April 13, 2021

Bureaucratic Indecision and Risk Aversion in India




The Indian bureaucracy suffers from indecision and risk aversion, resulting in an inordinate focus on routine tasks, coordination failure, process overload, poor perception, motivational issues and a deterioration in the quality of service delivery. We argue that bureaucratic indecision, in a large part, is a form of rational self-preservation exercised by bureaucrats from the various legal and extra-legal risks to their person, careers and reputation. These risks originate from problems of organizational design, institutional norms and other political factors. The research for this working paper included a review of interdisciplinary literature on bureaucracy and policy decisions, combined with semi-structured interviews. We interviewed current and ex-bureaucrats from India and other Asian and African countries, political scientists and other policy researchers. We also conducted a document analysis of historical and contemporary, administrative and legal documents including committee reports, acts and rules, annual reports and other government publications. We summarise the evidence on factors such as penal transfers, overload, inadequate training, process accountability, contradictory rules and political patronage. The paper concludes with a compilation of administrative and normative reform recommendations taking cues from history, state experiences and other country bureaucracies.

Key Words: Bureaucracy, Decision-making, Risk-aversion, Administrative, Reforms, Norms, Rule-based, Accountability.


Download the full-length pdf version of this paper here. A version of this paper was also published in the Indian Public Policy Review.





Table of Contents

Section 1: Introduction and Motivation

    1.1. Personal Risks and Bureaucratic Indecision

    1.2. Implications on Decision-making and Delivery

    1.3. Scope and Methodology of the Study

    1.4. Introduction to the Indian Civil Service

Section 2: The Causes of Bureaucratic Risk Aversion

    2.1. Organisational Design

        2.1.1. Intimidation through Overmonitoring

        2.1.2. Lack of Autonomy and Discretion

        2.1.3. Bureaucratic Overload and Inability to Specialise

        2.1.4. Candidate Selection and Recruitment

        2.1.5. Inadequate Training and Mentorship

        2.1.6. Career Incentives and Performance Pay

    2.2. Institutional Norms and Culture

         2.2.1. Complex rules and Administrative Inefficiency

         2.2.2. Accountability to Process rather than Outcomes

         2.2.3. Coordination Failures

        2.2.4. Corruption and Reputational Risks

    2.3. Political Alignment and Interference

        2.3.1. Political Support

        2.3.2. Political Pressure

Section 3: Recommendations for Enabling Decision Making

    3.1. Administrative Reforms

        3.1.1. Lessons from a State Government

        3.1.2. Lessons from Other Countries

        3.1.3. Lessons from Historical Reform Recommendations

    3.2. Normative Reforms




List of Abbreviations









The central tasks of a modern state can be placed in three broad categories — maintaining the rule of law, providing public goods to citizens and using fiscal tools to redistribute income. The specific combination of tasks in any one country depends on its history, social norms, constitutional mandates and the political economy of the day. A failure to deliver on these fronts pushes countries towards the anarchy of Matsya Nyay (Law of Fish).


An efficient bureaucracy is essential for a successful state. The German political economist Max Weber identified three sources of political legitimacy — the charisma of a leader, the authority of traditional social norms and modern legal rules. Weber wrote that a bureaucracy is central to the third type of political power, which manifests itself in the constitutional state. In his classic political treatise, Kautilya wrote about the importance of administration — or the role of the amatya (councillor) in his times — as a constituent element of the state.


The efficacy of a bureaucracy is dependent on the incentives or disincentives that individual civil servants face when they take decisions. Much depends on the decision environment. This paper looks at some of the formal and informal constraints that make Indian bureaucrats risk-averse. One of the results of such risk aversion is policy paralysis, a hot button issue in India a few years ago. 






Section 1: Introduction and Motivation



1.1. Personal Risks and Bureaucratic Indecision



Policies are frequently subject to failure. There may be financial obstacles, implementation delays or other unintended consequences. Often, these factors are not necessarily within the control of the bureaucrats who take the implementation decisions. Let us refer to these decisions as policy risks i.e. the risk of failing to deliver the intended public good or service. Such policy failures also have consequences for bureaucrats’ reputation, career progression, financial standing and motivation. These decisions therefore also carry personal risk for the bureaucrat i.e. the risk of being transferred, investigated, accused of corruption or reputationally maligned. In the Indian context, under its various capacity constraints, electoral and policy considerations are frequently at odds with one another, and nearly every policy risk often presents a significant personal risk for bureaucrats. This occurs because they are often individually, rather than institutionally, held accountable for bad policy outcomes. In this situation, bureaucrats exhibit indecision as they avoid personal and policy risk.




“Being a bureaucraat in the Indian Government aspiring to reach the top, the best strategy is not to take risks, not to take decisions, be circumspect in voicing opinions and seek safe jobs in established ministries. Over the last few years, decision making in the Indian Government has become crippled, thanks to the overhang of the 5Cs (CBI, CVC, CAG, CIC and Courts)”. 

— Nandan Nilekani (Swarup 2019).



Moreover, in recent times, it has been observed that even policy risk-free decisions present a significant personal risk to bureaucrats. This could be a routine decision that has near zero probability of a negative policy outcome. This occurs when bureaucrats are pulled up for decisions that are unpopular for extraneous reasons. These could be outcomes that are against political preferences or challenge other vested interests. In order to avoid experiencing these personal losses, bureaucrats either avoid making policy decisions or take decisions that minimise personal loss. This loss aversion strategy manifests itself when bureaucrats perceive losses more strongly than gains (See Box 1). This is often the case with policy decisions related to matters like procurement, infrastructural development and financial regulation.


A third scenario is one where there is no information about the probability of potential outcomes and the bureaucrats are uncertain about the policy decision. Here too, the fear of personal loss may cause bureaucrats to strategically avoid decision making and exhibit a bias toward the status quo. This prevents them from learning about potential policy outcomes for future decisions.


Much of this indecision, therefore, is a form of rational self-preservation exercised by bureaucrats who are otherwise left unprotected by the institutions that they serve. While the literature on governmental corruption, capacity constraints and failure to reform is relatively rich, the area of indecision is less explored. In this paper, we closely examine the nature of risk aversion in bureaucratic decision making in India, why it occurs, what are its implications on service delivery and how it can be ameliorated.



Box 1: Prospect Theory


Imagine a bureaucrat who is designing a new contract for employees in a public sector factory. The introduction of the contract, the bureaucrat’s economic advisors say, could result in two equally likely scenarios: in the first scenario, workers respond to incentives in the contract and production increases by 20%; in the second, the workers protest, causing unrest and production decreases by 2%. Since both events occur with the same probability, it would seem logical for the bureaucrat to propose the contract to the workers. However, most bureaucrats may choose not to do so, preferring to be risk-averse in the face of even mild downsides to their actions.


What does this tell us about the bureaucrat’s decision-making process?


In their seminal work in 1979, Kahneman and Tversky formalised “prospect theory”, which could provide some clues. A “prospect” comprises two elements: outcomes and probabilities. In our example above, there are two outcomes: first, a “good” outcome (G) where production increases by 20% and which occurs with a probability of 0.5; second, a “bad” outcome (B) where production drops by 2% which also occurs with a probability of 0.5. 


Prospect theory defines the value of the prospect as follows: p(G)*v(G) + p(B)*v(B). Here v(G) and v(B) are the values the bureaucrat places on the outcomes, and p(G) and p(B) are probabilities of the outcomes G and B, respectively.  In classical expected utility theory, v(G) would simply be 0.2 (or 20%) and v(B) would be -0.02 (or -2%). Hence, the value of the prospect would be: 0.5*0.2- 0.5*0.02 = 0.09. Since this is positive, the bureaucrat would be expected to undertake the prospect or, as in this case, propose the contract. 


However, under prospect theory, individuals display loss aversion. Loss aversion is a specific case of the more general principle of risk-aversion, where individuals prefer outcomes with less uncertainty to those with more. Under loss aversion, the aggravation one experiences with a loss is greater than the pleasure experienced by a gain. Thus, v(G) could be much lower than 0.2 and v(B) could be far less than -0.02. It could be that the overall value of the prospect 0.5*v(G) + 0.5*v(B) is negative. In these cases, the bureaucrat wouldn’t propose the contract at all. More generally, the “value function” of a prospect is an S-shaped curve as shown in the figure below. Formally, this implies that the curve is concave in gains and convex in losses. In other words, a bureaucrat would benefit more from avoiding a loss of x% in production than from increasing the production by the very same x%. 




The figure above plots a “value function” i.e a curve that plots the value to the bureaucrat (y axis) of a gain or loss in production of the factory (x axis) due to the introduction of the contract. For the same loss or gain x (shown in red in the figure), the value of the gains (y_g) is smaller than the value of the losses (y_l).   


One can now begin to see how this could cause policy paralysis: as the curve becomes flatter in gains and steeper in losses, the bureaucrat becomes more sensitive to the tiniest of losses and refuses to undertake any new actions altogether. The workers are never shown a new contract; the production never changes.  As we describe in this section, bureaucrats are also sensitive to the prospect of personal loss over and above loss to the project. 


Kahneman and Tversky showed, using a series of simple experiments, that loss aversion typified human behaviour more than the cold and calculating expected utility theory. Given the uncertain nature of the environments the bureaucrats work in and the ethos of loss avoidance that permeates the system, it is easy to see how they could behave in exactly the manner prescribed by prospect theory. 


This framework allows us to think through some broad policy principles: for instance, when assessing bureaucrats on ex post outcomes, it is important to consider their ex ante information set and ex ante decisions. In our example above, if the factory ends up in the scenario where production is reduced, the incentive structures for bureaucrats should take into account that this was a low probability event and it was sheer bad luck that things panned out in this manner. A related issue is to develop better, evidence-backed ways of predicting scenarios and observing outcomes. Both of these could help improve assessment of bureaucratic decision-making.  









1.2. Implications on Decision-making and Delivery



 The COVID-19 pandemic has brought to the forefront the importance of agility and quick decision-making in governance. Despite the considerable degree of uncertainty and risk1 around the novel Coronavirus, governments had to make rapid policy decisions in order to avert public health crises and ensure uninterrupted delivery of public goods and services. In many contexts, because time was of the essence, indecision had harmful consequences. For example, several national and regional governments were criticised for delaying airport screening, testkit procurement and, later, the resumption of essential services. Indecision on each of these fronts lead to substantial social and economic losses. Therefore, indecision, even in times of limited information, can have severe implications.


First, risk aversion has an impact on the nature of tasks that bureaucrats choose to take up. Bureaucratic fear of loss of control can lead to a singular focus on routine tasks (Diamond, 1985), particularly in generating documentation viz. paper trails justifying their decisions (Hull, 2012 in Heyman, 2012). Bureaucrats select out of certain kinds of tasks particularly those involving reputational/personal risk. There is a tendency to either deny the existence of the problem, procrastinate over it, pass the file over to another authority or dismiss the problem as altogether unsolvable. 


Second, bureaucrats tend to spread out policy risk by needlessly constituting committees for a decision, passing the decision on to a higher authority or requesting consultation with multiple departments. The proliferation of ministries over the last decades could also be indicative of this. There are 51 Union Ministries as of 2020, having grown three times from the 17 ministries of newly independent India. Even though this may be a positive indication of decentralisation of power, this proliferation hampers decision making by introducing delays and other coordination failures. 


Third, faced by external threats, bureaucrats can either become acquiescent to the system or are forced to find alternative ways to manipulate or bypass the system. One strategy is to collaborate with other bureaucrats to evade procedures (diversify personal risk)  and another is to rely on non-governmental actors like donors and multilateral agencies to support decisions (transfer project risk). When none of these options are viable, bureaucrats find themselves having to resign from the service. In a 2010 survey of civil servants in India, it was found that 33.6% of IAS respondents had considered resigning from the service at least once (DARPG, 2010). Of those who did, 80% mentioned political interference as reason and 73% mentioned frustration (Figure 1). 

(Source: Indian Civil Service Survey (DARPG, 2010))


Figure 1: Reasons for resignation among IAS officers (Base = 900)



Fourth, these motivational issues lead to arbitrariness and poor work performance (Gupta, 2012 in Heyman, 2012) which result in project delays, under-utilisation of resources (Choudhry and Mohanty, 2019), inter-stakeholder coordination failures (Ayyangar, 1949), and even shelving of important reforms. Projects that have very specific mandates and targeted actions are more likely to be implemented than those that have broader ambitions and uncertainty in outcomes. An alternative phrasing of this is that bureaucrats work better in mission mode when projects are “temporally concentrated” in that they have fixed timelines (Kapur, 2020), and their outcomes are predictable. In contrast, bureaucratic response to crises or disaster situations, where timelines and outcomes are unclear or uncertain, such as the ongoing pandemic, is often inadequate as bureaucrats are unable to deviate from rules and take necessary risks  (Singh, 2020).


