ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Middle Game in Coalition Politics

Formation and termination of coalition governments is not the end all of coalition politics. This article focuses on the post-formation dynamics of coalition governments in India at the centre. Drawing from organisational learning and institutional studies, the article argues that coalition experiments are learning entities. Learning leads to institutionalisation of relationships. Institutionalised relationships, the article concludes, help construct a shared world and fulfil a common vision.

Middle Game in Coalition Politics

Formation and termination of coalition governments is not the end all of coalition politics. This article focuses on the post-formation dynamics of coalition governments in India at the centre. Drawing from organisational learning and institutional studies, the article argues that coalition experiments are learning entities. Learning leads to institutionalisation of relationships. Institutionalised relationships, the article concludes, help construct a shared world and fulfil a common vision.


oalitions form, coalitions terminate. Formation and termination are easily recalled as most of the sound and fury of coalition politics occurs during these phases. Coalition politics in India, especially at the centre, has not been received well. One reason could be the excessive attention devoted to formation and termination. This consequently has left other aspects of coalition politics inadequately explored. Our focus is on the post-formation dynamics of coalition governments in India at the centre.

Drawing from organisational learning and institutional studies the paper argues that coalition experiments are learning entities. Coalitions as a form of governance develop gradually, with each experiment learning from the other. Learning leads to institutionalisation of relationships. The greater the degree of institutionalisation, the more complex the mechanisms become. Complexity leads to specialisation. Specialisation builds robustness and resilience, which in turn equips the coalition to better handle the pulls and pressures and the twists and turns. Institutionalised relationships we conclude help construct a shared world and fulfil a common vision.

Our purpose in this paper is twofold. One, to bring attention to the much neglected post-formation pre-termination phase of coalition governments and two, to show that coalition management is the key to sustainability and longevity. This paper is in three sections. The first section focuses on learning and coalition management mechanisms. The second section traces the institutionalisation of relationships. Summing up, the final section argues that institutions of coordination are integrating, expression-enabling and participation devices, enabling the development of cooperation and commonality, which in the final analysis is responsible for smooth maintenance.

I The Middle Game: Coalition in Action

Formation and termination are not the only elements of the absorbing game of coalitions. Although the action in between the two ends is an object of intense public attention and scrutiny, it has rarely received systematic and informed attention. Concentrating primarily on the two ends, both coalition studies literature and research on the Indian experience have by and large bypassed the middle game.1

What is the middle game about? Coalitions are the coming together of originally distinct elements for the purpose of joint action, usually limited to specific purposes. It is based on the recognition that their common goals are better served through joint rather than individual action. The middle game revolves around attempts to maintain unity in the face of the demands of diversity. It is about acting in concert while taking into account differential needs. The key to the balance between unity and diversity is coordination. Three specific tasks of the middle game which accentuate the need for coordination include (a) issues and concerns unforeseen at the time of formation, (b) managing relationships, and (c) running the government. Let us look at these a little more closely.

Tackling Unpredictability

In coalition studies, events or stochastic theorists were among the first who underscored the need to look beyond formation.2 Theories that concentrate on formation they argued do not automatically explain the continuance of coalitions beyond formation. Stability and termination would be better explained if we move away from formation and recognise that conditions change over the life of a cabinet. They noted that certain destabilising or critical “events” from the environment, independent of conditions existing at the time of formation could have an impact on stability and existence of the coalition.3 These events could be of various types and come under different circumstances.

A coalition encounters and overcomes myriad issues, of both exogenous and endogenous nature. These include disagreements and sparring, unexpected pinpricks, posturing by partners, supporters and opponents, unforeseen issues and incidents; all or any of them could change the original terms and conditions, which created the coalition. Besides this, the functioning of modern governments involves multifarious activities, the effects and products of which cannot be anticipated. Moreover, in a federal system where electoral cycles of states vary, the nature of party competition, the issues, the results and so on have consequences across the federation. The distinctive characteristic of all “events” is their unpredictability and uncertainty. There is neither any regularity in their appearance nor are they under the control of players.4 The differential impact of these events on coalition members is potentially destabilising. The middle game is therefore about engaging with unpredictability and uncertainty and responding to the unexpected.

Building and Consolidating Relationships

On a more positive note, it involves enlarging the arena of commonality, as greater the common worldview, the lesser the scope for friction. It implies consolidating, holding together and building on shared objectives. It is about creating a shared identity, structure and organising enduring relationships. This institutionalisation of relationships may be done through the construction of a common framework of thought and action and creation of a repertoire of expected actions and/or roles along with sanctions for improper or inappropriate behaviour.5 This framework puts members on a common platform and assigns roles individually and collectively, the fulfilment of which enables realisation of the larger goals of the alliance.

Efficient governance may necessitate a dialogue with those outside the alliance. The more open and frequent the dialogue, the better the prospects of reaching an understanding and therefore greater the possibility of anticipating actions and reactions. Consequently, the middle game is about building relationships, consolidating commonalities, ironing out differences and broadening the participatory base of governance; all this not being restricted to alliance members alone.

Business of Government

Coalitions not only need to anticipate uncertainty and build relationships but also run a government. Decisions regarding ministerial composition and portfolio distribution are no longer the prime minister’s sole prerogative. How do executive centred parliamentary systems face the challenge of the “logic of departmentalism”?6 Coordination at the party level might prove inadequate to the task of running a government and could necessitate institutional innovations at the governmental level.

The coalition game does not end with formation. There is an intense phase post-formation that needs close examination. In real time politics however, there are no watertight compartments as “everything is, in principle, connected to everything in politics”.7 There is therefore, not only an overlapping between parts of the game but also a continuum between the past, present and future experiments. Parties in one alliance today or in the past could be part of another at a different point of time. These linkages are the carriers of history and an important part of the learning process. The next part of this section sketches the broad contours of the coordination mechanisms used in the middle game. Real time mechanisms are however deeply influenced by the environment and circumstances in which they are created.

