Political Pressures and Bureaucratic Consequences: Vignettes of Janmabhoomi Implementation in Andhra Pradesh

Related Literature

Formally, India presents a modern, Weberian bureaucratic state (Kaviraj, 2005) projecting a democratic, development-oriented identity that was deeply subverted in practice (Jacob, 2016). Wood’s (2000, p. 222) observation for rural works programmes in Bangladesh is equally applicable for India:

Much of the thinking about good governance and development administration explicitly or implicitly makes judgements about performance (content, structure and outcomes) in terms of Weberian criteria concerning well-functioning bureaucracies operating within liberal-democratic, pluralist political systems. In other words, a particular concept of rationality is assumed and real societies are analysed by reference to it.

The literature has argued that the Indian state was shaped by dominant classes—industrial capitalists, rich farmers and the bureaucracy and salaried professionals (Bardhan, 1984)—and increasingly by corporate capital (Chatterjee, 2008). ³ However, such a macroscopic perspective necessarily reduces nuanced diversity within the state. Gupta (2012, p. 187) points out ‘the tendency in academic work to reify state institutions into overly purposeful, unitary, and motivated actors’. On somewhat similar lines, Fuller and Harriss (2001, p. 3) note that the macroscopic approach ‘seems to presuppose a unified intentionality and internal consistency, whereas in practice state functionaries often, or even normally, pursue competing agendas at cross-purposes with each other’.

Corbridge et al. (2005, p. 9) distinguish ‘sightings’ of the state by citizens, and in particular the poor, from ‘sightings made by government officers in different line departments’. ⁴ Although the literature focuses more on the former, that is, the interface of policy institutions with the public/citizenry (for instance, Vasan, 2002), scholarship has also turned to understanding the latter, that is, the internal business of policy institutions (for instance, Sridhar, 2008). However, micro-level, ethnographic studies of lower level officials and politicians continue to be inadequate. While we may know in the abstract that ‘[s]ince major government policies have their final point of implementation very low down in the bureaucracy, they are reinterpreted beyond recognition’ (Kaviraj, 1990, p. 13), we do not know enough about the ways in which internal government processes, across multiple long hierarchies, end up creating wedges between enunciated policy and implemented policy outcomes.

The literature on lower bureaucracy in India, including street-level bureaucracy, is thin. In the context of irrigation bureaucracy around the Narmada dam projects in Gujarat, Aandahl (2010, p. 305) notes that field-level offices are ‘overburdened with the task of managing both physical and social engineering in the command area’. Further

the power of the SSNNL [Sardar Sarovar Narmada Nigam Ltd, the official agency] engineers are limited both from above and from below, they have to balance the needs, demands and concerns of both project beneficiary farmers and policy makers and politicians. At the field level, the field engineers are unable to carry out efficiently their supervisory and advisory task because of the sheer size of the area they are supposed to cover. In addition, they told me about a heavily top-down communication structure where orders are handed down from above and feedback from the ground level is not welcomed but fake statistics are accepted as fact. (p. 306)

Mathur (1995) notes that policymakers seldom account for contextual and ground-level realities of implementation. Gupta (2012) argues that lower bureaucratic practices produce arbitrary outcomes because structural inequality is reproduced in its actions. Further, there is a tendency to equate programme success with meeting target outputs rather than explore (broader and deeper) programme impacts: as Guhan (1980, p. 1981) notes, ‘it was possible to meet the programme targets and miss the policy targets’. The first Administrative Reforms Commission (1967) noted that ‘performance is geared to inspections, inspections are based on targets and targets are related to financial provisions’ (cited in Mathur, 1995, p. 2706). Aiyar and Bhattacharya (2016, p. 63) argue that in the underlying tight top-down formal hierarchy, lower level functionaries see themselves merely as ‘“post offices”—used simply for doing the bidding of higher authorities and ferrying messages between the top and bottom of the education chain’. Nevertheless, as Lipsky’s (2010 [1980]) pioneering work shows, it is street-level bureaucrats that actually ‘make’ policy in its implementation, no matter what high policy formulations may articulate. Lipsky argued that ground-level pressures and differences in agendas within the bureaucracy together explain how policy is ‘subverted’ in the practices of street-level bureaucrats. Consistent with this, Ashwin (2016) argues that in fulfilling instructions passed on from above, lower bureaucracy is forced to employ unorthodox, non-Weberian practices (‘beech ka rasta’, middle path). Related coping strategies have also been investigated for lower health bureaucracy (Mishra, 2014).

