveness of implementation of a policy. One may note that the above framework of understanding state capacity does not consider it as a static quantity even when we restrict the definition to specific contexts. As per this framework, state capacity is a dynamic quantity, continuously changing not just with change in new rules (inherent norms) of organisation but also with every new public policy or its iteration.

[1] Fukuyama defines governance as a government's ability to make and enforce rules, and to deliver services, which is equivalent to our definition of state capacity.
[2] Pritchett, L. (2012) ‘The Folk and the Formula: Fact and Fiction in Development.’ WIDER Annual Lecture 16, Helsinki, UNU-WIDER.
[3] Fukuyama, Francis. 2013. “What is governance?” Center for Global Development. Working Paper 314, p:14
[4]Text of summary of Glewwe et al(2015) paper taken from Crawfurd, Lee. “Trillion dollar question: Six ways of spending money better on Learning”. RISE. 26 October, 2015. Accessed at

[i] Aiyar, Yamini, Vincy Davis, Ambrish Dongre. Oct 15, 2015. “Education reform and frontline administrators: A case study from Bihar I”, Ideas For India.
[ii] Mangla, Akshay. "Bureaucratic Norms and State Capacity in India: Implementing Primary Education in the Himalayan Region." Harvard Business School Working Paper, No. 14-099, (Revised August 2015).
[iii] Fukuyama, Francis, 2013. “What is Governance?” Governance: An International Journal of Policy, Administration, and Institutions 26(3):347-68.
[iv] Pritchett, L. (2012) ‘The Folk and the Formula: Fact and Fiction in Development.’ WIDER Annual Lecture 16, Helsinki, UNU-WIDER.
[v] Glewwe, Paul, Karthik Muralidharan. 2015. “Improving School Education Outcomes in Developing Countries: Evidence, Knowledge Gaps, and Policy Implications”. RISE. Working paper-15/001.
[vi] “The path through the fields”. The Economist. November 3, 2012. Accessed at
[vii] Aiyar, Yamini, Vincy Davis, Ambrish Dongre. “Education reform and frontline administrators: A case study of Bihar –II”. Ideas For India. October 16, 2015.
[viii] Joel Migdal, Strong Societies and Weak States: State-Society Relations and State Capabilities in the Third World (Princeton, NJ: Princeton University Press, 1988).

No comments:

Post a Comment