Fifth, and most gravely, the quality of public goods and services delivered deteriorates. As mentioned earlier, procurement and payment sanctioning are highly risk-ridden processes for bureaucrats. Risk aversion in these processes directly impacts the quality of public goods and services. For instance, there is a  tendency to accept the lowest tenders meeting the minimum specifications (regardless of a bidder’s ability to deliver). This is because price-based procurement decisions are easier to justify later, while quality-based procurement decisions are less straightforward to defend post-transaction. The selection of vendors or subsequent payment is a discretionary process because bureaucrats have to make decisions with incomplete information or based on approvals of officers at other departments/levels, yet they are held accountable for outcomes and therefore more exposed to the risk of corruption accusations. As a result, bureaucrats preemptively rely on price-based processes that then compromise on the quality of inputs that in turn reflect the quality of outputs like roads or irrigation. Even after the procurement process, bureaucrats who are fearful of accusations may choose to continually delay payments in the absence of more evidence of quality. On similar lines, bureaucrats are in the habit of opting for a legal dispute with contractors in case of an unforeseen situation during contract delivery (instead of renegotiating terms), buying from the public sector wherever possible, strict adherence to assessing quality in a quantifiable way, constituting committees with a diverse and large membership, and mechanically applying precedent. As Krishnan and Somanathan (2017) put it, it further embeds the culture of “procedure purity over substantive correctness”.


A final latent but important point is how the perception of bureaucrats (consider the connotation of the word bureaucracy) has suffered as a result of this non-performance.



“People often perceive the bureaucracy as an agent of exploitation rather than a provider of service. Corruption has become a low risk and high reward activity. Frequent and arbitrary transfers combined with limited tenures, are harming the work ethic and lowering the morale of honest officers. While expecting discipline and diligence from the administration, the political executive should self-critically review its own performance. Unless we do this, we cannot regain credibility in the eyes of the people who have elected us to serve them” 

— Prime Minister Atal Bihari Vajapayee, India (1999) at National Development Council Meeting 
(Government of India, 2001)




This is also evident in the common representation of bureaucrats in popular media and culture, associating them with inefficiency, red tape and corruption which in turn reinforces the reputational risks. For instance, a quick analysis of national media finds that “corruption” is the fifth most associated term with “bureaucracy” in the last two decades2






1.3. Scope and Methodology of the Study



The risks that bureaucrats face and the ensuing risk-aversion varies across levels of government, verticals of government and also within services. Anecdotally, we learnt that bureaucrats at higher levels of decision making (secretarial level), bureaucrats in ministries with heavy procurement responsibility (infrastructural ministries) and bureaucrats within services with regulatory responsibilities (for e.g. corporate governance) are more likely to be victims to risk aversion than others. In this paper, we choose to focus on the Indian Administrative Service (IAS) because it is a cadre that staffs officers across these functions. 


As we can see in Table 1 below, IAS occupies a significant proportion of secretarial positions, especially at the Joint Secretarial level. 



(Source: Krishnan and Somanathan, 2017)


Table 1: Distribution of IAS officers across levels of service in 2012


IAS officers are also exposed to a considerable number of legal threats. As we have discussed in Section 2.1.1., as many as 625 complaints were filed against IAS officers in 2018 (See Table 3). Often, such complaints and charges are not disposed of within the same year, which means that officers are likely to carry these charges over multiple years, regardless of their culpability. As a result, these charges are likely to have adverse reputational outcomes even if not legal ones. 


The research for this working paper included a review of interdisciplinary literature on bureaucracy and policy decisions, combined with semi-structured interviews. We interviewed current and ex-bureaucrats from India and other Asian and African countries, political scientists and other policy researchers. Through the interviews, we were able to understand the various types of bureaucratic risks, behaviour under risk and outcomes of this behaviour. 


We also conducted a document analysis of several historical and contemporary administrative and legal documents, including committee reports, acts and rules, annual reports and other government publications. This documentation helped validate the credibility of the legal and administrative threats that bureaucrats face. Further, we examined personal/reported experiences of bureaucrats' experiences such as autobiographical books, news reports and other media, to understand individual experiences of succumbing to or overcoming risk aversion.






1.4. Introduction to the Indian Civil Service



“Many of our civil servants would rank among the best in the world. They have done excellent work. They have delivered results in the face of adversity … Even as we celebrate our successes, we should also be honest in admitting our failures and our deficiencies … On their part, the civil servants in our country should fight the tendency of not taking decisions because of the fear that things might go wrong and they might be penalised for that. We cannot have a bureaucracy, which is 100 percent risk averse. In fact, we should encourage boldness in decision-making, provided that the decisions are well considered and according to the law of the land.” 

— Prime Minister Manmohan Singh, India (2012) Address on Civil Service day
(Business Standard, 2012)



In 1854, the Macaulay Committee recommended a merit-based, as opposed to patronage-based, recruitment process for the Indian Civil Services (ICS). Accepting the recommendation, from 1855, the ICS inducted recruits, from Oxford and Cambridge University, through an examination that was held in London. However, the service remained largely for Englishmen—applicants from Indian universities formed less than 4% of the service—who were tested for, among other things, their culture and ability in the Greek and Latin languages (Vanusia 2005). Then, from 1922, the exam was also held in Allahabad and Delhi on the recommendation of the Montagu Chelmsford Reforms, increasing the eligibility and intake of Indian applicants.


The present-day All India Services, including the IAS, have inherited several features from this pre-independent ICS including independent selection, entry rules, training, tenure, grading of positions and, importantly, the system of promotion and transfers based predominantly on seniority (ARC, 2008). 


Post-independence, the Union Public Service Commission (UPSC) was constituted and held responsible for the management of civil servants in all matters including recruitment, selection, promotion, deputation, induction, framing of rules and amendments, disciplinary action and advising government on matters referred by the President of India (see Articles 315 and 320 of the Constitution of India). 


A second important institution in the context of the civil service is the Ministry of Personnel, Public Grievances and Pensions, which includes the Department of Personnel and Training (DOPT), Department of Administrative Reforms and Public Grievances (DARPG), and Department of Pensions and Pensioners’ Welfare (DPPW). The Lal Bahadur Shastri National Academy of Administration (LBSNAA), under the DOPT, is one of the premier training institutions responsible for capacity building initiatives for IAS officers.


The UPSC examinations are highly competitive and applications for each position run into the hundreds. Table 2 below summarises the most recent statistics of applications and recruitment to the IAS. 




(Source: DOPT and UPSC Annual Reports (Government of India 2019))

Table 2: Applications to UPSC Examinations and IAS positions


This is primarily because the service is a highly aspirational occupation, associated with social service, prestige and job security (see Figure 2). 



(Source: Indian Civil Services Survey (DARPG, 2010))

Figure 2: Reasons for joining the Indian Administrative Service (Base = 3519)


The responsibilities of IAS officers change with their role during their service. Post induction and training, new entrants are given ‘field postings’. This includes a series of sub-divisional (e.g. Sub-Divisional Magistrate), district-level (e.g. District Collector), and then departmental positions (e.g. Chairperson of Board of Revenue). The primary responsibilities in these roles include law enforcement, policy implementation, revenue management and general administration. After field postings, which last about 8-12 years, an IAS officer is posted at a secretariat level. These positions involve policy development, political consultations and leading sectoral-specific governmental duties (Government of India, n.d.). 


Back to TOC 








Section 2: The Causes of Bureaucratic Risk Aversion


Despite the considerable decision-making power at senior levels of the service described above, there are several factors that hinder and constrain decision making. In the following sections, we delve into the organisational design, institutional norms and political interference and other factors highlighted above that prevent decision making and induce risk aversion among bureaucrats.


(Source: Indian Civil Services Survey (DARPG, 2010))

Figure 3:  Selected grievances of Indian civil servant (NB: Base differs for each grievance)

Figure 3 above represents selected grievances of civil servants from the 2010 Civil Service Survey (DARPG, 2010) (selected based on the relevance to the subject of this working paper), which describes a number of issues that we will be covering in the following sections.






2.1. Organisational Design



“Each individual possesses a conscience which to a greater or lesser degree serves to restrain the unimpeded flow of impulses destructive to others. But when he merges his person into an organizational structure, a new creature replaces autonomous man, unhindered by the limitations of individual morality, freed of humane inhibition, mindful only of the sanctions of authority.”

— Stanley Milgram (1974)



Policy discourse has analysed the state as an institution, but discussion of the state as an organisation has been limited (Kapur, 2020). In this section, we aim to do this by examining the processes and procedures governing recruitment, monitoring, training and performance appraisal, and access to resources. We discuss how organisational design failures lead to indecision and risk aversion. 






2.1.1. Intimidation through Overmonitoring


In a democratic framework, bureaucrats are unelected officials, but those that wield regulatory power; thus, it is essential to have mechanisms for oversight of their effectiveness and efficiency (Gruber, 1987) (Apaza, 2008). However, sometimes these monitoring mechanisms become tools for exercising control over the bureaucracy, which ultimately undermines their ability to do their job. Misuse of this monitoring system can also lead to adverse consequences for bureaucratic performance by becoming tools of intimidation. 


Baseless Investigations


60% of Indian IAS officers feel that their performance was affected by baseless complaints and investigations (DARPG, 2010). Table 3 below summarises the types and number of complaints pending or filed against IAS and other senior officers in a given year. 


(Source:  DOPT Annual Report (Government of India 2019))

Table 3: Complaints and Investigations against IAS officers in 2018


In many of these instances, the cases or complaints are eventually dismissed but the process itself is the punishment as several of these proceedings extend over years, having significant reputational consequences for the accused bureaucrat. 


Monitoring compliance with rules through (neutral) mechanisms such as audits and right to information is an integral part of governance and a key feature in the checks and balances on executive functioning and public service delivery. Despite that, research has shown that such features can become restrictive and conversely, engender opacity (Gerardino et al. 2017). For example, with the introduction of the Right to Information Act, 2005 (RTI Act), an unintended effect has been that the deliberative component of the bureaucratic process has come under scrutiny. While the aim of the RTI Act was and is to foster transparency in governance, its absorption into the bureaucratic work culture has followed a different path. As a result, bureaucrats now stay away from tendering frank advice (Krishnan and Somanathan, 2017).


Penal Transfers


Other causes of poor performance are disruptive transfers and arbitrary demotions through the ability to upgrade and downgrade posts (Krishnan and Somanathan, 2017). The authors contend that arbitrary transfers have led to short tenures in each post which has in turn diminished effectiveness. Every incumbent is still on a learning curve in nearly every post when they are being transferred; this prevents specialisation (discussed in detail in Section 2.1.5). Iyer and Mani (2012) present evidence of the use of transfers as political instruments and find that the average probability of transfer for an officer in a given year is 53% (data from 1980 - 2004).


Frequently transferred officials may not have adequate formal appraisals which affects their ability to get empanelled for promotions to seniors postings. Looking at the extent to which civil servants are transferred after an election resulting in a change in government, India does worse than all parliamentary systems other than Italy, even at the central level (Krishnan and Somanathan 2017). The 2nd Administrative Reforms Committee (ARC) recommended abolishing the system of penal transfers as it is demoralising and acts as a drain on the exchequer.


The threat of transfers and investigations make bureaucrats avoid decisions or take decisions in a risk-averse manner. This is perhaps why practices associated with monitoring have stronger negative associations on bureaucratic performance with more experienced bureaucrats (Rasul, et al., 2017) who are able to identify over-monitoring as a cause for delay and failure. Interestingly, in certain contexts, heightening monitoring and enforcement is also counterproductive for corruption as it displaces corruption into adjacent decision areas that are outside the purview of the monitoring system (Yang 2005) (Lichand, Fernandes 2019). 






2.1.2. Lack of Autonomy and Discretion


Apart from legal over-monitoring, bureaucrats are often also prone to administrative over-monitoring, in that their decision making is often subject to approval and authorisation by their superiors. This lack of autonomy not only prevents risk-taking but also disincentivises risk-free decisions. Giving bureaucrats discretion over their own decisions has multiple implications for decision making. Autonomy increases the speed and efficiency of project completion. For example, evidence from West Africa shows that a one standard deviation increase in bureaucratic autonomy increases project completion rates by 18% in the Nigerian civil service (Rogger and Rasul, 2016) and 28% in the Ghanain bureaucratic context (Rasul et al., 2018). In Ghana, the authors also investigated corruption practices and found little evidence of autonomy having a corrupting effect. 