Coalition Agreements and Coordination Committees

The two main coordination mechanisms that usually develop are coalition agreements and committees. They draw their inspiration from similar mechanisms used by political parties. Agreements take the place of ideology and associated policies and programmes of parties, while committees like the party organisation give the coalition a structure and lay down the basic rules of functioning. These institutional mechanisms give solidity, shape and direction to coalitions.

Agreements usually prepared after a consultation process serve three purposes. First, spell out policies, programmes and objectives. Second, lay down rules of the game or norms for interaction. Third, throw light on distribution of portfolios and “payoffs”. The three purposes have been listed in terms of their descending public visibility. The first purpose of agreements is the public face of the coalition in terms of its policies and goals and is usually available in the form of written documents. The second purpose, at one level reflects a certain degree of sophistication and institutionalisation of relationships. However, its primary thrust is internal. The third purpose reflects the opprobrious side, for which coalition politics has generally been damned.

Agreements enforce control directly by giving direction to the coalition, at the same time when they take on themselves a superior position in terms of being standards for justness and fairness they indirectly shape actions. The second and third purposes come to the public domain when used as moral sticks. For instance, “norms” are often used to bring recalcitrant allies into line. Or partners could display payoff agreements to show that the terms of the bargain are not being adhered to. In both cases the aim is to use perceived public pressure by vesting upon agreements a position of righteousness. All three purposes play a vital part in coalition management and effective governance.

As policy indicators, agreements spell out goals, purposes, priorities and costs involved in the achievement of the aims. They act as reference points not only for partners and supporters but also for the general public. It serves an internal function of binding the alliance around a common framework of action. At the same time it also has an external utility of informing the public of its objectives and agenda for action. At another level, agreements not only propose, shape and direct action but also subsequently endorse and legitimise action.

As rules of the game agreements lay down norms regulating the behaviour of members. It is primarily meant to check opportunistic behaviour and urge to push for short-term and limited gains. This code is usually taken for granted as it is assumed that members have a long-term vision and would not destabilise the arrangement in the short run. However, it could get codified when short sightedness creeps in and the danger of not sticking together increases. Uncodified rules are self-enforcing, however when codified, they are usually enforced by certain specialised agencies.

Agreements for Pay-offs

Dividing “spoils”, which fulfils the function of filling offices is generally the most private part of coalition agreements. In federal states, payoffs are not restricted to one level and could involve side payments at other levels. Evidences of such agreements are generally visible only when breached or there is a public display of unhappiness while bargaining.

Agreements as management mechanisms, spell out long-term goals, fulfil office and/or policy goals and give direction to the coalition. Agreements however, lack the dynamism to take on changing political circumstances. While they may factor in future contingencies they are ill-equipped to deal with unpredictability. Coordination committees fill in for this drawback and provide for real time management. While agreements are useful in macromanagement and bring solidity, permanence and order, committees help in micro-management and allow for adaptability and flexibility.

Coordination committees have generally served two purposes. One, function as safety valves, and two, as institutions to sustain togetherness. Committees provide a platform for regular interaction and also help take care of events. The middle game requires flexibility and hence there could be a variety of mechanisms, permanent or ad hoc, general or specific, intra-coalition or intercoalition, governmental or party level, national or state level and many more, depending on the needs of the coalition. Committees play a vital role, while on one hand they are able to nurture a shared vision; on the other hand, they explicitly recognise the contribution of the supporters. Support must not be taken for granted and needs visible recognition by keeping the stakeholders engaged.

As safety valves, committees give partners an opportunity to raise, criticise, deliberate and reach compromises away from public glare. This is crucial because a coalition is not union where they cohere to sink their differences. Parties represent specific interests and groups and continue to maintain their distinct identities within the coalition. Events consequently, have different meanings and effects leading to different reactions. This forum enables an exchange of information between the actors involved and if required a renegotiation of roles and responsibilities in the light of new developments. It acts as a sounding board and constrains both dissent and bargaining within specific contours, thus attempting to avoid fratricidal conflicts.

Coordination committees serve as safety valves by giving partners the space to express their distinct identities. Alternatively they also help construct a common identity by acting as bringingtogether institutions. As an integrative mechanism, committees’ nurture shared concerns and build understanding and cohesiveness among coalescing members.8 This bringing-together role also enables recognition of individual efforts, roles and contributions that may not be visible in the larger joint effort. By creating this sense of belonging, it reduces the chances of alienation. This understanding enables them to work effectively and successfully towards common goals.

Logic of Appropriateness and Learning Process

Every domain according to the “logic of appropriateness” has its own particular rules and practices and they by and large represent lessons of learning from past experiences.9 This logic basically means rule based behaviour where a given set of rules, codes, procedures, roles, conventions, traditions and so on direct action, suggesting appropriate behaviour for situations. In coalition politics this logic we argue is codified in terms of agreements and coordination committees which assign actors particular roles and guide behaviour. Embedded in the logic is a normative element in terms of what is “good” or what fulfils common interest. This element dissuades selfish actions and directs members towards the achievement of common objectives putting aside individual labels. Furthermore in “novel” situations, the logic is handier as it would suggest actions “appropriate” to circumstances without jeopardising the alliance.

The logic and its subsequent codification are neither inherent nor are they instantaneous processes but are actually learnt and shaped. Our thesis is that coalitions learn to play the game. The logic in terms of agreements and committees develop as a consequence of learning. We postulate an evolutionary learning process in which there is an accretion of knowledge. From unstable and shaky beginnings, over a period of time coalitions develop mechanisms and best practices ideal for the game. This repertoire represents the development of new knowledge. This line of exploration enables us to avoid being both descriptive and functionalist. We leave the descriptive task to the media. If we explain the emergence of coordination mechanisms simply based on the function they serve, then we would be erring on the side of functionalism. Both these approaches cannot capture the essence of the middle game. A learning based explanation situates it within the larger game and may well also throw light on the vexing question of longevity.

Until now we have discussed the main issues of the middle game and postulated that coalitions are learning entities. But how does this learning take place? How do coalitions establish mechanisms of governance? How do these mechanisms evolve? The final part of this section attempts to answer these questions.