A larger question hangs above the entire enterprise of a gigantic bureaucracy-led system for government programmes: the difficulty of making the system do the bidding of heroically planned, non-contextual, non-socially embedded development activities. Guhan (1980, p. 1977) notes that for purposes of targeted poverty alleviation schemes, enormous and high-quality bureaucratic resources—‘a Tughlaqian enterprise’—would be needed to scrutinize poverty levels of households:

it will involve the identification of the poorest (viz., the basic units for poverty eradication) in the vast universe of the poor… and not just the mere verification … of the genuineness of alleged need. … An army of over 100,000 officials will be required to complete these enquiries…. (emphases in the original)

This point turns out to be pertinent for the vignettes and narratives we present.

Background

Below, we sketch out the basic elements of Andhra state politics and the governing party’s Janmabhoomi programme.

Fieldwork

The focus of this study is on ration card distribution. Among Janmabhoomi’s several welfare schemes, one was to rapidly increase ration card distribution. According to the official website, the goal was the ‘[d]istribution of 12.43 lakh new ration cards during the programme after ensuring purification of data and by generating exceptional reports’ (italics added). ⁸ This article explores the implementation of procedures for scrutinizing ration card applications—‘purification of data … by generating exception reports’—at the district and mandal levels. Our focus is on the ‘white’ ration cards intended for BPL households. ⁹ These cards entitle holders not only to subsidized food items but also serve as a gateway for other subsidies ranging from education (including higher education) to health to agriculture.

In the ration cards component of Janmabhoomi, detailed procedures were specified for inviting and scrutinizing applications and awarding (or rejecting) ration cards. These are listed in Table 1. ¹⁰ We develop vignettes regarding how these procedures were actually implemented in Shodhana mandal of Kiranam district of Andhra Pradesh. ¹¹ For reasons of confidentiality, we use pseudonyms for district, mandal and individual names. Field research was conducted in November–December 2015, and follow-up visits were undertaken in January and March 2016, in addition to several telephonic conversations with mandal officials in the intervening period. There are about 50 mandals in Kiranam district, of which almost 60 per cent—including the study mandal—are fully rural. Census 2011 data indicate that in terms of ‘main’ occupations, agricultural labourers constitute the majority of the working population of Shodhana mandal (about 60 per cent), and cultivators about 15 per cent. ¹²

OPEN IN VIEWER

Table 1.

Procedures for Processing Ration Card Applications

1A	When the government calls for applications (once in every one to two years), individuals can apply at Meeseva centres. (This information gets automatically collated electronically and is accessible to the RI, CSDT, and Tahsildar (all mandal-level officers).19
1B	Alternatively, individuals can apply through village Janmabhoomi Committees during Janmabhoomi Grama Sabhas (village meetings).
2	All applications made via Meeseva or Janmabhoomi cumulatively appear on the Civil Supplies portal.
3	Each VRO receives the applicant list of the concerned jurisdiction via the mandal office.
4	Each VRO makes physical enquiries regarding each applicant, verifies her/his suitability, writes a report, and submits it to the RI.
5	The RI recommends or rejects applicants based on the VRO’s reports.
6	The RI’s recommendation is electronically transmitted to the CSDT, who also has to recommend / reject those on the list.
7	The CSDT’s recommendation goes to the Tahsildar, who puts the final digital signature.
8	This information is transmitted electronically to the district office, which arranges for printing and disbursement of cards.
9	Cards are distributed in a public meeting (Janmabhoomi) or through the dealer.

Source: Authors’ observations, validated by a Deputy Tahsildar (DT) and a Village Revenue Officer (VRO).