Ineffective Supervision


One mechanism for this increased efficiency is the removal of undue influence or delays by senior monitors. In a study of procurement officers in Punjab, Pakistan, it was found that shifting decision-making power (on audit documentation) from a senior monitor to the implementing procurement officer improves procurement decisions, especially when the monitor is ineffective (Bandiera et al. 2020)3.  


Discretion, Choice and Quality 


A second mechanism is the increased discretion exercised by bureaucrats to make better choices such as while choosing contractors/suppliers. Studying bureaucratic discretion in procurement decisions in Italy, Coviello et al. (2018) found that even though discretionary procurement decisions led to repeat purchases from past firms (indicative of favouritism), the decisions themselves were associated with lower work duration and administrative costs, and improved quality and quantity of firms. “Discretion may then help, rather than harm, as it allows incomplete contracts to be complemented with dynamic informal governance mechanisms” (Coviello et al., 2018). However, this trade-off between efficiency and leakage is sensitive to the conditions of discretion and auction design. In contrast, in the Hungarian context, discretionary procurement decisions were strategically exercised by bureaucrats who were politically connected and therefore procurement outcomes suffered and winning firms were less productive and more expensive (Szucs, 2020). Therefore considerable thought needs to go into the design of autonomy and task assignment and the incentive compatibility of this design. 


Room for Experimentation


A third mechanism, most relevant to the subject of this paper, is that autonomy provides bureaucrats with the legal and administrative room to experiment, fail, learn and specialise in certain tasks without the fear of personal threats. While this section discusses the independence of bureaucrats, the independence of the bureaucracy as an institution is a closely related topic, which has been discussed in more detail in Section 2.3. 






2.1.3. Bureaucratic Overload and Inability to Specialise


Another salient reason for decision paralysis in the Indian bureaucracy is administrative overload and the inability to specialise. The Figure 4 below indicates the level of understaffing in some Indian states. 






(Source: (Krishnan and Somanathan, 2017))

Figure 4: State IAS Representation against Population


Note : Rate of understaffing is calculated by taking the percentage difference between the state’s share of the IAS cadre and the state’s share of national population


Sub-optimal Time Use


Dasgupta and Kapur (2017) led a study where block development officers (BDOs) were asked to record their time use across the following tasks: filing forms, managing office and staff, planning and budgeting regarding schemes, field visits, handling individual complaints and requests from citizens, local politicians or legislators, unrelated duties, or other. The authors find that these officials are required to multi-task excessively, losing focus on managerial activities, impacting project implementation and related outcomes. The authors find that BDOs are able to allocate, on average, only 22% of their time on planning and budgeting for programs.


Apart from managerial activities, another set of functions that have a bearing on public service delivery are coordination activities. Kalaj et al. (2020) study time-use among bureaucrats in Ethiopia and find that civil servants that spend more time in coordination activities (meetings and interfacing with clients) work in organisations with significantly better service delivery outcomes. In contrast, time spent working alone is worse for service delivery even if there are better incentives and more discretion for the bureaucrats. 


Lack of Continuity and Specialisation


Closely related to the problem of suboptimal time use is the problem of specialisation and competence. One of the adverse outcomes of frequent transfers is the inability for bureaucrats to specialise (Khosla and Vaishnav, 2016). However, the relationship between specialisation and service delivery, especially by IAS officers, is not straightforward. On one hand, the allure of the IAS is precisely that it is a service for generalists, giving officers access to multiple levels and departments of government. “Make the IAS officer a specialist and he is like anyone else” (Ghate, 1998). The officers are expected to have a certain number of sector-agnostic qualities like leadership, team management, coordination, problem-solving or as (Agarwal and Somanathan, 2005) put it, “intelligent and informed generalists”.


On the other hand, the reduction of the service into a series of “stints” implies a lack of continuity for the bureaucrat and thus lack of accountability for service delivery. On-the-job learning opportunities also shrink and career advancement suffers. Ferguson and Hason (2013) show that specialisation aids career advancement for bureaucrats in India at all career stages. For early-stage bureaucrats, specialisation serves as a positive signal of ability, while for senior bureaucrats, it aids advancement through the actual ability and skills gained. Specialisation and the development of a focus are important from the perspective of decision making, because expertise provides bureaucrats with the confidence to take risks and manage its consequences.


In Section 2.1.5, we discuss how formal training and capacity building can facilitate specialisation. 






2.1.4. Candidate Selection and Recruitment


Reforming Selection Criteria


Another factor worth examining in the context of competence for decision making is the selection and recruitment process of IAS officers. Does the current recruitment process elicit the application of, and subsequently select candidates with entrepreneurial qualities? At present most discussion about reform to recruitment processes is focussed on quantitative criteria like number of attempts, age limits, selection criteria and examination design. This is because successful candidates are getting older, are less likely to hold a postgraduate degree, and take an average of four attempts to pass the entrance exam (Vaishnav and Khosla, 2016).


Addressing these challenges, the 10th report of the Second Administrative Reforms Commission (2nd ARC) recommended reduction in maximum permissible age for appearing in the civil service examination (to 25 for general candidates) and the maximum permissible attempts (to three for general candidates) to address the increasing average age of entrants (See Figure 5)  and the number of prior attempts.


(Source:  ARC, 2008)


Figure 5: Maximum Eligible Age and Average General Age of IAS Candidates 


Personality, Talent and Motivation


While these indicators are reflective of the skill and talent of candidates, the discussion on recruitment reforms often leaves out selection on the basis of qualitative criteria. Candidate personality is particularly important in the realm of improving decision making and delivery. Are Indian bureaucrats risk averse by selection? 


Behavioural research, such as by Hanna and Wang (2017), provides some evidence on how current advertising and positioning of government jobs fail to attract candidates with desirable qualities. They find a correlation between the propensity for dishonesty and preference for government service in India. Interestingly, even after screening for ability, the pool of candidates does not improve on measures of dishonesty. 


What motivates candidates to apply to the civil service is also an indicative consideration. Are they motivated by career progression (career oriented) or by social impact (prosocial)? Studying health workers in Zambia, Ashraf et al. (2020) found that when advertising for civil service recruitment, respondents to career-oriented advertisements are likely to be more talented and more effective at each step of service delivery as opposed to prosocial advertisements5. Therefore, a combination of attracting career-oriented applicants and screening by ability leads to the selection of the most objectively worthy candidates. This finding has been in contrast with previous research which made the case for selecting candidates precisely for their prosocial motivation (health inspectors in Pakistan (Callen et al., 2015) and graduates in Indonesia (Banuri and Keefer, 2013)), indicating that this is a context and role-specific dynamic, worth investigating further in the Indian administrative context.






2.1.5. Inadequate Training and Mentorship


IAS officers initially complete a year of training at the LBSNAA which includes a foundational course and Phase 1 training. This is followed by a year of district-level on-field training which is then followed by Phase 2 training. From 2007 onwards, a comprehensive system of mid-career training was introduced whereby officers need to complete 8 weeks of training between their 7th and 9th year and 14th and 16th year and 4 weeks between their 26th and 28th year of service. This training is necessary for promotion to the next level. There is also a plethora of optional training including options for foreign training. We know from prior research that higher marks on training is a predictor of on-the-job performance (Bertrand et. al, 2015). Data has shown that there are clear performance benefits for IAS officers who invest in training (Khosla and Vaishav, 2016). But how can bureaucrats be trained to become better decision makers? 


Lack of leadership training


A repeated issue with bureaucratic training programmes is that the training focuses on lower order, typically task-based assignments and doesn’t take into account the need for training that gives officers domain expertise or skill development (ARC, 2008).  It also does not focus on attitudinal and behavioural skills.


Second, there is inadequate training on developing leadership skills and team building, which are essential to the functioning of a complex bureaucracy. Low appeal amongst the senior officers and short-term versus long-term tradeoffs of such training result in a lot of top officers not undertaking such training (ARC, 2008). In this respect, the Indian bureaucracy can learn a lot from the example of Singapore and Malaysia. In Singapore, bureaucrats receive consistent leadership training and development. They are taught to understand challenges and to break down complex problems into smaller parts. Additionally, there is an entitlement of 100 hours of training available to the bureaucrats a year. In Malaysia, yearly training is a component for performance evaluation and career progression. Other such learnings from international contexts are described in detail in Section 3.1.  


Third, bureaucrats are often trained to be policy implementers rather than policy makers. They are experts at execution, even under constrained situations, but not necessarily at big picture ideation and experimentation. This may be another factor leading to risk aversion when making decisions.

Developing Competency


One way of resolving the trade-off between generalist and specialist expectations  is by giving bureaucrats the opportunity to train in the subject of their current posting on a real-time basis. To this end, the recently launched Mission Karmayogi (PIB, 2020) is a step forward in the right direction because it aims at “matching competencies to post requirements” rather than relying on bureaucrats’ prior education. Further, it aims to make capacity building a continuous rather than episodic process. 

Access to Mentorship


As mentioned earlier, a significant grievance (highlighted in DARPG, 2010) has been the lack of mentorship by senior civil servants for junior officers. This is particularly the case at sub-state levels. Mentorship is potentially a channel that can plug gaps in conventional training and through which bureaucrats can learn about qualitative aspects of the bureaucracy, particularly leadership, team management, problem solving and risk taking.


Interestingly, some observers of the civil service believe that it is precisely this lack of contact that isolates district and junior officers from the risk aversion that manifests at higher levels, as a result of which they are able to make decisions, delegate and have trust in the decision-making apparatus. This is another way of saying that current mentorship propagates the culture of risk-aversion, because of senior bureaucrats’ lived experience in dealing with investigations and charges. 






2.1.6. Career Incentives and Performance Pay


As discussed earlier, while intrinsic motivation in the public sector may be higher than in the private sector (Andersen and Seren 2012; Cowley, 2013; Bellé, 2012), the implication of this for bureaucratic decision making and effectiveness is not straightforward, especially in the absence of career incentives. 


Rigid Progression Rules


Promotion in the IAS is tenure based, occurring at the intervals of 4, 9, 13, 16, 25 and 30 years of service. While at the junior levels, officers usually get promoted soon after they hit these tenure durations; at the senior levels, there is a discrepancy between the sanctioned and actual time as promotions are often dependent on vacancies and performance reviews (Bertrand et al. 2019). After 16 years of service, officers are also eligible to be posted to the central government, through a process called empanelment, which is considered highly prestigious. However, empanelment isn’t a transparent process and often lies in the hands of politicians (Krishnan and Somanathan, 2017).


Bertrand et al. (2015) have shown that these fixed progression rules are particularly demotivational to officers who enter the service at an older age. The higher average age of entrants into the IAS has significantly reduced the eligibility of officers to secure higher-level positions as shown in Figure 6 (ARC, 2008). A little over 10% of officers who entered the service in 2005 are eligible to be promoted to the position of chief secretary. 


(Source: Government of India 2008)


Figure 6: Percentage of IAS officers eligible for chief-secretary


Lack of Career Planning


In addition to career progression rules, the annual performance review process also has many flaws (See Box 2). Officers tend to be judged on adherence to process instead of outcomes, and as a result, do not have a stake in producing good results. Further, the focus of the performance review tends to be on numerical grades and evaluation for past performance instead of a holistic approach that includes setting goals and career planning (ARC, 2008). It also occurs at a fixed time every year instead of being an ongoing process. The performance review process takes place between an officer and his superior, and there are no departmental goals or alignment between officers in the same department, let alone inter-departmental cooperation.



Box 2: APAR Process


The Performance Appraisal of members of the bureaucracy is conducted as per the All India Services (Performance Appraisal Report) Rules, 2007. There are three levels of authorities with cascading degrees of seniority involved in this process—the reporting, reviewing and accepting authorities. The reporting authority supervises the performance of a service member. The reviewing authority supervises the reporting authority and the accepting authority supervises the reviewing authority. These officers vary by position—for instance at a Joint Secretary level, a Secretary is the reporting authority, a relevant minister of state is the reviewing authority and a Cabinet Minister is the accepting authority. The performance review consists of 5 sections—one for basic information, one for self-appraisal by the candidate and sections for appraisal, review and acceptance by the reporting officer, reviewing officer and the accepting officer, respectively.