It is now beyond dispute that certain unobservable changes in the mental activity or cognitive processes accompany observable changes in behaviour.10 Cognitive science tells us that learning is not simply a changed pattern of behaviour or a response to stimuli (R-S phenomenon) but is a mental activity involving certain thoughts and strategies.11 At the core of cognitive science lies the assumption that the mind constructs associations based on its experiences, interpretations and information. These mental constructs form a reservoir of knowledge and constitute the basis of memory. They are not fixed, but are plastic in nature and learning is not simply a recall of previously stored information.12 Instead, learning is a result of the interaction between constructs in our mind and our new experiences and information. Learning is a change in mental knowledge, i e, it involves a transformation in mental associations. In the face of a new experience or information, the mind consults the mental construct and reviews its new experience with the old construct, thereby transforming the mental association no matter what.13In sum learning is guided more by past experience than by future anticipation.

In organisational learning literature, the equivalent of mental constructs or associations are routines. For organisational theorists March and Levitt, the term “routines” “includes the forms, rules, procedures, conventions, strategies, and technologies around which organisations are constructed and through which they operate. It also includes the structure of beliefs, frameworks, paradigms, codes, cultures and knowledge that buttress, elaborate and contradict the formal routines”.14 Routines tell us how and why to do certain things in a particular context. Further these constructs or routines are captured by history in a way that makes “the lessons, but not the history, accessible to organisations and organisational members who have not themselves experienced the history”.15 Neisser similarly observed that the mind stores not products but traces of earlier cognitive acts.16 This makes it possible for routines to achieve a degree of autonomy and survive a change in personnel making them transferable across time and space. Thus routines, which guide human action, become standard operating procedures and usher in predictability, essential for collective action. In the context of organisations, March and Levitt note that “routines are transmitted through socialisation, education, imitation, professionalisation, personnel movement, mergers, and acquisitions”.17 Given that in politics there are neither beginnings nor endings and that everything is connected to every other thing, conditions are fertile for the transmission of knowledge.

How does the above help us understand coalition politics? The question of learning has primarily been the domain of psychology, however social scientists have begun to use findings from this advancing domain to explain social phenomenon.18 Learning could be a valuable explanatory factor in coalition studies, when we move from simple episodic studies to longitudinal analyses of coalition experiments and from a fragmentary approach towards a study of the whole coalition game. From cognitive science we know that past experiences encoded in mind constructs are valuable markers for future action. When we argue that coalitions learn to play the middle game it is not a simple transformation from the initial amateurish attempts to more sophisticated techniques of coalition management. But rather this transformation involves a complicated learning process, a processing of knowledge that is building or constructing associations to fulfil the needs of the emergent new situations. The learning process evolves through multiple stages.

Coalition agreements and committees are basically constructs and routines of coalition governance. When coalitions formed in India for the first time after a long period of single party rule, mental constructs of coalition experiences were likely to have been absent. Under these conditions the coalition is more likely to draw first from constructs and routines that have been used most recently in similar situations.19 Furthermore, some of these constructs may be of the general type essential to the functioning of the system itself. Political parties for example, have wellorganised structures, rules for functioning, clearly established ideologies and concomitant policies and programmes. These features or routines are now not only characteristic of political parties but have in a way become part and parcel of the system. A coalition organically lacks these elements and the construction of some structure, rules of organisation and a common programme of action therefore follows. Besides existing constructs, coalition experiences elsewhere could also be instructive. For example, in a federal system coalition experiments in the units could become exemplars.

A series of coalition experiments over time enables the formation of a database of experiences. Every experiment contributes to the storehouse of knowledge or the system’s memory bank. This enables each experiment to learn not only from its own experiences but also from that of others. This is an evolutionary process of learning, where exploitation of knowledge from the bank and an exploration process go hand in hand.20 Exploitation basically means using knowledge that is readily available. Where as in exploration there is a search for new possibilities and this could take different forms. For instance, there could be refinements, reconstruction, experimentation and improvisation of old techniques, besides construction and creation of new mechanisms. Both exploitation and exploration aid in the development of the repertoire of routines. With each new experience and information, the constructs in the database are transformed.

Evolutionary learning is nevertheless is not necessarily efficient, i e, progressing to a higher or superior stage. The intention may be to move to better coordination and management mechanisms but this does not always happen. There are problems of superstitious learning, improper recording of experiences, competency traps, faulty interpretation paradigms and frameworks, which may not result in “superior” mechanisms.21 The next section analyses the coalition experiments at the centre since 1989 in a comparative framework with special reference to the evolution of management mechanisms.

II Coalitions and Mechanisms

The National Front (NF), the first in the series of coalitions in the competitive multiparty era, neither had any formal coordination mechanism nor any common programme.22 The Left Front and the Bharatiya Janata Party (BJP) were supporting the alliance on the Janata Dal (JD) manifesto. This was obviously not conducive to the development of a common worldview and consequently among others, the friction between the twin goals of Mandal and Mandir led to the break up. The Front however did have some cabinet panels but the government did not last long to tell us how these mechanisms functioned.23

After the eleventh Lok Sabha elections, when the United Front (UF) was being formed, V.P. Singh, who headed the first federal coalition in 1989, was asked at a press conference as to what was needed to ensure the durability of a coalition. He laid down three factors, one, there was need to consolidate solidarity within the third Front (used to refer to the group of parties who sought to keep themselves distinct from the two polity wide parties, the Congress and the BJP) for this they had to devise modes of cooperation among various partners; two, there had to be an agreement on policy parameters among the coalition constituents and its supporters from outside; and three, coping with areas of political conflict at the state level. He also stressed the importance of setting up “contact groups” between the Front and the Congress. This message was clearly the wisdom of experience and lessons learnt. The NF experiment was singularly lacking on all three counts. The UF quickly moved on to establish a three-tier coordination mechanism. Though there was a leadership change in the UF during its term, in this paper the two experiments are clubbed together.24

Common Minimum Programme

At the first tier of the UF management mechanism was a programme spelling out its goals. This common minimum programme (CMP) outlining the social, economic and political agenda of the government was drafted by a subcommittee of the Front with an aim to narrow down differences and synchronise ideas and was supposedly based on a synthesis of the constituents’ manifestos and a Congress document.25 The UF, as one its senior ministers, remarked was committed to “reconciling the differences and enlarging the areas of consensus and the CMP was its modus vivendi”.26 The CMP attempted to achieve the highest common factor on which there was commonality and consensus among both partners and supporters.