The administrative structure of the Revenue Department in the district is presented in Figure 1. The Joint Collector (JC) is the second highest officer (after the Collector). The JC runs the Revenue administration and is concerned chiefly with land matters and civil supplies (including ration cards policy implementation).

Inside the Lower Bureaucracy: Vignettes of Collectorate–Mandal Relations

The District Collector’s office in Kiranam district holds a weekly video conference call with mandal officials. During each such call, there is typically a different prioritized theme for discussion. On a specific date in early December, the theme for the video conference was issuance of new ration cards. The JC made a video presentation of the procedures to be followed—listed in Table 1—to a video audience of Civil Supplies Deputy Tahsildars (CSDTs) and Revenue Inspectors (RIs) of all mandals in the district (see Figure 2 for administrative structure).

OPEN IN VIEWER

Figure 2.

Source: Authors’ observations.

As the JC commenced his video presentation, at the Shodhana mandal office, the Tahsildar saw the presentation in the office conference room. She asked the office attendant to call the RI. The attendant literally ran to the RI’s office to convey this. The RI came in a hurry, carrying a diary and pen. She sat next to the Deputy Tahsildar (DT) and looked worried. Since by then Shodhana mandal was already at step 2 of ration card procedures listed in Table 1 (i.e., village Janmabhoomi Committees had created recommended applicants lists), the Tahsildar asked the DT to print the Janmabhoomi Committee lists, and the DT asked the typist to print these out. ¹³

The next video conference with the JC occurred, as scheduled, exactly 1 week later. As the conference was underway, it became clear that the JC’s concern was that the district would fall behind the rest of the state in ration card approvals, and be upbraided for this by superiors in the state government administration. It was apparent that he had come to know that very few mandals in the district had made progress on ration card approvals, and that in fact many mandals had not even started this work. This is confirmed by field observations at the Shodhana mandal office in the days prior to and following the first video conference: there was little action, interest or discussion on the matter of ration card applications. Rather, mandal officials were busy following up on several other orders on land-related matters coming from higher-ranking officers at the district level.

Given this immediate context, the JC became visibly angry during the video conference when confronted with inaction from the mandals. He shouted at the Revenue Divisional Officers (RDOs) and Tahsildars:

It has been a week since I asked you to clear the ration cards, a large portion has still not been completed. Janmabhoomi is coming closer, what are you actually doing, are all Tahsildars and CSDTs sleeping? What are you RDOs doing?

Progress on ration card approval was particularly poor in Karkalli division, and the JC proceeded to upbraid its RDO, Tahsildars and CSDT. The RDO demurred (‘Sir, we are doing…’), upon which the JC shouted at the RDO:

Shut up… What are you doing? What are you actually doing? Where is your CSDT and what is he doing?

When the CSDT was identified on video, the JC turned on him:

Ayya, ¹⁴ what are you doing, are you playing games, are you play-acting? Instead of working, what are you thinking actually? If you don’t work, I will suspend you, be careful.

Everybody on the video conference across the district saw the CSDT try to respond, with tension and shivers:

I am working, Sir, but the internet is not working properly…

The JC interrupted him:

Shut up, does this look like fun and games to you? Don’t give me long-winded excuses, you should be seen working. What are you thinking, you idiots, useless fellows, you are not capable of work. I don’t know what you will do, I want everything to be cleared by tomorrow evening. [Turning to the corresponding RDO] RDO, follow-up.

The JC then warned the RDOs, Tahsildars and CSDTs to get the work done as soon as possible. He emphasized, again, the need to exercise care in cases where applications passed by Janmabhoomi Committees were being rejected.

Zooming out from the Vignettes: Clientelist–Populist Politics

Placing the above vignette in its larger political context, we argue that the electoral imperative was manifested in clientelist–populist politics, which in turn produced political pressures on the bureaucracy. The argument is summarized in Figure 1. The present section focuses on argument (1) in Figure 1, clientelist–populist politics.