The self-appraisal includes a work plan the reporting authority signs off on; an officer’s accomplishments are assessed based on this. Officers also need to reflect on their performance and indicate areas where they made significant contributions and those where an upgrade in skills is necessary, and the proposed steps/training programmes needed to achieve that. The reporting authority is supposed to write the appraisal report every financial year. This consists of the assignment of numerical grades on the assessment of attributes like attitude to work and decision-making capabilities (70% weightage) and assessment of work output like the accomplishment of planned work and quality of output (30% weightage). Numerical grades are given from 1-10 to each officer, relative to the performance of the officer’s peers. Grades 1-2 and 9-10 are given only in rare cases and need to be justified by specific accomplishments or lack thereof. The reporting officer also comments on the officer’s integrity. The reviewing authority reviews the reporting officer’s work and also assigns grades in the appraisal report and the accepting authority accepts the report. In cases of differences of opinions with other officers, the reviewing and accepting officers must note the reasons. 


The officer has 15 days to reply based on factual information in the report, and similarly, the accepting authority reviews the comments and forwards it down the chain to get the reviewing and reporting authorities’ comments. If the officer disagrees with the report, he can present his case to the Referral Board, whose decision is final. For IAS officers this board usually consists of: (i) Chief Secretary of the State Chairperson, (ii) Senior-most Secretary in the State Member, and (iii) Secretary (Appointments) Convener for state-level officers; and (i) Cabinet Secretary Chairperson, (ii) Secretary (Personnel) Member, and (iii) Establishment Officer Convener for Central officers.



(Source: Government of India 2007)


Because of its ineffectiveness in affecting personal or career growth, officers engage with APAR minimally. Reviews are often filled post the assessment period when necessary to merely provide a stamp of approval for promotion. Goals are sometimes written retrospectively, i.e. once tasks are completed, and bureaucrats often have the power to influence data that they are assessed on. Further, there is a perception among officers that promotions are secured unfairly. In the 2010 DARPG survey, 78% of respondents believed that some or most officers used influence to gain sought-after positions. Only 24% believed that these postings were merit-based (Khosla and Vaishnav, 2016).

Pay Compression


Finally, there is also significant pay compression in government services. Lower-level bureaucrats are overpaid and top-tier ones, including those of the IAS, are significantly underpaid relative to what they could make in the private sector. As shown in Figure 7, the ratio of the maximum wage to the minimum wage has fallen from 54.5 post-independence to 12.50 in 2015. Higher wages attract more able applicants (Dal Bó et al., 2013) and the lack of financial incentives causes self-selection in qualified applicants who would rather work in the private sector. Another issue with low salaries is that it could encourage officers to make money in other ways while in office (Khosla and Vaishnav, 2016).



(Source: ARC 2008)


Figure 7: Pay compression - Ratio of max. Salary to min. Salary over time


In conclusion, in the absence of appropriate career and monetary incentives to motivate bureaucrats, bureaucrats are likely to take a conservative, self-preservational approach to decision making and avoid risks. 


Back to TOC






2.2. Institutional Norms and Culture



“A management style [that] is relatively authoritarian, a high degree of control, top-down communication, individuals search for stability, limited initiatives, and centralised decision making”

 — (Hendryadi, 2019)



The Indian bureaucracy has a 150 year old colonial legacy, and the institutional norms and culture that have evolved from this legacy play a significant role in current risk aversion. This is because bureaucratic norms tend to guide how state capacity develops (Mangla 2015) and what is documented as “official” can have a significant impact on culture (Dery 1996). Routine behaviour leads to the desire for maintaining the status quo, reinforcing isolationist tendencies and defensive attitudes (Diamond 1985).


This thread of institutional legacy is apparent across countries once governed by the United Kingdom. However, some former colonial countries, most notably Singapore, have steered the government machinery towards reform. The government has been able to instil a culture promoting performance to enable officers to make decisions. As Best et al. (2019) put it, governance outcomes and “state effectiveness [are] to a large extent embedded in the individuals and organisations of the bureaucratic apparatus. [Those] tasked with implementing policy are important sources of variation in states’ productivity”. 







2.2.1. Complex rules and Administrative Inefficiency


Pay Compression


The pre-colonial Indian Civil Service (ICS) was a heavily centralised institution, designed for a small group of officers to control a vast country of over 300 million people (Merchant, 2012). This applied not only to the organisational structure but also to decision making. Officers made decisions based on adherence to rules rather than outcomes. This legacy of centralised decision making and rule-based administration from a distance had a deleterious effect on the risk-taking ability of officers. Rule-based administration contracts, if not eliminates, all discretionary powers of the officer. As a result, decision making suffers and many proposals are lost or rejected without adequate consideration.


Contradictory Rules


In an ideal world all these laws, rules, and guidelines would be well defined, in congruence with one another, would not contradict each other and would have clearly outlined incentives/disincentives. In the context of the Indian bureaucracy, this is often not the case. Bureaucrats face difficulty in taking prompt and appropriate action as there are laws that contradict each other, and in some cases, there are laws formed in such a manner that the incentives or deterrents are either too low or too high. This produces administrative inefficiency in the form of delays, coordination failures or simple avoidance. Consider the following excerpt from the RTI Act: 



An Act to provide for setting out the practical regime of right to information for citizens to secure access to information under control of public authority in order to promote transparency and accountability in the working of any public authority, ...

— Right To Information Act, 2005





Given that the RTI Act was enacted to provide citizens with access to information in the control of any public authority, consider the following features of certain other Acts that contradict the spirit of the RTI Act and hinder its implementation:


  • The Official Secrets Act, 1923 does not define ‘secret’ and makes it an offence for public officials to communicate any information to an unauthorised person. Since what is secret is not defined, this becomes a catch-all phrase to restrict the flow of information.

  • Sections 123 and 124 of the Indian Evidence Act, 1872 restrict a government servant from citing evidence from unpublished documents without the permission of the head of the department. They are not allowed to disclose any information that may cause harm to the public interest.

  • Paragraphs 116 and 118 of the Manual of Office Procedures prohibit officials from communicating information in any form to anybody without a general or specific order.

  • There are also state-specific laws that restrict public servants from revealing official documents or parts of it to any person.


These features of multiple laws on the same subject create confusion in implementation, ambiguity and needless litigation regardless of implied repeal (ARC, 2008). 








2.2.2. Accountability to Process rather than Outcomes



Building on the extensive usage of written procedures of the Mughal empire, the British administration, which was deeply distrustful of the native Indian functionaries, created a “surveillance and control” regime using papers, files and procedures (Hull, 2003). The paper-trail based method of administration remains entrenched in the organisational norms of the Indian bureaucracy. While there has been a shift from being a colonial state with the primary functions of taxation and public order to a welfare state, the bureaucracy has not been wholly recast.


 Accounting based Accountability


The clamour to keep the bureaucrats accountable to citizens with a view to ‘discipline’ the unruly has further hampered the organisation’s culture. The task of fixing accountability has led to top-down, rule-based, process-driven accounting systems, leading to “accounting based accountability” (Honig and Pritchett, 2019).  Invariably, this has had a far-reaching impact on discretion, with a tendency to put the process front and centre, ahead of outcomes. It has proved to be detrimental to bureaucratic decision making, becoming a roadblock to the ‘compliant’ rather than penalising the ‘non-compliant’. Bureaucrats across levels of responsibility describe themselves as “cogs” in the machine responding to orders rather than citizen needs, breeding a culture of “order followers” rather than “active agents of change” (Aiyar et al., 2015; Aiyar et al., 2020). 


In discussing the limits to accounting-based accountability in education (and far beyond), Honig and Pritchett (2019) highlight its pernicious effect on welfare. They caution against managing civil servants primarily or exclusively via “accounting-based accountability” as it could “demotivate those in the system and spur departure of the differentially positively motivated”. The authors go on to say that “accounting-based accountability can be a kind of self-fulfilling prophecy, creating the problem it is meant to solve”. 


Apart from breeding deep-seated distrust in the system, it fosters “legalistic” norms vis-à-vis a more conducive “deliberative” approach to decision making within the bureaucracy. “Legalism” promotes rule-following and deference to a hierarchy, often at the cost of public service delivery while a “deliberative” model of bureaucracy encourages outcome orientation (Mangala, 2015). Since norms are unwritten codes of conduct, any push to reform the bureaucracy and the state needs to alter the norms and organisational culture from within (Aiyar, 2018)


In Box 3 below, we summarise all the central rules and regulations and some state laws that govern the IAS, revealing the extent of micromanagement of the service and loss of discretion.





Box 3: List of rules/regulations that govern IAS (non-exhaustive)




Central Acts, Rules and Regulations


1.      Indian Evidence Act, Section 123 and 124, 1872

2.    Official Secrets Act, 1923

3.    Delhi Special Police Establishment Act, 1946

4.    Constitution of India, 1950

5.    All India Service Act, 1951

6.    All India Services (Compensatory Allowance) Rules, 1954

7.    All India Services (Medical Attendance) Rules, 1954

8.    All India Services (Travelling Allowance) Rules, 1954

9.    IAS (Recruitment) Rules, 1954

10.   Indian Administrative Service (Cadre) Rules, 1954

11.    Indian Administrative Service (Pay) Rules, 1954

12.   Indian Administrative Service (Probation) Rules, 1954

13.   Indian Administrative Service (Recruitment) Rules, 1954

14.   All India Services (Leave) Rules, 1955

15.   All India Services (Provident Fund) Rules, 1955

16.   Indian Administrative Service (Recruitment by Competitive Examination) Regulation, 1955

17.   Indian Administrative Service (Appointment by Promotion) Regulations, 1955

18.   Indian Administrative Service (Probationer Final Exam.) Regulations, 1955

19.   All India Services (Special Disability Leave) Regulations, 1957

20.  All India Services (Death-Cum-Retirement Benefits) Rules, 1958

21.   All India Services (Remittances Into and Payments from PF and family pension Funds) Rules, 1958

22.  All India Services (Commutation of Pension) Regulations, 1959

23.  All India Services (Conditions of Service-Residuary Matters) Rules, 1960

24.  All India Services (Study Leave) Regulations, 1960

25.  Indian Administrative Service (Fixation of Cadre Strength) Regulations, 1966

26.  All India Services (Conduct) Rules, 1968

27.   All India Services (Discipline and Appeal) Rules, 1969

28.  All India Services (Confidential Roll) Rules, 1970

29.  All India Services (Dearness Allowance) Rules, 1972

30.  All India Services (Joint Cadre) Rules, 1972

31.   Departmental Inquiries (Enforcement of Attendance of witnesses and Production of Documents) Act, 1972

32.  Former Secretary of State Service Officers (Conditions of Service) Act, 1972

33.  All India Services (Leave Travel Concession) Rules, 1975

34.  All India Services Indemnity Act, 1975

35.  All India Services (House Rent Allowance) Rules, 1977

36.  All India Services (House Building Advanced) Rules, 1978

37.   All India Services (Group Insurance) Rules, 1981

38.  Administrative Tribunals Act, 1985

39.  Indian Administrative Service (Regulation of  Seniority Rules, 1987

40.  Prevention of Corruption Act, 1988

41.   Compilation of Fundamental Rules and Supplementary Rules (Part-I), 1989

42.  Revised Scheme for Redeployment of Surplus Staff, 1989

43.  Indian Administrative Service (Appointment by Selection) Regulations, 1997

44.  All India Services (Prevention of Sexual Harassment) Regulations, 1998

45.  IAS Promotion GuideLines, 2000

46.  Compendium of Welfare Programmes Instruction, 2001

47.   Central Vigilance Commission Act, 2003

48.  Central Vigilance Commission Amendment Act, 2004

49.  General Financial Rules, 2005

50.  Right to Information Act, 2005

51.   AIS (Performance Appraisal Report) Rules, 2007

52.  All India Services (Performance Appraisal Report) Rules, 2007

53.  Central Vigilance Commission (Staff) Rules, 2007

54.  Draft Public Services Bill, 2007

55.   IAS (PAY) Rules, 2007

56.  Administrative Instructions on Departmental Canteens in Government Officers, 2008

57.   IAS (Pay) 2nd Amendment Rules, 2008

58.  Civil Services Performance Standards and Accountability Bill, 2010

59.  Public Interest Disclosure and Protection to Persons Making Disclosure Bill, 2010

60.  Citizen's Charter and Grievance Redressal Bill 2011, 2011

61.   Right to Public Service Act, 2011

62.  Right To Information Rules, 2012

63.  Lokpal and Lokayuktas Act, 2013

64.  Delhi Special Police Establishment (Amendment) Act, 2014

65.  WhistleBlowers Protection Act, 2014

66.  Manual of Office Procedure, 2015

67.   Central Vigilance Commission (Staff) Amendment Rules, 2017

68.  Lokpal (Complaint) Rules, 2020

69.  Extracts of provisions in F.R. 56, n.d.

70.  Fundamental Rules (FR 11 and FR 52, 53 and 54), n.d.

71.   Public Servants Inquiries Act, n.d.




State level Acts, Rules and Regulations



Bihar Reorganisation Act, 2000

2. Madhya Pradesh Reorganisation Act, 2000

3. Uttar Pradesh Reorganisation Act, 2000

4. Maharashtra Government Servants Regulations of Transfers and Prevention of Delay in Discharge of Official Duties Act, 2005