Steering Committee

At the second level was the Steering Committee (SC), which included parties not only within the governing coalition but also those outside it (but not all). This complex mechanism was developed by “exploiting” knowledge of coordination experiments in Kerala.27 No agreement however, was reached on a similar arrangement between the Congress and the UF.28 The SC was to be the voice of the UF and a forum for consultation and discussion. Along the way the UF I set up a Core Committee or Standing Committee of the Steering Committee (SCSC), ostensibly for the purpose of better coordination. In practice it duplicated the work of the SC, diluted its importance and soon emerged as a big brother to the SC.29

The Prime Minister and the Main Supporting Party

The third mechanism of the UF, direct interaction between the prime minister and the Congress was heavily dependent on the personal relationship and goodwill between two individuals. P V Narasimha Rao and H D Deva Gowda gelled well, but with a leadership change in the Congress, the relations also underwent a change. Sitaram Kesri, the new Congress president and Gowda were uncomfortable with each other from the start.30 Notwithstanding the leadership change, Gowda continued to meet Rao and it was after one such meeting that a furious Kesri pulled the plug.31 This uninstitutionalised mechanism was the weakest link of the UF coordination chain. At the time of the formation of the second UF government, the president reflecting on the fragility of personal relations suggested the establishment of an institutionalised coordination mechanism between the Congress and the UF.32 This advice went unheeded and personal relations continued to play a vital role.33

The UF mechanisms were top-heavy and ill suited to the needs of a federal polity.34 Another reason for souring of relations besides the question of personal equations was the UF refusal to support the Bahujan Samaj Party (BSP), a Congress electoral ally in Uttar Pradesh. The UF was constrained by the Samajwadi Party (SP) from acceding to the Congress demand.35 This noncooperation not only broke the Congress-BSP alliance but also enabled the BJP to subsequently form the government in Uttar Pradesh.

The concentration of power caused more problems for the Congress than the UF making Kesri’s short reign a challenging one.36The Congress was in completely unfamiliar territory, being out of power and supporting a government from outside. There were numerous conflicting views within the party, which included among others, joining the government, heading the government and even forming “secular” government with support from the UF constituents. As subsequent actions testify, the withdrawal of support by the Congress was more a result of an internal power struggle rather than any major disagreement with the UF.37

National Democratic Alliance: Strength-to-Strength38

For the purpose of this discussion, the BJP plus allies experiment of 1998-99 and the National Democratic Alliance (NDA) experiment that followed are clubbed as NDA. The NDA learnt quickly and grew from strength to strength. Exploring and exploiting available mechanisms it used multiple coordination agencies stretching from the political to the governmental sphere.

The National Agenda for Governance (NAG) like the CMP formed the foundation. It committed the coalition maker, the BJP to adhere to a moderate agenda and to eschew its religious and thereby essentially divisive programme. This assurance removed the “untouchability” tag, allowing formation of the coalition. The NAG was an integral part of the daily political discourse and a crucial stabiliser unlike the CMP, which appeared to have been ignored after its publication.

At the second level, the NDA had a Coordination Committee (CC), which met more frequently and regularly than the UF mechanism.39 Notably wherever possible, it extended to the state level in which the dominant partner in the state donned the leadership mantle. Inter-party matters handled there did not find a place in the national level agenda. As a conflict resolution device it cauterised conflicts and enabled disposal of contentious issues without holding the coalition to ransom.40 This split coordination pattern, a reflection of the federal reality enabled a degree of flexibility as it could take into account the heterogeneity and diversity, increased consultation avenues and helped maintain a balance of power between partners.

Coalition Dharma: Code of Conduct

Exploring and refining mechanisms, the NDA laid down a framework, “tenets of a dharma of coalition” basically a call for appropriate behaviour in a coalitional situation. It was to enable resolution of differences between partners and regulate its internal functioning.41 This declaration was intended to strengthen the government by checking actions which were tarnishing the alliance and also bring “internal cohesiveness and singularity of purpose” in its functioning.

Notwithstanding the dharma doctrine, continuous carping and sniping by allies and fraternal organisations of the BJP did little to enhance the prestige of the ruling alliance or confidence in coalitions in general. After a particularly severe bout of indiscipline in 2001 when the prime minister was targeted, a consensus emerged in favour of laying down a “code of conduct”.42 While the dharma doctrine was self-regulatory the new code was to be enforceable. It was resolved “to formulate a code of conduct applicable to all to ensure adherence to the canons of coalition politics”.43 This was also a reaction to the possible re-entry of some of the allies who had deserted the coalition during the assembly elections.44 Subsequently it was decided that there would be a cool off period between rejoining and re-induction into the ministry.

The NDA also extensively used both the all-party meeting, one of the oldest parliamentary devices, and the ad hoc executive federalism mechanism and intergovernmental agency, the chief ministers’ conference (CMC) for coordination. As a coordination device, all party meetings brought together political parties on a common platform and allowed taking into account the concerns of not only partners but the entire political spectrum represented in and sometimes outside parliament. The CMC involved states in national decision-making.

These institutions served several purposes. First, it enabled consultation with the external support partners who were part of the legislative coalition but did not take part in CC meetings. Second, the lack of requisite numbers in the Rajya Sabha necessitated the use of the bipartisan route more frequently. Finally, it also allowed the alliance to move away from tight coalitional constraints and get a wider backing as and when required.45

The NDA machinery was not limited to the political level but also extended to the governmental level. The traditional essence of the cabinet being a small inner circle of government ministers lost meaning. It was now a heterogeneous body whose members sat at the cabinet table not at the invitation of the prime minister but as a matter of right. The decline of the principle of collective responsibility and immobilism has been a major lament with the coming of coalitions. The NDA attempted to tackle this problem innovatively.