It turns out that perceptions around village Janmabhoomi Committees are crucial in understanding how governance played out in the above vignette. ¹⁵ In the video presentation, the JC asked mandal officials to accept most of the applications sent by the Janmabhoomi Committees. He emphasized this point repeatedly, noting that strong justification had to be provided if a mandal official rejects an application. ¹⁶ What was left unsaid but understood by all was that since Janmabhoomi Committees were the brainchild of the governing TDP, going against their wishes would pit the revenue bureaucracy against governing party politicians, both those at the state-level and those at the district-/mandal level. In fact, the TDP’s political opponents have long complained that Janmabhoomi Committees are a political ruse to provide partisan administrative favours. For instance, the leader of the opposition (YSR Congress leader) Y.S. Jagan has demanded the cancellation of Janmabhoomi Committees:

He opined that benefiting TDP leaders in the name of Janmabhoomi Committees was unfair…. Authority was being placed in the hands of a handful of TDP activists. Sanction of pension and ration cards also was dependent on the signature of those activists. ¹⁷

The logic of Janmabhoomi Committees can be placed within a broader understanding of the historical trajectory of electoral politics. In the early, post-independence decades, consonant with the rest of India, politics in the state was based on vote bank politics through patron–client relationships mopped up by dominant party, dominant caste politicians for electoral purposes (Reddy, 1989). In several parts of India including Andhra Pradesh, this started changing in the 1970s and was influenced by national politics: Indira Gandhi’s rivalry with the Congress ‘Syndicate’ produced a move to left populist politics and the fraying of patron clientelism. ¹⁸ Robinson (1988, p. 209) provides a village-level perspective from Mallannapalle of Medak district, noting how in the 1970s’ elections traditional patrons began to be bypassed by politicians arriving to seek the people’s vote:

[Congress Party politician] G. Veeraiah came to Mallannapalle to campaign for the Congress. He did not meet [the traditional patrons] Lakshma and Narsimha privately as on previous occasions, but instead held an open meeting in front of their house. He urged the people to ‘support Indiramma’ and to vote for the ‘party which cares about the poor people’…. Many Mallannapalle villagers commented that this was the first election ever held in which they had not been instructed by the patels in how to vote.

The turn towards populism brought a slew of development yojanas (schemes), for instance more widespread distribution of government credit and the creation of multiple small farmers’ programmes. Wyatt’s (2013, p. 366) analysis for the Indian context (and neighbouring Tamil Nadu in particular) is applicable here: ‘populism as an ideological construct that celebrates the importance of the ordinary people, asserts these people should not be divided by social hierarchy and justifies improvements in their welfare’. With the arrival of NTR in state politics in the 1980s, this populism received a cultural slant as well (Kohli, 1988).

Nevertheless, while populism replaced patron clientelism (with the 1970s as the turning point), the essential logic of clientelist transactions—exchange of votes for favours/benefits—has continued. That clientelism survives and even thrives despite increased political competition is not necessarily paradoxical, as has been observed by Levitsky (2007) for Latin America and Lindberg and Morrison (2008) for Africa. As Reddy and Haragopal (1985, p. 1155) note of the middleman (pyraveekar) who arranges such local-level exchanges:

The institution of pyraveekar has its own well-established political linkages to both the local power structure in the villages and the outside political framework. … The politicians contesting elections have come increasingly to lean on these pyraveekars at election time, and in the absence of organized local units of political parties, the pyraveekars have come to play the role of party cadres.

The literature is consistent with the above presentation of clientelism coexisting with populism. Srinivasulu (2004, p. 3851) notes:

[The] organisational base of the TDP at different levels began to comprise of people belonging to the class of contractors, builders and even speculators. … This contractor class, for whom politics is primarily a business proposition, the developmental work and even the FFW [Food For Work] programme has become a ‘feeding channel’.