5. Madhya Pradesh Lok Sewaon Ke Pradan Ki Guarantee Adhiniyam, 2010

6. Bihar Lok sewaon ka adhikar Adhiniyam, 2011

7. Chhattisgarh Lok Seva Guarantee Bill, 2011

8. Delhi (Right of Citizen to Time Bound Delivery of Services) Act, 2011

9. Himachal Pradesh Public Services Guarantee Act, 2011

10. Jammu and Kashmir Public Services Guarantee Act, 2011

11. Janhit Guarantee Act, 2011

12. Karnataka (Right Of Citizens to Time Bound Delivery Of Services) Bill, 2011

13. Rajasthan Public Service Guarantee Act, 2011

14. Right to Service Act, 2011

15. Uttarakhand Right to Service Act, 2011

16. Assam Right to Public Services Act, 2012

17. Kerala State Right to Service Act, 2012

18. Odisha Right to Public Services Act, 2012

19. Goa (Right to Time-Bound Delivery of Public Services) Act, 2013

20. Gujarat (Right of Citizens to Public Services) Bill, 2013

21. West Bengal Right to Public Services Bill, 2013

22. Haryana Right to Service Act, 2014

23. Maharashtra Right to Public Services Ordinance, 2015





Fear of Unknown Unknowns


The other concern that impacts decision making is the constant worry about risk and uncertainties. The former is defined as one which can be “predicted, measured and quantified” whereas the latter relates to “unknown unknowns” and “known unknowables” where outcomes cannot be meaningfully defined (Sanyal, 2020). Tackling the latter requires building systemic trust, best done through active supervision rather than stringent and stifling rules and regulations. Rules and regulations intended to cover each and every deviation creates a system of “opacity” and “complexity” that burden the majority with compliance. Further, the burdensome regulation can be throttling, with “endless box-ticking and excessive requirements”. For the system to perform, it would be far better to have a simpler regulatory framework supplemented by active supervision (Sanyal, 2020).  


The hesitation to take decisions also stems from the fuzziness of locus of accountability. Often, senior bureaucrats have to make large payment decisions based on unverified information provided by those on the ground. The payment system needs to transition from one relying on mere relay of data and submission of ex-post utilisation certificates to one where personnel directly responsible for overseeing the product certify the delivery as per the contract. This needs to be supplemented with active monitoring by immediate seniors to foster a system of responsible supervision. 



Corruption Perceptions


In 2019, India ranked 80 out of 180 countries in the Corruption Perception Index published by Transparency International. Moreover, 69% respondents in 2017 said that they paid a bribe when accessing public service in the last 12 months, as part of their Global Corruption Barometer report. Corruption is symptomatic of poor management as witnessed by bureaucrats using public institutions for “personal enrichment and the provision of benefits to the corrupt” which can work towards “undermin[ing] the legitimacy and effectiveness of government” (Rose-Ackerman et al., 2016). 


Experts posit several reasons for this increase in corrupt practices in the higher levels of bureaucracy. Krishnan and Somanathan (2017) suggest that concerted effort at pay compression between the highest and lowest ranks and rising private sector salaries led to a decrease in morale and an increase in corruption. Other mechanisms of bureaucratic corruption include conferment of special rewards beyond the normal conditions of service through extensions of service and post-retirement employment opportunities in regulatory or quasi-government bodies. The creation or filling up of such posts are timed with the retirement of senior officers, which allows extraction of pensions as well as the drawing of new salaries (Krishnan and Somanathan, 2017). 


Anti-corruption Policy


Alongside these developments came the introduction of anti-corruption legislation, including the Prevention of Corruption Act, 1988, the Central Vigilance Commission Act, 2003, the Lokpal and Lokayukta Act, 2013, and the WhistleBlowers Protection Act, 2014. However, rather than reducing corruption, they have had the effect of displacing corruption by bringing in certain adaptive behaviours in bureaucratic decision making. They have reduced the willingness of officers to offer clear policy advice on potentially controversial issues, who take the safer route of holding up matters, templatising decision making to minimise the risk of prosecution and forming committees (Krishnan and Somanathan, 2017). 


For example, the Prevention of Corruption Act (PCA) 1988, Section 13(1)(d)(iii), states that “a public servant is said to commit the offence of criminal misconduct if he, while holding office as a public servant, obtains for any person any valuable thing or pecuniary advantage without any public interest.” The absence of ‘intent’, ‘negligence’, ‘recklessness’ or ‘prior knowledge’ in establishing guilt under this section has created an environment which suppresses decision making. As Leaver (2009) puts it, the avoidance of criticism prompts even public-spirited bureaucrats to behave inefficiently. 


In a welcome move, the passage of Prevention of Corruption (Amendment) Act, 2018 redefined provisions related to public servants. It made “bribery” a direct offence and introduced intentional enrichmentin Section 13, whereas in the earlier version, intent was not included in the definition of criminal misconduct by public servants. There is also now a provision of prior permission for investigating serving and retired bureaucrats.


Paper Trail Paralysis


The fact that a decision could unintentionally, unknowingly or tangentially cause private gains to some party, even if there is no loss to the government, has created an environment of decision paralysis.  Central to bureaucrats’ self-preservation strategy is the ability to manage files, papers and procedures. The paper trail becomes overwhelmingly important to preserving an untarnished reputation. 


There is undue reliance on investigating the paper trail instead of following the money trail for prosecuting ‘criminal conspiracy’ by a public servant. An unintended consequence of this line of action is that innocent and honest bureaucrats could be prosecuted, tarnishing their reputation and destroying careers, even as the corrupt continue to conceal the act of bribery. The practical effect of this tendency to “investigate by file reading” has led bureaucrats to feel that it is beneficial to “initially oppose on file any request from the private sector, or indeed a private citizen, even if genuine” as a way to manage the risk of prosecution (Krishnan and Somanathan, 2017). 







2.2.3. Coordination Failures


Lack of Team-orientation




A hierarchical and pyramidal structure with multiple levels, both salient features of a Weberian bureaucracy, characterise the Indian state well. Several countries have attempted to ‘flatten the structure’ as well as infuse team orientation and organisational goals in an effort to break silos and respond to complex needs of governance (2nd ARC, 2008). 


The role of intrinsic motivation, team composition, distribution of bureaucracy across different functional lines and tiers of local, state and federal government, and coordination, both horizontally and vertically, play a critical role in the organisational design and effectiveness of the bureaucracy (Kapur, 2020). The culture and practices within bureaucracy need to be reoriented towards collaboration to achieve the common organisational target. Otherwise, bureaucrats will spend more time in coordination work rather than on improving service delivery and related outcomes (Kalaj et al., 2020) 


Inter-governmental Coordination


Given the complexity of public administration, the ability to coordinate different functions is key to achieving goals. Ethnographic studies suggest that “fragmentation and arbitrariness in bureaucratic work” create incoherence within the government and lead to poor service delivery despite genuine commitment (Heyman, 2012). The fragmented structure translates into coordination failure within and between departments. Therefore, an important piece of successful policy and decision making lies in organisational design, enabled by integration and coordination between departments and units within them. This needs to be vertical as well as horizontal in nature. A network between civil servants which enables interactions and problem solving and a common understanding of the problems and desired outcomes is key to better coordination (Peter, 2018). 






2.2.4. Corruption and Reputational Risks




Theoretical work has long established that the structure of government institutions is a very important determinant of the level of corruption (Shleifer, Vishny 1993). The incentive structure for bureaucrats needs to be carefully designed in order to prevent malfeasance, especially when corruption detection capability is limited (Becker and Stigler 1974).



Back to TOC







2.3. Political Alignment and Interference

So far, our discussion of bureaucratic decision making has been removed from the political economy in which it operates. While colonial legacy can explain inter-governmental differences, it cannot explain within-country or inter-organisational differences. One explanation for this could be the bureaucrat-politician dynamics, which vary across states and ministries. In the following sections, we discuss how these interactions can both facilitate and hinder risk-taking.








2.3.1. Political Support


A common view of the politician-bureaucrat relationship sees the politician determining policies and the bureaucrat implementing them. This view, while simplistic and stylised, conveys a broad sense of the roles these actors play. Typically, it is the case that large policy reform requires political will. Furthermore, it is also true that politicians do not usually involve themselves in the nitty-gritty of policy implementation. However, the nature of the politician-bureaucrat relation is complex. There is now a new set of studies in political economy that helps us understand how this plays out. 


There is evidence to suggest that bureaucrats work more efficiently when they work with politicians who (i) are aligned with the party in power, (ii) command a large vote-base, and (iii) are the sole representative liaising with the bureaucrat.  


Electoral Support


For example, BDOs have more resources at their disposal when they work in constituencies whose politicians are aligned with the party in power at the centre (Dasgupta and Kapur, 2017). Moreover, Nath (2018) argues that bureaucrats respond better to politicians who are elected with large vote margins. The paper matches politicians to bureaucrats from across India. As an outcome, the paper focuses on project time taken by the bureaucrat to approve projects under the Member of Parliament Local Area Development scheme. The main finding is that bureaucrats working under politicians who have larger vote margins—i.e when the constituency is a politician’s stronghold—take 11% less time to approve projects. 


Clear Reporting Channel


Bureaucrats work more efficiently when they are directly under the control of a single politician (Gulzar and Pasquale, 2016). In many instances, a single bureaucrat (say, the District Collector) is answerable to multiple politicians (say, Members of the Legislative Assembly). There are at least three reasons why this could be problematic: first, politicians may not be able to claim full credit for the bureaucrat’s performance and this could weaken their incentives to monitor performance; second, some politicians may choose to free-ride on other politicians responsible for the bureaucrat; third, politicians may jostle for control, which could have negative effects on development outcomes. Gulzar and Pasquale (2016) argue that when it comes to implementation of the National Rural Employment Guarantee Act, 2005, the credit-claiming story may be the most likely cause for single politicians’ superiority over multiple politicians.   







2.3.2. Political Pressure


While bureaucratic decision making benefits from political will or support, it suffers greatly under political interference. Lack of independence and political interference makes bureaucrats subservient to the political class, to whom they pass on their decision making to avoid reputational risks. 


Political Punishments


The incidence of this interference with bureaucrats’ careers is evident in the pattern of bureaucratic transfers which are known to rise when new politicians are appointed (Mani and Iyer, 2012). The authors exemplify this with the state of Tamil Nadu, where the spike in transfers corresponds clearly to the change in chief minister. Khemani (2007) has also shown that discretionary transfers executed by the then Planning Commission were often distributed to favour states that were politically important for the central power.


The literature on this subject points to two factors that further exacerbate this dynamic. First, it appears that bureaucrats at senior levels are more likely to experience political interference and loss of decision-making power, as can be seen in Figure 8 below where Apex Scale is one of the highest grades in the civil service, typically signifying a position such as a Chief Secretary of State. 




(Source: Indian Civil Servants Survey (DARPG, 2010))



Figure 8: Perception of civil servants on undue external influence (Base = 4517)



Second, it has been argued that bureaucrats that are assigned to their home states are perceived to be more corrupt and less able to withstand political pressure (Xu, Bertrand and Burgess, 2020). In both cases, we can see that organisational and social proximity to political decision makers translates to higher political pressure or influence.



Resultant Rent-seeking



The impact of this interference and exposure to political corruption takes multiple forms. First, “endemic political interference can lead to rent-seeking behaviour even for honest officers” (Khosla and Vaishnav, 2016). This takes place through activities like garnering favours, directing decisions toward self-benefit or simply indecision. As you can see in Figure 9 below, bureaucratic effectiveness closely relates to control over corruption, and countries which exercise high control of corruption are also the same countries that show most bureaucratic effectiveness. India ranks 106/214 on corruption control and 76/214 on bureaucratic effectiveness.