It used inter-ministerial groups in addition to the council of ministers for decision-making at the governmental level. These groups popularly called group of ministers (GoM) usually composed of three to four members of the cabinet.46 As a coordination mechanism GoMs helped minimise differences of opinion and conflicts within the cabinet. They were also used for making policy decisions or vetting policy and programme recommendations. Additionally they were used to look into matters of special interest to the allies.47 The GoM functioned at the governmental level but took care of the party-government interface.

The NDA like the UF managed its relations with its outside support partners through direct relations between the top echelons. The NDA therefore not only made use of previous experiences but also introduced newer institutions into the repertoire of coordination mechanisms.

United Progressive Alliance: Towards Autonomous Institutions

The United Progressive Alliance (UPA) since May 2004 has put in place a complex web of mechanisms spanning the political and governmental level attempting to plug the weak links of previous experiments. Though it has yet to utilise mechanisms like all party meetings and chief ministers’ conferences in a substantial way, it has tweaked the GoM mechanism to make it an instrument for both coordination and control.

The basic management tool of the UPA is the CMP, prepared in consultation with partners after a study of their manifestos.48 The Left parties were shown drafts of the CMP and their suggestions incorporated.49 While the Left broadly agreed with the CMP, it did not endorse the economic policies and the position regarding the formation of new states. It was agreed that they would present alternate economic policies while continuing to support the government. The agreement to disagree is refreshingly mature in so far as it helped remove unpredictability, enabling partners to estimate actions and reactions in advance.

The cabinet endorsement of the CMP as the national common minimum programme (NCMP) transformed it to a national commitment. Subsequently, ministries have been directed to prepare roadmaps along with legislative proposals for its effective and speedy implementation. The catch however was that wherever appropriate, this preparation was to be in consultation with the planning commission, finance and the law and justice ministry.50 While it may bring about greater inter-ministerial coordination it also facilitates new forms of control, as these departments are under the coalition maker.

The UPA instituted a power sharing formula for portfolio distribution. It was supposedly agreed to that for every three representatives, a party would be allocated one ministerial berth. This agreement signalled a move to establish some principles for portfolio allocation. The second step of this agreement was the actual distribution of offices among allies. As noted earlier such agreements if even they existed are not usually publicly discussed or displayed but only used as moral sticks. When the Dravida Munnetra Kazhagam (DMK) did not get the berths that it had been promised, it made the existence of such an agreement public.51

The bringing-together institution for the alliance is the UPA coordination committee (UPACC) which provides a platform for consultation and discussion. It primarily comprises of government partners and some who have declined to join the government. However, the Left and other supporting parties have also been invited for discussion and consultation to this forum. This forum was stated to meet every month.

Continuing with the NDA tradition, the UPA has already set up more than 60 GoMs to deal with multiple issues ranging from the Dabhol power project and restructuring of Delhi and Mumbai airports to affirmative action and 350th year of Taj Mahal celebrations. Care has been taken to ensure representation of different alliance partners based on their specific concerns.52 The staffing pattern of the GoM reveals that it has been used not only for coordination but also as an instrument of control over ministerial functioning.53Most GoMs are headed by the coalition maker thus keeping a check on its direction.

There are two levels of mechanisms between the UPA and the main supporting partners. First there is a Left Front coordination committee which brings together the four Left parties, enabling them to “interact” and iron out differences while ensuring the commonalities are achieved. It was to act as a “sentinel” keeping a close watch on the implementation of the NCMP and ensure that despite “differences” with the UPA the government does not fall.54

At the second level, there is the UPA Government-Left Coordination Committee (ULCC) which links the government and the Left front. All previous experiments lacked an institutionalised coordination mechanism between the executive coalition and non-governing partners, relying instead on the personality oriented ad hoc mechanism of leader-to-leader interaction. The ULCC reluctantly set up at the insistence of the Left parties, is therefore an important development.

Exploring further, the UPA set up a national advisory committee (NAC).55It was entrusted with two main tasks, (a) to advise the government on implementation of the NCMP and (b) to provide input for the formulation of policies by the government. The NAC as constituted by the government included academics, economists and representatives of voluntary organisations among others. It is meant to act as an independent advisory body to the government and its recommendations were to be subject to the normal scrutiny and processing of government. The government however, is not bound by its decisions.

The role of the Left as supporting partners has been instrumental in the UPA adding new routines to the coalition bank, systematising their functioning and attempting to make them more meaningful.

III Coalition Management: Institutionalisation of Learning

Our study of a neglected feature of coalition politics, the middle game reveals that the coalition game is not composed of discrete events but is in fact a continuous process. Parties consider past history and learn both from their past and others experiences at multiple levels. At the same time they also anticipate future questions and problems.56 In this connection we saw that institutionalised coordination mechanisms have played a crucial role in bringing stability and resolving the problem of maintaining unity and diversity.

The distinct characteristic of the middle game in India has been the gradual institutionalisation of learning, indicating a transfer of the lessons of history. From merely seat adjustments and no agreements, the repository now includes electoral alliances, common agendas, all party meetings, chief ministers’ conferences, multiple level coordination committees, inter-ministerial groups and even agencies outside the sphere of conventional politics. Committees have acted as safety valves and as bringingtogether channels. Agreements have served as both a gel and a torch. The NAG and now the NCMP went beyond the purpose of internal utility and have become public reference points. It would be fair to conclude that federal coalitions through the processes of exploitation and exploration evolved newer and sturdier governance mechanisms.

In this evolution, it is possible to discern three generations of coalition management mechanisms. The NF had no machinery worth its name and was consequently completely unprepared. Actions unleashed without systematic consultation and discussion led to cross connections between individual, party and alliance goals. Defeating the Congress, capturing power and the subsequent ecstasy drowned the need for further consolidation. It did not attempt to build bridges, secure similarities, dilute differences or even agree to disagree. Though the alliance was aware of the need for a common programme it only paid lip service to it. This exploitation following the organisational learning pattern indicates that in novel situations, in this case the formation of a coalition government after years of single party rule, routines used most recently and frequently find favour. However, the alliance floundered in the absence of an organised and systematic organisation. Nevertheless a learning process had been set in motion.