So, while competitive populism is a reality of state politics—before Naidu was elected in 2014, his Congress Party predecessor YSR also engaged freely in populist politics—clientelism and ‘leakage’ from the government apparatus continue. Elliott (2016, p. 29) observes in a recent study of Andhra Pradesh:

[Naidu’s] initiatives did little to disrupt the ongoing system of extracting bribes for ration cards, siphoning off food from government allocations to resell in the market and paying overcharges to local contractors … Mooij (2003) reports that reforms in AP did not lead to a fundamental break in either pro-poor schemes or in ways in which resources are distributed through political networks and used for local party-building purposes. ¹⁹

Top-level Political Pressures on Bureaucracy

Having established the nature of the state’s clientelist–populist politics, we now turn to its implications for top-level pressures on the bureaucracy (argument (2) in Figure 1). Popular and media accounts attest to the pressure that Naidu and his administration bring to bear on the bureaucracy to implement government programmes, and particularly flagship programmes with high political stakes. In the context of the vignette, news reports suggest that the state government was placing considerable emphasis on increasing ration card distribution in a very short span of time. For instance, Minister for Civil Supplies Paritala Sunitha announced in late December that as part of the 3rd phase of the Janmabhoomi programme (to be launched in January 2016), fresh ration card applications would be allowed, and these would be processed and ‘new ration cards would be given by January-end’, totalling over 12 lakhs—of which four lakhs were related to ‘rectification of data and inclusion of members’. It is inconceivable that the procedures described in Table 1 could be properly implemented on such a scale over the space of a few weeks. Such frenetic push for ‘progress’ likely put considerable pressure on JCs to show substantial increase in ration card approvals in their districts. There was also pressure to ensure that appropriate stocks were maintained at ration shops.

In June 2015, initiating Janmabhoomi—Maa Vooru, in Chittoor District, Naidu is reported to have pressured the bureaucracy thus:

Naidu lost his cool and warned officials against being negligent in completing their respective tasks. … Stating the government increased the salaries of employees by giving 43 percent fitment despite financial crunch, he said the government also provided health cards for the welfare of employees hoping for their active involvement in implementing government programmes for the development of the state. ‘But, if you fail to fulfil the same, I will not spare you,’ he said. (New Indian Express, 4 June 2015)

Similarly, in November 2015 in Nellore district, Naidu chided the bureaucracy while organizing a whirlwind of teleconferences with them:

Chief Minister N Chandrababu Naidu expressed dissatisfaction over poor response from the district administration as well as TDP cadres in the district after huge disaster struck the district twice. He reportedly warned the negligent officials that they would be shifted to other places basing on their performance. … Chief Minister organised tele-conference with tahsildars, MPDOs [Mandal Parishad Development Officers], Panchayat Secretaries, and other revenue officials and Janmabhoomi Committee members on the relief and rehab [rehabilitation] measures and the flood situation in the areas. (The Hans India, 22 November 2015)

Similar observations have been made in the academic literature regarding Naidu’s relationship with the bureaucracy. Manor (2002, p. 54) quotes a district-level bureaucrat: ‘he personally hounds us constantly’. For the most recent (post-2014) Naidu regime, Elliott (2016, p. 29) writes: ‘Naidu devoted enormous energy to building up party cadres and local bureaucrats, keeping watch over them on a daily basis via video conferences’ and the bureaucracy ‘often felt harassed by his computer-aided monitoring sessions’. Implicit is the portrayal of the leader as a progressive populist fighting for ‘the people’ against ‘the system’, an image cultivated earlier by Indira Gandhi at the national level: ‘the populist image of Indira Gandhi fighting vested interests on behalf of the poor majority was set against the “implementation failures” of planned development, attributed to the reactionary or corrupt tendencies of provincial-level government and party leaders’ (Witsoe, 2013, p. 49).

Bottom-level Political Pressures on Bureaucracy: A Vignette

We now turn to the second set of pressures on the lower bureaucracy, also derived from the logic of clientelist–populist politics (argument (3) in Figure 1). Below, we present a vignette from the mandal revenue bureaucracy on digitization of land records, which is another function (with attendant procedures for verification) that the mandal bureaucracy has been tasked with from above.

The Sarpanch of Narayanapuram, along with an ally, met the DT in his office in order to fast track his (private) application for verification of land records. The corresponding VRO happened to be sitting at the office computer engaged in some election-related work. The DT told the VRO: ‘Clear the applications of the Sarpanch on a priority basis.’ He then turned to the Sarpanch and said: ‘Sir, please give your acknowledgment numbers [application details] to the VRO, he will do [the work].’ The VRO deferentially said that he was presently processing applications from Naryanapur village and that he would start on Narayanapuram after that. The DT affirmed, but asked the VRO to process the Sarpanch’s application on a priority basis. With this, the Sarpanch and his ally left the office.