(Source: World Bank, 2018)



Figure 9: Global Mapping of Corruption Control and Bureaucratic Effectiveness


Note: Expressed in units of a standard normal distribution


Interestingly, there is also early evidence that better financial incentives, especially through higher public sector wages, are not associated with lower corruption, indicating that deeper administrative reforms are required (Jakiela 2018), which we discuss in the following section. 






Back to TOC








Section 3: Recommendations for Enabling Decision Making




“We cannot march through the twenty-first century with the administrative systems of the nineteenth century.”




 — Prime Minister Narendra Modi, India (2016) at NITI Aayog Transforming India Lecture in 2016






In this section, we discuss administrative and normative reforms to tackle bureaucratic indecision and risk aversion. We assess learnings from state governments and other countries, as well as historically proposed reforms in the Indian context. 







3.1. Administrative Reforms


There are several areas to focus reforms on in order to create a ‘positive professional identity’ for the bureaucracy, as highlighted by Krishnan and Somanathan (2017) and the two administrative reform commissions. 


A certain degree of bureaucratic independence is essential for quick decision making and preserving the efficiency of governance. E-governance is touted as one such approach for bringing about increased efficiency in the bureaucracy. Firstly, enabling e-governance via digitisation and automation is critical for documentation and recordkeeping. Moreover, it aids in consolidation of instructions, improving searchability, creating an electronic trail for decision making and reducing the time taken for various processes. However, in the absence of behavioural changes, technological changes remain superficial. Training modules need to be targeted, kept up-to-date and should extend beyond mere procedural aspects such as procurement and finances to systemic subjects such as decision-making abilities. 


There is an urgent need to prioritise training on leadership and self development to allow bureaucrats to understand challenges, break complex issues into smaller pieces and address them. For example, re-architecture of the public finance systems towards using workflow, internet-based IT platforms, in-built third party back-checks and automatic audits at the stage of decision making (instead of a post facto assessment) could boost bureaucrats’ confidence in the decision-making process. Quick third party back-checks will allow for immediate verification of data instead of ex-post utilisation certification-based audits which bring opaqueness in the system and disincentives decision making. Payment platforms designed to build automatic limits alongside self-certification requirements for each drawing and disbursing officer will enable financial autonomy as well as foster confidence between different levels of bureaucracy (Mathew et al., 2017). 


When focusing on these areas of reform, several global norms are essential to keep in mind as model examples. A citizen-centric approach should be fostered within the bureaucracy. In order to facilitate inter and intra-departmental or ministerial cooperation, a lead agency must be identified and tasked with securing a unified outcome, including being empowered with a veto. Without adding another layer to the process, a senior minister with the necessary visibility and authority could be entrusted to step in, in case of gridlock. 








3.1.1. Lessons from a State Government


Strengthening internal communication has the potential to significantly improve coordination within the government and lead to superior governance outcomes. Clear communication, including shared knowledge of targets and a two-way channel to ascertain information and receive feedback, is a means of achieving clarity of objectives and required action. The focus of this should be on improving managers’ ability to communicate and lead by example (Nsubuga, 2017). 



For instance, after the 2018 floods in Kerala, the Government of Kerala (GoK) established working groups (State Planning Board, 2019) to enable local governments to learn from the crisis. The state was amongst the first to engage in active multi-sector consultation and coordination during the first wave of COVID-19. 24/7 control rooms were created at the state and district levels on 1 February 2020. Along with that, state medical boards and rapid response teams were constituted to enable coordination between different parts of pandemic response. Figure 10 gives a glimpse of GoK’s coordination bodies and their composition. 





(Source: Sneha et al. 2021)

Figure 10: Various GoK and external COVID-19 Coordinating bodies and their composition







3.1.2. Lessons from other countries


In Singapore, public sector wages are indexed to those of the private sector and hence, move in line with the market (though they are likely to remain below private sector wages in most cases). Performance-based promotion is practised and civil servants experience mobility within the service in that they have the ability to transfer and work for various government organisations over time. These two aspects provide sufficient monetary and career incentives for bureaucrats to perform their job well. Before a new post is created, there is an assessment done to determine the opportunity cost of the job for a typical applicant and the wage level, so a qualified candidate can be hired. In terms of performance review, bureaucrats are reviewed both according to their current job and according to how high they can rise in their careers, reflecting different skill sets valued at different levels. Further, every 12 months there is a performance ranking of bureaucrats. There is also a ‘mid year structured performance review’ in which staff and supervisor exchange feedback on their performance and other workplace issues. There is also a workplace suggestion scheme and a culture of upward feedback that promotes transparency and empowerment. 


In Malaysia, since decision-making authority for Ministries is typically with the respective Minister, civil servants focus on anticipating the issues that may arise and take necessary steps to provide recommendations. Committees comprising civil servants are usually set up to facilitate this process. Records of discussions are shared with all relevant agencies with clear timelines to ensure implementation. There is a structured appraisal and evaluation process for all levels of bureaucracy that occurs twice a year, linked to key performance indicators. Every bureaucrat has to undergo training for a certain minimum number of days per year and can liaise with the ministry for their training needs in line with skills they need. Ministries have a training budget allocated towards in-house or external training. Additionally, before confirmation of promotion, bureaucrats need to undergo training. 


Indonesia was able to simplify and standardise processes in its Ministry of Finance in order to improve employee performance, improve accountability and gain public acclaim in a department renowned for venality and unaccountability. Brazil was able to reduce public corruption through setting up and implementing a code of conduct. Malaysia was able to develop a system that tied performance management to service delivery at the organisational level (Iyer, 2011). 







3.1.3. Lessons from Historical Reform Recommendations


Since independence, India has constituted several civil service reform commissions and committees that have closely examined the Indian bureaucracy, studied other contexts and made recommendations toward reform. These bodies looked into key themes including the role, structure and organisation of the civil services, recruitment and training, performance and transparency, politicisation and corruption, personnel management, service delivery, and public grievances, among others. 






Figure 11: Historically Proposed Reforms 


Multiple 2nd ARC reports have made recommendations about the issues discussed in this paper. In Box 4 below, we summarise recommendations from the 1st Report (Right to Information: Master Key to Good Governance), 4th Report (Ethics in Governance), 10th Report    (Refurbishing of Personnel Administration and Scaling New Heights), 13th Report (Organisational structure of Govt of India) and 14th Report (Strengthening Financial Management Systems) (ARC, multiple) and also the 43rd Law Commission of 1971.




Box 4 : Summary of Recommendations from relevant ARC reports


Recruitment and Selection: The 10th Report suggested the creation of national curricula for undergraduate and graduate-level education in public policy to address gaps in pre-entry education levels. Offer bridge courses to those who do not graduate from these curricula. It also recommended the reduction of the maximum permissible age for appearing in the civil service examination (25 for general candidates) and the maximum permissible attempts (3 for general candidates) to address the increasing average age of entrants and the number of attempts. Further, it suggested modifications to the structure of preliminary and main examinations for better order, pace, selection and exam content. It also recommended that the rules of induction of state officers to IAS be revised to make induction more regular and fair.

Training and Performance: Multiple reports in the 2nd ARC stressed the need to use training and performance management systems to inculcate features of cutting edge level functionaries to work in tandem with the organisational goals. It suggested the need to expose civil servants “to new management skills as well as concepts such as teamwork, public-private partnership, managing networks and issues relating to tackling corruption”. It also recommended integrating organisational goals with the performance of constituent units and ultimately with all individuals working in it. It suggested setting goals at all levels of bureaucracy, monitoring individual and unit performance within the organisation, investing in capacity building, rewarding good performance, and managing underperformance.

Legal Monitoring:

  • In order to safeguard against malicious RTI requests, the ARC recommended that Public Information Officers should exercise the right to provide information if the request is “manifestly  frivolous or vexatious” and communicate the same to the applicant within 15 days of application. However, how such applications will be identified or classified was not clarified.

  • The Prevention of Corruption Act 1947 requires that the court receive approval of a sanctioning authority before it can take cognisance of an offence by a public servant. The committee recommended that this protection be extended to retired civil servants as well.

  • The 43rd Law commission (1971) also recommended that the National Security Act be designed to encompass the multiple conflicting legislations on security discussed in Section 2.2.1, including the Official Secrets Act. However, the NSA, promulgated in 1980, simply replaced the Maintenance of Internal Security Act, as a result of which the contradictions highlighted earlier persist. 

Career Progression: The ARC recommended that the current empanelment-based system for Senior Administrative Grade (SAG) level should be replaced with a more transparent and objective procedure. This requires ministries to classify their SAG positions by functional domain so that the Civil Service Authority, to be established under the Civil Service Bill, can assign officers according to their competencies. Additionally, for positions at SAG and higher levels, the Civil Service Authority will accept and assess applications from all services to increase competition.





Back to TOC 






3.2. Normative Reforms


Reforming Managerial Practices



It is important that private sector managerial practices are not simply imported onto the complexities of public sector organisations (Sinclair, 2008; Krishnan and Somanathan 2017). As Best et al (2019) put it, “the magnitude of the potential benefits of designing policy to match the effectiveness of the implementing agents is especially important … since altering human resource practices can be infeasible or costly in the public sector”. It is important to take underlying culture into consideration and design systems which adapt to it. Platforms for performance management, project management, training and feedback must be customisable to allow each organisation to use them effectively.


Along with financial incentives, it is important to socially incentivise bureaucrats, emphasise the importance of their role and provide social recognition for their services to citizens. For example, incorporating success stories in the training of officers may help build a positive narrative in the bureaucracy and improve their self-image and confidence. Social incentivisation has proven to significantly interact with productivity and motivation (Ashraf and Bandiera, 2018). 


Evidence and Institutional Memory


In the context of bureaucrats working toward policy reforms, they must be encouraged to elicit and collect data on citizen demand and/or document any other evidence motivating the need for such reforms. Providing this evidence helps make an electoral argument for these reforms when political support for it is otherwise minimal. Further, bureaucrats should be encouraged to document their effort toward a reform, especially when met with political opposition in a given time, so that the work can be utilised by future governments.


Another important factor for institutional memory is data on bureaucratic grievances and victimisation. At present there is no systemic record of cause for bureaucratic transfers, legal accusations, convictions, career progression and other complaints. While departments like DOPT and DARPG do publish data on some of these subjects on an ad hoc basis, it is not published consistently, periodically and in easily accessible formats.


Tolerance for Failure


Most importantly, the government must have tolerance for failure. This critical mindset change will help move from a ‘fail safe’ approach, where there is no tolerance for failure, towards a ‘safe fail’ paradigm that allows bureaucrats to fail within reason.

Indian bureaucratic culture also lacks a design thinking approach to problem solving in policy implementation. For example, inculcating the practice of seeing an issue as  a design challenge, prototyping and iterating solutions can help identify bottlenecks in policy frameworks and facilitate implementation. This can be particularly useful in public-facing systems where user feedback and perspective can be incorporated into government processes. Creating this space for experimentation and iteration will encourage bureaucrats to take reasonable risks and navigate uncertainty. 


A political leadership interested in overcoming the challenges of indecision in the bureaucracy must demonstrate its intent to empower bureaucrats. This requires a visible devolution of power such that bureaucrats are actively discouraged from seeking approvals from senior leadership and are given the mandate to take decisions at their level. 



Back to TOC 










“Public policy, to put it flatly, is a continuous process, the formulation of which is inseparable from its execution.” 



— Carl Friedrich




In this paper, we discuss the various reasons for bureaucratic indecision, uncertainty and risk-aversion. This includes flaws in organisational design, restrictive institutional norms and political misalignment. Following this we discuss administrative and normative reforms, historically proposed in the Indian context and from other country contexts, summarised below:



  1. Protection from personal risks: Provide bureaucrats protection from political transfers, false charges, political interference and prosecution on deliberative processes. Eliminating these personal risks will enable the bureaucrats to take policy risks without the fear of consequences to their person, reputation and career. 

  2. Autonomy and Discretion: Enable autonomy and discretion in decision making whenever feasible. Performance management systems need to be adapted so that bureaucrats are accountable for outcomes (within their control) rather than mere adherence to processes.

  3. Reform rules of career progression: Career progression should not be simply a factor of seniority but also performance. Even though performance-based pay does not address all the incentive problems in the Indian bureaucracy, it does disincentivise extra-legal rent-seeking. How performance is measured and graded is a key determinant to the success of this strategy. 

  4. Competency and Training: Train bureaucrats to perceive themselves as policy makers rather than as policy implementers. This requires bureaucrats to have leadership skills, team-building skills and other sector-specific expertise. It is important to regularly and systematically assess competencies and facilitate specialisation.