V P Singh’s four-point advice to the UF in 1996 is testimony to that learning. Singh recognised that seat adjustments and coalition formation was not the game but a part of it. He identified the problematic areas and mechanisms needed for skilful management. Compared to the UPA’s complex array of institutional mechanisms, the UF’s efforts were raw. The UF however established the first generation of management mechanisms. Two of its mechanisms, common programme and coordination committee are now minimum requirements. Its collegiate style of drawing up its programme has been particularly instructional to coalition experiments that followed. The UF itself exploited the storehouse of routines provided by the Kerala experiments to set up its coordination committee and the steering committee of the coordination committee.

Learning from previous experiences every subsequent alliance attempted to avoid past mistakes.57 One of the major challenges for federal coalitions is to remain alive to local particularities and challenges while taking advantage of the federalised competition pattern. Federal-state, state-state stress at the intergovernmental level could cause enormous coordination problems. Notwithstanding an increasingly federalised party system, activities in the same geographic area could lead to intracoalitional conflicts. Consequently, there has been a need felt for clear-cut division of responsibilities not merely to avoid conflict but to also avoid duplication of efforts. Standard and fixed governance mechanisms have proven impractical. Instead units could be allowed to adapt mechanisms to their local needs.

The NDA ushered in the second generation of mechanisms. Exploring new devices, it took coordination to the state level. Coordination committees no longer served merely as safety valves but attempted to transform partners into stakeholders. The NDA’s distinct contribution to the lexicon of coalition politics has been the GoM mechanism. It also used the intergovernmental mechanism of chief ministers conference as a coordination device. Furthermore, its use of all-party meetings recognised that the middle game required efforts not only within the coalition but also across the political spectrum. This instrument gave the alliance the “room to manoeuvre” and seek a broader support base. Through its code of conduct it attempted to lay down the parameters of appropriate behaviour, bring in elements of justice and fairness in actions and ensure greater predictability and unity in action.

Constant refinement has led to increasing complexity and specialisation. While the UF relied more on exploitation, the NDA was prepared to experiment and try out new devices. Moving on the innovation path, the UPA has created a multidimensional web of coordination devices. Rather than a single bring-all-together body, it has specialised agencies. While the central coordination agency, the UPACC deals with the UPA as a whole; the ULCC has been designed to address the specific concerns of the Left Front. Having been involved as part players in some of the previous experiments, the Left and the Congress were aware of the dangers of leaving coordination to individuals and have established collegiate structures of synchronisation. The ULCC is emblematic of the relentless institutionalisation that has taken place. The Left has its own committee for management of relations within the Front. This has not only enabled it to put up a united face and but has also increased its bargaining power vis-à-vis the UPA. Another UPA contribution has been the formula for the distribution of offices. By establishing clear-cut norms it sought to regulate bargaining and avoid post formation tussles regarding office.

From learning, there was institutionalisation and then multiplication, sophistication and specialisation followed. Institutional mechanisms were initially valued simply for their role in managing the alliance. Over a period of time, mechanisms have also obtained an identity of their own, beyond the instrumental value. The establishment of the NAC ostensibly for inputs from civil society is representative of this movement. It has a certain degree of autonomy not available to the other coordination tools. This insulation and autonomy enables it to provide a vision and a direction that looks beyond mere survival.58 Its composition and focus on the common goals has also given it a degree of credibility that other coordination mechanisms may not enjoy. Its establishment reflects the high level of maturity coalition management has reached in a little more than a decade.

While newer and sturdier coalition management mechanisms have been established, the learning process does not necessarily result in inefficient institutional mechanisms giving way to superior ones. The NDA’s reliance on personal contact with allies providing outside support and the UPA’s avoidance of fullfledged state level coordination are examples of inefficient mechanisms continuing despite a learning process. It may be concluded that the nature of the actors has a decisive impact on the establishment of coordination mechanisms. The Congress had been the principal opponent to all until the breakdown of its dominance. Despite the breakdown, it continues to be a critical player in many states making it difficult for it to enter into arrangements at the state level. The BJP could establish state level mechanisms as it was secondary player in many states and hence was not seen a threat.

How do we evaluate the role of institutionalised governance mechanisms? Focusing merely on whether they do or do not prevent termination is a narrow and inadequate measure. The middle game is about maintaining a balance between unity and diversity and here it is the processes involved that make a difference. Agreements and committees should be valued not for the results they produce but for the processes they initiate. Institutionalised management sets in motion a whole process of consultation, dialogue and deliberation. The characteristic feature of which is the bringing-together of different partners. This process creates an atmosphere of togetherness and builds confidence among partners that they are being involved and are active participants in the game.

Coordination mechanisms prevent dominant partners from riding roughshod over the smaller partners. The coalition maker would be acutely aware that all decisions would be intensely scrutinised in the coordination committee. The partners are also in the know that they can voice their suggestions/opinions. Participation in terms of actually making a contribution finally is secondary. What counts is the recognition of the right to participate and the right to be consulted in the decision-making process. The web of management institutions and the processes involved communicate the message, that the coalition game is a collective affair.

Reduction of subsidies for instance, has been an issue that has stuck in the craw of all coalition regimes. All coalitions have faced dissent on the question of reducing or phasing out subsidies in the oil and gas sector. The script goes something like this. A price hike raises an instant reaction. There may be series of committee meetings or consultation rounds before a price hike. Another round of meetings and consultations follow the hike and the government would come up with either a partial rollback or promises one in the future. These numerous rounds of consultation and meetings, if we focus merely on the outcome would appear to be a charade to delay the inevitable. However, as institutional studies have shown us this process or the “drama of decision-making” is crucial.59

The role of the UF’s steering committee (SC) highlights the “drama” vividly. In this drama it emerges as a hero in its role as a safety valve and as a bringing-together institution. Many of the UF constituents had gone to the elections on a reform rollback platform but the government chose to build a consensus in favour of reforms. The success of reforms critically hinged on the reduction of fiscal deficit. Though cutting subsidies was a big task, the UF covered good ground. The SC provided a forum for dialogue. A study of the news reports of coordination meets showed that considerable amount of time was expended on the issue of subsidies. Subsequently the government came out with novel solutions, the targeted public distribution system and oil bonds, to check the deficit and lessen the pain on consumers while moving towards a rational pricing scheme.60 Though they might not have been optimal solutions that satisfied all, the process by which the programme was conceptualised and implemented was creditable.