After an hour, by about 7 pm, another team from the same village visited the DT’s office with a similar request. This team consisted of TDP affiliates: the Vice Mandal Praja Parishad member (MPP) and his ally. A similar routine was played out: the DT directed the VRO to take acknowledgment numbers and process their applications on a priority basis, the VRO agreed and the visitors departed. The DT then remarked to the VRO: ‘See, you are young at this job. If we do the work of these people, they will control [prevent] other villagers from coming to the office [with requests], so [you should] process their applications on a priority basis.’ The VRO nodded: ‘OK, Sir.’

After about 10 days, the MPP was back at the office. He asked the VRO whether he had started processing applications from his village. The VRO replied: ‘Ayya, I started [processing applications from Narayanapuram] only yesterday.’ The MPP asked that his ally’s application be processed the same day, post lunch. Since this required a village visit (‘field enquiry’), the VRO said he would do it the next day, but was pressured by the MPP to do it the very same day. Accordingly, he visited the village and made enquiries regarding five applications from the MPP’s ally.

Meanwhile, the Narayanapuram Sarpanch and his ally came to know about the VRO’s village visit on behalf of the MPP. When the VRO got back to his mandal office, he was met there by the Sarpanch and about eight henchmen, who questioned him as to why he had not processed their application yet despite the DT’s strictures. The group was enraged, and the situation chaotic. The VRO tried to placate them by saying that he would process the application the next day itself. The visitors issued dire warnings: ‘We will see your end, what do think of us? Cannot do anything? See what will happen to you now.’ The worried VRO tried pleading with the Sarpanch, who stayed silent. By about 7 pm, after 20 minutes of shouting by the henchmen, they left saying that they would take up the matter immediately with the DT. Meanwhile, the VRO started looking into the relevant papers.

After 10 minutes, an attendant from the DT’s office came by motorcycle and told the VRO that the DT required his presence at his office. Shouting could be heard from outside the DT’s office, all the way to the road. Upon entering, the DT rudely upbraided the VRO, giving him no chance to explain, and calling him ‘useless’. The DT then asked the attendant to bring the relevant (‘1B’) land register for Narayanapuram and asked the VRO to locate the Sarpanch’s application. The VRO was ordered to immediately write out his ‘enquiry report’ exactly according to the existing record and also to immediately update it online. The VRO went back to his room to get the relevant applications of the Sarpanch. The henchmen waited outside the DT’s office while the Sarpanch and the DT—along with a journalist affiliated with the Sarpanch—sat talking inside. The VRO came back with the relevant applications, instantly ‘verified’ them, wrote out his ‘report’, got the DT’s signature and gave it to the typist to make the digital entry. All applications by the Sarpanch and his allies were thus processed, except for two which could not be found. The VRO promised to process these the next morning. The DT reminded him: ‘Your first duty in the morning is this [to clear those two applications], and then only [should you do] everything else.’ The whole process took an hour. The VRO took their permission to leave, while the Sarpanch, journalist and DT continued their casual political discussion for another half hour.

The above vignette speaks to the pressures brought on lower bureaucracy directly by local politicians due to the latent power conferred by the logic of clientelist–populist politics. If lower bureaucracy does not respond adequately, local politicians can and do complain to state-level politicians—who depend on them for local vote mobilization—and who then apply renewed pressure on the lower bureaucracy through the bureaucratic hierarchy.

Consequences of Political Pressures: Further Vignettes from Lower Bureaucracy

The previous three sections argued that clientelist–populist politics simultaneously produced high-level political pressures on the bureaucracy and concomitant local pressures on lower bureaucracy. To explore consequences of these pressures, we turn again to the ration card narrative from Shodhana mandal. It turns out that the repeated pressures placed on mandal officials by district officials and others produced arbitrary outcomes, as described below.