  5. Clear Mandate and Preventing Overload: Equip bureaucrats with the necessary technological, educational, administrative and human resources they need for decision making to prevent overload. Bureaucrats and their teams must have clearly defined terms of reference and reasonable mandates.

  6. Trust and Motivation: Demonstrate trust in bureaucrats’ abilities and decisions and policy experimentation. Bureaucrats need to be motivated to engage in the complicated and time-consuming process of policy reform. Apart from appropriate financial incentivisation, this also requires social incentivisation.

  7. Citizen demand and Institutional memory: Enable bureaucrats to collect data and document citizen demand for certain reforms. They must be encouraged to document their work toward a reform. The Indian government must invest in studying its own and other bureaucracies by collecting and maintaining periodic data on bureaucratic demographics composition, grievances, outcomes, transfers, legal accusations, convictions and career progression etc. Moreover, this data should be published in easily accessible formats, amenable to regular analysis by journalists, researchers and citizens.


While this paper focuses on IAS officers, much of its learnings are applicable to state, district and block-level administrations as well. Local administrators have even less access to resources and training compared to IAS officers and are, under certain circumstances, less empowered to take autonomous decisions and more exposed to personal risks. This has scope for further study and policy attention. 


Back to TOC 





1. While risks are decisions where potential outcomes and their likelihoods are known, in advance, decisions under uncertainty take place without either information. For example, during the initial weeks of the COVID-19 pandemic, when little was known about viral transmission rates and the media of transmission. At the time, governments had to decide about mobility and access restrictions under uncertainty. However, later stages of the pandemic, when the risk of transmission in different circumstances could be estimated based on case data and other information collected up to that point.


2. Author analysis using www.mediacloud.org.


3. Authors measure this in terms of decrease in procurement prices and find that autonomy granted to officers with ineffective monitors are able to reduce price by 20%.


4. Note that BDOs are civil servants, not IAS officers.


5. In fact although applicants may be less prosocial on average, when put through a recruitment process (that focuses on talent) the ultimately selected candidates are equally likely to be pro-social. 





We are grateful to Niranjan Rajadhyaksha for his guidance at all stages of developing this working paper, from conceptualisation and methodological set-up to analysis and editorial feedback. We are also grateful to Sharan M.R. for his regular inputs and written contributions. We would also like to thank Vikram Sinha, Gyanendra Badgaiyan and Tanvi Ravel Mehta for their comments and suggestions and Sridhar Ganapathy for his assistance with web-design.

This paper benefited greatly from the insights of our interviewees, Gyanendra Badgaiyan (IDFC Institute; former civil servant, Government of India), Yamini Aiyar (Centre for Policy Research), Gulzar Natrajan (civil servant, India; Global Innovation Fund), Momodou K. Dibba (civil servant, Government of The Gambia), Del Francis Wreh (Executive Director, LIMPAC (Policy Think Tank), Government of Liberia), M N Reddi, (IDFC Institute; former civil servant, Government of India), Devesh Kapur (Paul Nitze School of Advanced International Studies, John Hopkins University), Daniel Rogger (World Bank), Wu Wei Neng (Chandler Institute of Governance; former civil servant, Government of Singapore), Aneetha Warusavitarana (Advocata), Ravi Ratnasabapathy (Advocata), Sarinder Kumari (former civil servant, Government of Malaysia), Siddharth George (Boston University; former civil servant, Government of Singapore), Santhosh Mathew (Bill and Melinda Gates Foundation; former civil servant, Government of India), Bhuvana Anand (Co-founder, Trayas) and K. P. Krishnan (former civil servant, Government of India).



Back to TOC




Appendix: List of Abbreviations



AIS                All India Services
APAR            Annual Performance Appraisal Report
ARC               Administrative Reforms Commission
BDO              Block Development Officers
CAG               Comptroller and Auditor General of India
CBI                Central Bureau of Investigation
CCS                Central Civil Service
CIC                Central Information Commission
COVID          CoronaVirus Disease
CVC               Central Vigilance Commission
DARPG         Department of Administrative Reforms and Public Grievances
DOPT            Department of Personnel and Training
GOK              Government of Kerala
IAS                Indian Administrative Service
ICS                Indian Civil Services
LBSNAA       Lal Bahadur Shastri National Academy of Administration
NITI              National Institution for Transforming India
NSA               National Securities Act
OSA               Official Secrets Act
PCA               Prevention of Corruption Act
RTI                Right to Information
SAG               Seniors Administrative Grade
SPSC             State Public Service Commissions
UPSC            Union Public Service Commission



Back to TOC







Ashraf, Nawa and Bandiera, Oriana (2018), “Social incentives in organizations”. Annual Review of Economics.


Administrative Reforms Commission, Government of India (2008). “Second Administrative Reforms Commission: Refurbishing of Personnel Administration: Scaling New Heights” 10th Report. Department of Public Grievances and Redressal. https://darpg.gov.in/sites/default/files/personnel_administration10.pdf 


Administrative Reforms Commission, Government of India (Multiple). “Second Administrative Reforms Commission. All Reports. https://darpg.gov.in/arc-reports 


Agarwal O. P. and Somanathan T.V. (2005) Public Policy Making In India: Issues and Remedies. Working Paper. Centre for Policy Research.


Aiyar, Yamini and Krishnamurthy, Mekhala (2020) “Lateral entry, biometric attendance won’t fix Modi’s bureaucracy issue. Reinvest in the state”, ThePrint, https://theprint.in/opinion/lateral-entry-biometric-attendance-wont-fix-modi-bureaucracy-issue/372836/


Aiyar, Yamini, (2020) “Look beyond rules to reform the bureaucracy”, Hindustan Times, https://www.hindustantimes.com/opinion/look-beyond-rules-to-reform-the-bureaucracy/story-9UNqqyjR8MNNJjEVo0035O.html


Aiyar, Yamini; Dongre, Ambrish and Davis, Vincy (2015), “Education Reforms, Bureaucracy and the Puzzles of Implementation: A case study from Bihar”, International Growth Centre, Working Paper


Amanda Sinclair (2008) Public Sector Culture: Managerialism or Multiculturalism? Australian Journal of Public Administration 48(4):382 - 397. DOI: 10.1111/j.1467-8500.1989.tb02240.x


Andersen, Lotte Bogh, and Seren Serritzlew (2012). Does Public Service Motivation Affect the Behavior of Professionals? International Journal of Public Administration 35(1): 19-29. https://www.tandfonline.com/doi/pdf/10.1080/01900692.2011.635277?needAccess=true.


Apaza, Carmen R. (2008) The Importance of Bureaucratic Oversight Mechanisms: The Case of the Inspector General. The American University. https://www.bpastudies.org/bpastudies/article/view/67/145


Ashraf, Nava, Bandiera, Oriana, Davenport, Edward and Lee, Scott S. (2019). Losing Prosociality in the Quest for Talent? Sorting, Selection, and Productivity in the Delivery of Public Services.https://ashrafnava.files.wordpress.com/2019/10/losing_prosociality_final.pdf


Ayyangar, N. Gopalaswami Ayyangar (1949) “Report on Reorganisation of the Machinery of Government”Government of India Press.


Bandiera Oriana, Best Michael Carlos, Khan Adnan Qadir & Prat Andrea (2020) "The Allocation of Authority in Organizations: A Field Experiment with Bureaucrats,". NBER Working Papers 26733. National Bureau of Economic Research, Inc.


Banuri, Sheheryar and Keefer, Philip (2013) Intrinsic Motivation, Effort and the Call to Public Service World Bank Policy Research Working Paper No. 6729, Available at SSRN: https://ssrn.com/abstract=2370149


Becker, Gary S., and George J. Stigler (1974) "Law Enforcement, Malfeasance, and Compensation of Enforcers." The Journal of Legal Studies 3, no. 1: 1-18. http://www.jstor.org/stable/724119.


Bellé, Nicola (2013) "Experimental Evidence on the Relationship between Public Service Motivation and Job Performance." Public Administration Review 73, no. 1 : 143-53. http://www.jstor.org/stable/23355447.


Bertrand, Marianne, Burgess, Robin, Chawla, Arunish and Xu, Guo. (2019) The Glittering Prizes: Career Incentives and Bureaucrat Performance. https://faculty.chicagobooth.edu/-/media/faculty/marianne-bertrand/research/april-2020-update/glitteringprizes_r3_final.pdf


Bertrand, Marianne, Burgess, Robin, Chawla, Arunish and Xu, Guo. (2015) Determinants and Consequences of Bureaucrat Effectiveness: Evidence from the Indian Administrative Service. http://economics.mit.edu/files/11199/


Bó, Ernesto Dal, Frederico Finan, and Martín A. Rossi (2013) "Strengthening State capabilities: The role of financial incentives in the call to public service." The Quarterly Journal of Economics 128, no. 3 : 1169-218. https://www.jstor.org/stable/26372521.


Business Standard (2012)  “We cannot have a 100% risk-averse bureaucracy: Edited excerpt from Prime Minister Manmohan Singh’s speech on Civil Services Day on April 21 in New Delhi”, https://www.business-standard.com/article/opinion/-we-cannot-have-a-100-risk-averse-bureaucracy-112042200007_1.html


Callen, Michael and Gulzar, Saad and Hasanain, Syed Ali and Khan, Muhammad and Rezaee, Arman (2015) Personalities and Public Sector Performance: Evidence from a Health Experiment in Pakistan . NBER Working Paper No. w21180, Available at SSRN: https://ssrn.com/abstract=2607368


Choudhury, Mita & Mohanty, Ranjan (2019) Utilisation, Fund Flows and Public Financial Management under the National Health Mission. Economic and political weekly. 54. 49-57.


Coviello Decio and Guglielmo Andrea and Spagnolo Giancarlo (2016) The Effect of Discretion on Procurement Performance . CEPR Discussion Paper No. DP11286, Available at SSRN: https://ssrn.com/abstract=2783286


Cowley, Edd and Smith, Sarah. (2013) Motivation and mission in the public sector: Evidence from the World Values Survey. The Centre for Market and Public Organisation. http://www.bristol.ac.uk/media-library/sites/cmpo/migrated/documents/wp299.pdf.


Dasgupta Aditya and Kapur Devesh (2017) The Political Economy of Bureaucratic Overload: Evidence from Rural Development Officials in India . Available at SSRN: https://ssrn.com/abstract=3057602 or http://dx.doi.org/10.2139/ssrn.3057602


Department of Public Grievances and Redressal (2010) “Survey of Civil Servants” Government of India. https://darpg.gov.in/sites/default/files/Civil_Services_Survey_2010.pdf 


Dery, David (1998). "Papereality" and Learning in Bureaucratic Organizations Administration & Society - ADMIN SOC. 29. 677-689. 10.1177/009539979802900608. https://www.researchgate.net/publication/249625243_Papereality_and_Learning_in_Bureaucratic_Organizations


Diamond, Michael A. (1985) “The Social Character of Bureaucracy: Anxiety and Ritualistic Defense.” Political Psychology 6, no. 4 (1985): 663-79.. doi:10.2307/3791022.


Ferguson, John-Paul, and Sharique Hasan. “Specialization and Career Dynamics: Evidence from the Indian Administrative Service.” Administrative Science Quarterly 58, no. 2 (2013): 233-56. http://www.jstor.org/stable/43546649.


Ghate, Prabhu (1998) “Reforming the Civil Service: Meeting Crucial Need for Expertise.” Economic and Political Weekly 33, no. 7 : 359-65. http://www.jstor.org/stable/4406409.


Government of India (2001) “Approach paper to the Tenth Five Year Plan” Planning Commission.


Government of India (2019) “Department of Personnel Affairs Annual Report 2018-19” Ministry of Personnel, Public Grievances and Pensions.


Government of India “Service Profile for IAS” (n.d.) Department of Personnel Training.  https://cseplus.nic.in/Home/DisplayPDF?streamId=PCsUnEplvZihdzEe8FEMf/ozdm2Z1SV9EtWCRovKRC3lakWb5Az6zBHuA5aOBE/F63T7xmhv7oprudau8IxE7qwEbpfY7FdzRY38CgzX0Ow3cxPusYL1aWOaLZjLhzSg 


Griffith, T.L., Nordbäck, E., Sawyer, J.E. et al. (2018) Field study of complements to supervisory leadership in more and less flexible work settings. J Org Design 7, 10https://doi.org/10.1186/s41469-018-0034-5. https://jorgdesign.springeropen.com/articles/10.1186/s41469-018-0034-5


Gruber, Judith (1987) Controlling Bureaucracies: Dilemmas in Democratic Governance. Berkeley: University of California Press, http://ark.cdlib.org/ark:/13030/ft1g500470/ 


Guilherme Lichand and Gustavo Fernandes (2019) The Dark Side of the Contract: Do Government Audits Reduce Corruption in the Presence of Displacement by Vendors? https://www.econ.uzh.ch/dam/jcr:ec360d66-9272-4a8d-9e4d-65245e586d32/TheDarkSideOfTheContract.pdf


Gupta, Akhil. (2012) Red Tape: Bureaucracy, Structural Violence, and Poverty in India. Duke University Press. 10.1215/9780822394709.