The analogy of a drama perhaps most effectively highlights the nature of decision-making in a coalition situation. A plot takes numerous twists and turns before the final curtain call. These twists and turns ensure all actors a moment of glory and also indulge the audience before the end resolution. Similarly, there are often long drawn out episodes when it appears that all is lost, when the unpredictable happens. These episodes provide actors roles or parts to play. Even a decision by some actors not to engage or play a role is actually part of the design. A “wooing” back process follows this act.61 It is all this that makes the coalition plot intense, gripping and dramatic. From the point of view of coalition management this drama plays a vital role. It enables what we called as the building of confidence among the partners that they are part of the game. It informs partners that their voices are important inputs into decision-making and they are being heard. It also demonstrates that the final decisions are outputs of the whole consultation process rather than a one-sided proclamation by the coalition maker.

A refrain common to all coalitions has been the demand by the partners and supporters that they should be consulted or that they have not been consulted on such and such issue. This demand when juxtaposed with the subsidy example above shows that the process is the key to relationship management. The issue of subsidies was highlighted because it happens to be the most popular drama in town. The drama analogy may be used in other areas as well. The point is that the motions preceding any decision by the coalition are critical to relation management. The process should be or at least appear to be inclusive rather than exclusive, one that brings together partners, enabling them to share their concerns, allowing them not only to speak but also to be heard. It should transform them from mere allies into stakeholders. The process is therefore crucial to the maintenance of the balance between unity and diversity.

Political culture and ideology may be inadequate cohesion material in the face of weak institutionalised coordination mechanisms. Consequently they are less than adequate explanatory factors for durability and longevity. Drawing upon the concrete experience in India this paper has postulated an evolutionary learning process in which there is a slow institutionalisation of coordination machinery. As parties begin to work together and learn the “art of associating together”, to use Tocqueville’s phrase, they develop arenas for interaction. Over a period of time these arenas solidify into robust institutions, which then develop into complex systems. It is this machinery in which the logic of appropriateness is embedded that enables the coming-together, builds trust and gives the coalition the strength to coexist in the face of pressures of diversity.

Coalitions that have not put in place an institutionalised framework for coordination have floundered in the face of individual whims and transient situational demands. The difference between weak and unstable as opposed to strong and stable is the degree of institutionalisation. While institutionalised mechanisms are insurance policies against uncertainty and unpredictability, their real utility lies in their role as bringing-together institutions. Coordination mechanisms help in the creation of a shared world and common understanding by ironing out differences and building on the commonalities and it is this common understanding that ensures that the coalition functions efficiently and successfully. The art of coalition politics therefore lies in the middle game.




[I would like to thank Balveer Arora, Jawaharlal Nehru University, Peter Ronald deSouza, Senior Fellow, CSDS, E Sridharan, Academic Director, UPIASI and Ashutosh Kumar, Panjab University for their advice and helpful comments on an earlier version of this paper. I also thank an anonymous referee whose comments helped substantially improve the paper. Nonetheless all shortcomings that remain are mine alone.]

1 There is considerably rich literature on formation and termination. See for example, William Riker, The Theory of Political Coalitions, Yale University Press, New Haven, 1962; Abram de Swaan, Coalition Theories and Government Formation,Elsevier, Amsterdam, 1973; Michael Laver and Normal Schofield, Multiparty Government: The Politics of Coalition in Europe, Oxford University Press, Oxford, [1990] 1998; Michael Laver and Ian Budge (eds), Party, Policy and Government Coalitions, St Martin’s Press, London, 1992; Paul Warwick, Government Survival in Parliamentary Democracies, Cambridge University Press, Cambridge, 1994; Michael Laver and Kenneth A Shepsle, Making and Breaking Governments: Cabinets and Legislatures in Parliamentary Democracies, Cambridge University Press, Cambridge, 1996. Literature on the Indian experience include, Subash Kashyap, Coalition Government and Politics in India, Uppal, New Delhi, 1997; E Sridharan, ‘Principles, Power and Coalition Politics in India: Lessons from Theory, Comparison and Recent History’ in D D Khanna and Gert W Kueck (eds), Principles, Power and Politics, Macmillan, New Delhi, 1999, pp 270-90; Balveer Arora, ‘Negotiating Differences: Federal Coalitions and National Cohesion’ in Francine R Frankel et al (eds), Transforming India: Social and Political Dynamics of Democracy, Oxford, New Delhi, 2002, pp 176-206; Paranjoy Guha Thakurta and Shankar Raghuraman, A Time of Coalitions, Sage, New Delhi, 2004; M P Singh and Anil Mishra (eds), Coalition Politics in India: Problems and Prospects, Manohar, New Delhi, 2004; Bidyut Chakrabarthy, Forging Power: Coalition Politics in India, Oxford, New Delhi, 2006.

2 Eric C Browne, John P Frendreis and Dennis W Gleiber, ‘An “Events” Approach to the Problem of Cabinet Stability’, Comparative Political Studies, 17(2), July 1984:167-97.

3 Ibid, p 180. 4 Ibid. 5 We take a “normative” view of institutions by March and Olsen. Guy Peters

uses this term to differentiate the March and Olsen thesis from other brands of new institutionalism. For different versions see B Guy Peters, Institutional Theory in Political Science: The ‘New Institutionalism’, Pinter, London, 1999.

6 Michael Laver and Kenneth A Shepsle, Making and Breaking Governments: Cabinets and Legislatures in Parliamentary Democracies, Cambridge University Press, Cambridge [1996], 1999, p 13. They argue that those who head and control a department/ministry have a significant role in directing the decision-making and implementing process. The logic accords considerable discretion to individual ministers to act independently of other ministers and also reduces the scope of constraints that could come in the way of their functioning.