Even as the JC’s rant—about lower bureaucracy being ‘useless idiots’ for not following the official (and impossible) ration card procedures—was beamed into the Shodhana mandal office on video, the Tahsildar called the typist, who dropped all her work and came running. The Tahsildar told the typist to open the e-governance web portal (‘RI login’) and start clearing ration card applications waiting for approval. ²⁰ The Tahsildar asked the DT to follow up on this. The DT replied, ‘Ma’am, I will take care of it.’ The RI, sitting right there, was merely a spectator. The DT told the typist to accept the vast majority of applications through the RI login—clearly with the JC’s stinging remarks in mind—and to randomly reject some so that it would appear that discretion was indeed used. Since night had fallen by now, the typist said she would do the needful the next morning. The DT was satisfied, and the officials finally left the office by about 9 pm.

The next morning, the DT received a phone call from the RDO to clear all pending ration card applications as instructed by the JC. The DT again asked the typist to open the RI login and clear all applications. She cleared all the applications by 11 am and informed the DT, who in turn informed the Tahsildar. The Tahsildar then called the CSDT and asked him to do his part (viz., sign off on the changes made by the typist on the RI login). By about 4 pm, the CSDT had finished this work, again without appropriate scrutiny. By about 4:30 pm, printouts of the final approved list of accepted applicants were given by the typist to all the VROs of Shodhana mandal.

The next day, the VROs were having a mid-morning chat at the mandal office. One VRO turned to another and noted:

Anna [Brother], how the new Sir [JC] struck off as he liked… In my list there are teachers (government employees) who were approved for ration card… ²¹ By chance if they [higher level bureaucrats] come for enquiry in future, we will be the first to be suspended, no? There are many blunders in this list of mine.

In response, the second VRO said to the first:

What can we do, Thammudu [Brother], except do what they say? These things are not new for us, right? In my list also, there are many people who should not get [ration cards]—government employees and rich landlords—but they got them.

In later conversations with a VRO from Shodhana mandal, several specific cases of ‘Type 1’ and ‘Type 2’ targeting errors in ration card distribution emerged (eligible households not receiving cards and ineligible households receiving cards, respectively). For instance, the VRO mentioned that two constables from a village received white ration cards although they were ineligible, while other eligible persons from their village did not get cards. Even 3 months after issuing the cards, a constant stream of people was coming to the Tahsildar office for corrections. This is consistent with Powis’s (2012, p. 131) observation of a ‘disjuncture between the image and practice of Janmabhoomi’:

All ‘grievances’ raised during Janmabhoomi were meant to be recorded and ‘Action Taken Reports’ were to be read out in the following round to report on progress of applications received and the progress of the inquiry. Official data shows that nearly all applications made in the JB [Janmabhoomi] had been ‘resolved’, though this was not widely reflected in the discussions with villagers. As one man told us, ‘The joke is that the officials are taking our applications and selling them as scrap paper to wrap palli (groundnuts) in.’

Discussion and Conclusion

This paper has used ethnographically informed vignettes from the lower bureaucracy in Andhra Pradesh to open up a discussion of broader political dynamics. To throw light on our first vignette (on bureaucratic pressures at the mandal level), we used a political framework driven by clientelist–populist electoral logic manifested in a dual set of pressures on the bureaucracy (Figure 1): pressures from state-level politicians on higher bureaucracy and passed through to lower bureaucracy and pressures from local politicians directly on lower bureaucracy. The vignettes sketch out a story of political pressure to fast track ration card approvals—implicit more than explicit pressure—on the district administration, translating vertically down the hierarchy to pressures on the mandal administration, with the result that the only way the mandal administration could ‘deliver’ on ‘progress’ was through subversion of official Weberian procedures.