Hanna, Rema, and Shing-Yi Wang (2017) “Dishonesty and Selection into Public Service: Evidence from India.” American Economic Journal: Economic Policy 9 (3): 262-90. DOI: 10.1257/pol.20150029


Hendryadi, Suratna, Suryani & Budi Purwanto | Md. Mahmudul Alam (Reviewing editor) (2019) Bureaucratic culture, empowering leadership, affective commitment, and knowledge sharing behavior in Indonesian government public services, Cogent Business & Management, 6:1, DOI: 10.1080/23311975.2019.1680099. https://www.tandfonline.com/doi/full/10.1080/23311975.2019.1680099


Heyman, Josiah (2012) "Deepening the Anthropology of Bureaucracy." Anthropological Quarterly 85, no. 4. doi:10.2307/41857300.


Honig, Dan and Pritchett Lant (2019) “The Limits of Accounting-Based Accountability in Education (and Far Beyond): Why More Accounting Will Rarely Solve Accountability Problems.” Center for Global Development, Working Paper 501.


Hull, Mathew S. (2012) Government of Paper: The Materiality of Bureaucracy in Urban Pakistan, Berkeley (California): University of California Press, 320 pages.


Hull, Matthew S.(2003) “ The file: agency, authority, and autography in an Islamabad bureaucracy”, Language & Communication Journal 23(3):287-314, doi:10.1016/S0271-5309(03)00019-3


Iyer, D. (2011). Tying Performance Management to Service Delivery: Public Sector Reform in Malaysia, 2009-2011. https://successfulsocieties.princeton.edu/sites/successfulsocieties/files/Performance%20Management%20Malaysia_ToU_1.pdf


Iyer, Lakshmi, and Anandi Mani (2012) “Traveling Agents: Political Change and Bureaucratic Turnover in India.” Review of Economics and Statistics 94 (3) (August): 723–739. doi:10.1162/rest_a_00183. http://nrs.harvard.edu/urn-3:HUL.InstRepos:33782993


Jakiela, Pamela  (2018) “Three Lessons from the World Bank’s New Worldwide Bureaucracy Indicators Database” Commentary and Analysis, Centre for Global Development. https://www.cgdev.org/blog/three-lessons-world-banks-new-worldwide-bureaucracy-indicators-database 


Kahneman, Daniel; Tversky, Amos (1979). "Prospect Theory: An Analysis of Decision under Risk". Econometrica. 47 (2): 263–291. doi:10.2307/1914185. ISSN 0012-9682. JSTOR 1914185.


Kalaj Jozefina, Rogger Daniel and Somani Ravi (2020) “Bureaucrat Time-Use and Productivity: Evidence from a Survey Experiment”. Working Paper. 




Kapur, Devesh (2020) "Why Does the Indian State Both Fail and Succeed?" The Journal of Economic Perspectives 34, no. 1 : 31-54. 0. doi:10.2307/26873528.


Kennedy, John F. (1963) “Remarks Prepared for Delivery at the Trade Mart in Dallas, TX”. https://www.jfklibrary.org/archives/other-resources/john-f-kennedy-speeches/dallas-tx-trade-mart-undelivered-19631122. 


Khemani, Stuti (2007) “Does delegation of fiscal policy to an independent agency make a difference? Evidence from intergovernmental transfers in India” Journal of Development Economics. Volume 82, Issue 2,


Krishnan, K.P. and Somanathan, T.V. (2017) “The Civil Service.” Rethinking public institutions in India. Edited by Devesh, Mehta, Pratap Bhanu, and Milan Vaishnav. 2018. Oxford University Press. Print. Pp 339-416


LaForge, G. (2016). Changing a Civil Service Culture: Reforming Indonesia’s Ministry of Finance, 2006-2010. https://successfulsocieties.princeton.edu/sites/successfulsocieties/files/GL_CC_Indonesia_MOF_FORMATTED_5May2016_ToU_1.pdf


Leaver, Clare. (2009) "Bureaucratic Minimal Squawk Behavior: Theory and Evidence from Regulatory Agencies." American Economic Review, 99 (3): 572-607.


Levitt, Steven D. and Dubner, Stephen J. (2006). “Freakonomics: A Rogue Economist Explores the Hidden Side of Everything”, p.22, Penguin UK


Lichand, ‪Guilherme and Gustavo Fernades (2019), “The Dark Side of the Contract: Do Government Audits Reduce Corruption in the Presence of Displacement by Vendors?”, University of Zurich, Switzerland, mimeo.


Majeed, R. (2010). Strengthening Public Administration: Brazil, 1995-1998. https://successfulsocieties.princeton.edu/sites/successfulsocieties/files/Brazil%20Public%20Administration%20ToU_1.pdf


Mangla, Akshay (2015) "Bureaucratic Norms and State Capacity in India: Implementing Primary Education in the Himalayan Region." Asian Survey 55, no. 5 : 882-908. https://www.jstor.org/stable/26364317.


Maria Paula Gerardino & Stephan Litschig & Dina Pomeranz. 2017. “Distortion by Audit: Evidence from Public Procurement,” NBER Working Papers 23978, National Bureau of Economic Research, Inc.


Mathew, A. Santhosh and Purohit, Bhumi (2020) “India’s public finance architecture is the real culprit in Gorakhpur and Bhagalpur”, LiveMint. https://www.livemint.com/Opinion/I6UMVabJwiukGs7rB5bmSM/Indias-public-finance-architecture-is-the-real-culprit-in-G.html


Maurice Yolles. (2016). Understanding Bureaucracies. DOI: 10.13140/RG.2.1.2125.0809. Liverpoool John Moores University


Md. Morshed Alom (2020). Public Sector Organizational Culture: Experience from Frontline Bureaucracies [Online First], IntechOpen, DOI: 10.5772/intechopen.91177. Available from: https://www.intechopen.com/online-first/public-sector-organizational-culture-experience-from-frontline-bureaucracies


Media Cloud. MIT Center for Civic Media and the Berkman Klein Center for Internet and Society at Harvard University.


Merchant, Minhaz. (2012) "Indian Asphyxiation Service (IAS)". Economic Times Blog. https://economictimes.indiatimes.com/blogs/headon/indian-asphyxiation-service-ias/.


Michael Carlos Best, Jonas Hjort, David Szakonyi. (2019) Individuals and Organizations as Sources of State Effectiveness. https://blogs.cuit.columbia.edu/mcb2270/files/2019/10/BestHjortSzakonyi-StateEffectivenessProcurement_Oct2019.pdf


Milgram, Stanley. (1974) "Obedience to Authority : An Experimental View". Published by Harper and Row.


Mishra, Twesh (2016) “Cannot Progress in 21st Century With 19th Century Administration: PM Modi”, BloombergQuint, https://www.bloombergquint.com/politics/cannot-progress-in-21st-century-with-19th-century-administration-pm-modi


Muralidharan, Karthik (2016) "A New Approach to Public Sector Hiring in India for Improved Service Delivery," India Policy Forum, National Council of Applied Economic Research, vol. 12(1), pages 187-236.


Namutebi Edith Nsubuga (2017) “Coordination Failure in Government Offices: A Case of Uganda”, Dissertation submitted for Degree of Doctor of Philosophy in International Development Studies, National Graduate Institute for Policy Studies (GRIPS). Performance Incentives


Peter, Guy B (2018) “The challenge of policy coordination”, Taylor & Francis Online, https://www.tandfonline.com/doi/full/10.1080/25741292.2018.1437946


Prendergast, Canice (2001) The Limits Of Bureaucratic Efficiency. University of Chicago & NBER. http://web.mit.edu/14.282/www/Prendergast_BureauctraticEfficiency.pdf


Press Information Bureau (2020) “Cabinet approves "Mission Karmayogi"- National Programme for Civil Services Capacity Building (NPCSCB)” https://pib.gov.in/PressReleseDetailm.aspx?PRID=1650633


Press Trust of India (2019) “153 IAS officers appointed on central deputation in 2018-19, lowest in last 5 years” The Economic Times. https://economictimes.indiatimes.com/news/politics-and-nation/153-ias-officers-appointed-on-central-deputation-in-2018-19-lowest-in-last-5-years/articleshow/71792555.cms?utm_source=contentofinterest&utm_medium=text&utm_campaign=cppst 


Rasul Imran, Rogger Daniel, Williams Martin J (2018) “Management and Bureaucratic Effectiveness Evidence from the Ghanaian Civil Service. World Bank Policy Research Working Paper 8595. World Bank Development Economics Development Research Group


Rasul, Imran., Rogger, Daniel., and Williams, Martin J. (2017). Management and bureaucratic effectiveness: A scientific replication. International Growth Centre. Reference number: S-33301-GHA-1.




Rema Hanna and Shing-Yi Wang (2017) Dishonesty and Selection into Public Service: Evidence from India. American Economic Journal: Economic Policy, 9(3): 262–290 https://doi.org/10.1257/pol.20150029


Sanyal, Sanjeev (2020) “Risk vs Uncertainty: Supervision, Governance & Skin-in-the-Game”, Government of India, Ministry of Finance, Discussion Paper No. 1/2020-DEA


Schultz, Thomas M (2005) “Organizational culture: An impetus to influence organizational behavior and decision making” . Graduate Student Theses, Dissertations, & Professional Papers. 8752. https://scholarworks.umt.edu/etd/8752. https://scholarworks.umt.edu/cgi/viewcontent.cgi?article=9787&context=etd


Shleifer, Andrei, and Robert W. Vishny (1993) "Corruption." The Quarterly Journal of Economics 108, no. 3 : 599-617. http://www.jstor.org/stable/2118402.


Singh Arkaja (2020) “Hunger, COVID-19 and The Indian Administrative State”, Centre for Policy Research Working Paper.


Sneha P., Shah K., Mariwala V., Varghese A., Das M. (2021)  The First Wave of the COVID-19 Pandemic in Kerala - A State Capacity Perspective, Working Paper, IDFC Institute


Susan Rose-Ackerman, Bonnie J. Palifka. (2016). Bureaucratic Corruption. In Corruption and Government: Causes, Consequences, and Reform (pp. 51-92). Cambridge: Cambridge University Press. doi:10.1017/CBO9781139962933.004


Swarup, Anil (2019) “Not Just a Civil Servant” Unicorn Books Pvt Ltd. 


Szucs, Ferenc (2020) “Discretion and Corruption in Public Procurement” Working Paper https://drive.google.com/file/d/1-2f_OEZp6j1cSxeTR1YEUtHzA-SDNnzG/view


ThePrint Team (2018) “Parliament amends anti-corruption act: Will it weaken the law or restore fairness?”, ThePrint https://theprint.in/talk-point/parliament-amends-anti-corruption-act-will-it-weaken-the-law-or-restore-fairness/88379/


Vaishnav Milan, Khosla Saksham (2016) “The Indian Administrative Service Meets Big Data” Carnegie Endowment for International Peace https://carnegieendowment.org/2016/09/01/indian-administrative-service-meets-big-data-pub-64457


Vanusia, Phiroze (2020) “Greek, Latin and The Indian Civil Service.” The Cambridge Classical Journal 51 (2005): 35-71. http://www.jstor.org/stable/44698241.


Williams, Martin J. (2015) From Institutions to Organizations: Management and Informality in Ghana’s Public Bureaucracies. https://martinjwilliamsdotcom.files.wordpress.com/2015/09/martinwilliams_frominstitutionstoorganizations_150915.pdf


World Bank, “Worldwide Governance Indicators Indicators 1.1.” https://info.worldbank.org/governance/wgi/  https://datacatalog.worldbank.org/dataset/worldwide-bureaucracy-indicators 


Yang, Dean (2004) Can Enforcement Backfire? Crime Displacement in the Context of Customs Reform in the Philippines. University of Michigan, Ford School of Public Policy Working Paper No. 02-010, Available at SSRN: https://ssrn.com/abstract=546523 or http://dx.doi.org/10.2139/ssrn.546523 




Back to TOC



x Close Window

Please verify your email address to access this content