While the individual components of the argument summarized in Figure 1 have been noted in the literature, our contribution lies in pulling them together into a common framework linking the ‘politics game’ and the ‘bureaucracy game’ (the left and right sides of Figure 1, respectively) in a grounded manner. In particular, a crucial part of official procedures (shown in Table 1) was that the VRO would be the frontline official in the field to scrutinize the legitimacy of applications recommended by (partisan) village Janmabhoomi Committees; VROs were to write up separate enquiry reports for each applicant, allowing higher levels of the mandal and district administration to make sound decisions. The vignettes detail how political and bureaucratic pressures interfered with these procedures. Pressures faced by lower bureaucracy have been detailed in other contexts as well, for instance field engineers in the Narmada dam projects (Aandahl, 2010) and local Irrigation Department functionaries in the Tungabhadra dam project in Karnataka (Mollinga, 2003). However, Janmabhoomi is distinctive in being a flagship programme of a governing party engaged in an extreme form of competitive populism, with the resultant pressures unleashed by it being that much greater. When passed through to lower bureaucracy, this extreme pressure translates to widespread subversion of Weberian procedures. A 2014 report in Andhra Pradesh by the Chief Rationing Officer submitted to the Commissioner of the Civil Supplies Department noted that in Hyderabad district the number of ration cards issued exceeded the population by almost 30 per cent (see Lasania, 2014). The study has sought to provide ethnographically informed vignettes of the underlying realities that produce such aggregate outcomes.

The vignettes also suggest the importance of understanding the lived realities of street-level bureaucrats—in this instance, mandal administration officials—akin to the emphasis placed in Lipsky’s (1980) classic work and in recent work for India (Aiyar & Bhattacharya, 2016; Ashwin, 2016). The master narrative that emerges out of the vignettes is not that of venality or petty bureaucratic corruption, but rather how implicit political pressures in a hierarchical system produce subversion of official procedures by street-level bureaucrats faced with impossibly difficult workloads. This also differs from Gupta’s (2012) emphasis on structural inequality driving bureaucratic practice (including seemingly arbitrary actions). While fieldwork did indeed throw up cases where some local elites successfully received ration cards despite their ineligibility, suggesting that the final approval process was not fully random, the master narrative that emerges is nevertheless about political pressures shaping bureaucratic practices. Interestingly, these political pressures from above (state political administration) are tied to a particular kind of partisan participatory democracy from below (village Janmabhoomi Committees), which can also be interpreted as a form of patronage politics.

Note that our argument—about competitive electoral politics manifested in a clientelist–populist manner creating political pressures on programme implementation bureaucracy—is focused on implementation rather than directly or explicitly on development outcomes. The fact of the matter is that ration cards do get distributed in enormous numbers (albeit without appropriate verification), and more generally, various welfare schemes do find their way to the voting public. Our vignettes, and overall argument, are consistent with positive development outcomes, although the implicit counterfactual is that such outcomes are likely to have been even better had political pressures for clientelist–populist delivery not produced subversion of (sound) implementation procedures. Nevertheless, as in the case of Tamil Nadu (Wyatt, 2013), populist pressures do pass through to the citizenry. Elliott (2016, p. 27) notes that competitive populism between the TDP and Congress in Andhra Pradesh has meant that

the question before political parties at election time has not been whether they would sustain the [rice] subsidy, but how much rice and what level of subsidy they would provide. The same dynamic has developed in promises for free power for farmers, initially offered by Congress in 1999. By 2009, this had become an expected benefit, and the competition was only over how many hours of free electricity would be provided.

We conclude with a puzzling implication of our observations. Widespread subversion of official procedures, due to the impossible pressures faced by lower bureaucracy, must be known to those who frame these procedures (both higher bureaucracy and the politicians to whom they answer). According to Powis (2012, p. 123), such ‘significant scope for distortion in the process of implementation … was part of the design’. Extant literature suggests that this is partly to generate opportunities for rent-seeking and corruption, but partly also because populist programmes need Weberian garb for regime legitimacy. According to Mooij (2007, p. 335), in this double-speak (i.e., claiming one thing but doing another), ‘it is the gesture of reaching out that may be enough, rather than the outcomes in terms of social development, welfare or expenditure levels’. However, extant literature does not sufficiently explain, for instance, why pressures around Weberian procedures for bureaucratic verification and approval of ration card applications—the subject of our vignettes—need have occurred when the top bureaucrat in the agency distributing rations (the Chair of the State Civil Supplies Corporation) has said: ‘Everyone who has applied for a ration card would get it. The State government wants to see everyone in the State has a ration card…’ (The Hindu, 10 February 2016). This paradox calls for exploration and explanation in future research.