7 James G March and Johan P Olsen, Rediscovering Institutions: The

Organisational Basis of Politics, Free Press, New York, 1989, p 26. 8 For institutions as integrating mechanisms see ibid, pp 118-42. 9 Ibid, pp 23-24.

10 Neisser used the term cognition, to refer to “all the processes by which the sensory input is transformed, reduced, elaborated, stored, recovered, and used. It is concerned with these processes even when they operate in the absence of relevant simulation, as in images and hallucinations. Such terms as sensation, perception, imagery, retention, recall, problemsolving, and thinking, among others, refer to hypothetical stages or aspects of cognition.” See Ulric Neisser, ‘Cognitive Psychology’ in Margaret P Munger (ed), The History of Psychology: Fundamental Questions, Oxford, New York, 2003, p 448.

11 Cognitive neuroscience has shown with the help of advanced imaging techniques, that specific parts of the brain are used for particular functions. Among other things it also showed that the neural networks within the brain function differently when confronted with a new situation as compared to dealing with familiar objects, leading to a time lag in the observable behaviour. The functioning of neural mechanisms and time lag is evidence to the mental activity. See Michael I Posner, ‘Current Research in Cognitive Neuroscience’ in Indramani Singh and Raja Parasuraman, Human Cognition: A Multidisciplinary Perspective, Sage, New Delhi, 1998, pp 41-54.

12 Cognitive science goes beyond what Neisser called the “reappearance hypothesis” which postulates that the mental processes do not create something new but only remember something that already exists. Ulric Neisser, 2003, op cit, p 453.

13 Neisser called this the “utilisation hypothesis” which holds that the mind does not simply recall objects or responses but there is an active process of reconstruction involved, ibid, p 455.

14 Barbara Levitt and James G March, ‘Organisational Learning’, Annual Review of Sociology, 1988, p 320.

15 Ibid.

16 Neisser, 2003, op cit, p 455.

17 Barbara Levitt and James G March, 1988, op cit, p 320.

18 C Mantzavinos, Douglas C North and Syed Shariq, ‘Learning, Institutions, and Economic Performance’, Perspectives on Politics, 2:1, March 2004, pp 75-85.

19 Barbara Levitt and James G March, 1988, op cit, p 328.

20 James G March, ‘Exploration and Exploitation in Organisational Learning’ in James G March (ed), The Pursuit of Organisational Intelligence, Blackwell, Malden, Massachusetts, 1999, pp 114-36.

21 Barbara Levitt and James G March, 1988, op cit, pp 322-28.

22 V P Singh in an interview to N Ram and Sukumar Muralidharan, in 1996 noted that the JD was the dominant partner in relation to the regional parties in 1989 and “though consultation took place, it was more out of courtesy than because of the clout of the regional parties”. Frontline (Chennai), 13(12) June 15-28, 1996. In another interview to Shekar Gupta in 2005 he said that the Left and the BJP met in the prime minister’s residence every Tuesday over dinner. The Indian Express, New Delhi, July 5, 2005.

23 These included panels on price rise, panchayati raj, autonomy to the electronic media, right to work, Punjab and Jammu and Kashmir and the right to information. The Hindu, Chennai, December 8, 1989.

24 This is not to deny that the leadership change was insignificant. UF I and UF II were headed by H D Deva Gowda, June 1, 1996 to April 21, 1997 and I K Gujral, April 21, 1997 to March 19, 1998, respectively.

25 Its formulation began almost immediately after the third front re-christened itself as the UF and elected Deva Gowda as its leader. The members of the drafting committee included P Chidambaram (Tamil Manila Congress), Sitaram Yechury (CPM), D Raja (CPI), and Jaipal Reddy (JD). The Congress also came out with a comprehensive note on the framework within which it expected the UF to function. It did not lay down any specific conditions but said it supports policies on which there is a “broad national consensus”. Secondly, it also hoped there would be a continuity “of such policies which have acquired the imprimatur of history”, The Hindu, Chennai, June 2, 1996.

26 P Chidambaram’s statement in The Hindu, Chennai, June 3, 1996. V P Singh speaking on the CMP said “it has to be a sort of averaging. And what, after all is consensus in a democracy? It is aggregating and averaging”, Frontline, Chennai, June 15-28, 1996.

27 The liaison committee mechanism of the United Democratic Front (UDF), the “apex body of Front constituents” has no formal role in the day-today functioning of the government; however, all major policy decisions are first discussed here and only if there is a consensus here does the government move ahead. This Kerala example was supposedly cited as a way of ironing out differences between political parties, The Hindu, Chennai, May 26, 1996.

28 Rejecting the proposal for a formal coordination committee the Congress spokesman, V N Gadgil said, “We are not in favour of it. If any talks are to be held at all, these must take place between the prime minister and the leader of the Congress parliamentary party”, The Hindu, Chennai, January 17, 1997.

29 The Standing Committee of the Steering Committee was formed on February 12, 1997.The only major difference that could be gathered was that while the SC met at the UF headquarters, the core committee met at the prime minister’s residence. The two committees almost always met on the same day and the SCSC meet usually preceded the SC meeting. The SCSC composed of the more influential members was restricted to 11 members of the Front and was formed to discuss and deal with day-to-day affairs of the Front.

30 Their political careers were at different levels. Kesri in his first press conference had appealed to the UF not to try to break the Congress and he is reported to have informally indicated that it was trying to lure away Congress members, The Hindu, Chennai, October 7, 1996; another point of irritation was that a PIL had been filed in the Delhi High Court asking the CBI to investigate into Kesri’s assets, The Hindu, Chennai, November 15, 1996. After the fall of the Gowda headed-UF government, Indrajit Gupta (CPI) noted that Gowda and Kesri did not get along well. Kesri is supposed to have told him “this man, the prime minister, is treating us like an enemy and wants to destroy my party and I could not sit quietly and tolerate it. Ever since I became president of the Congress he seems to think I am a potential enemy or rival of him, which I am not”, The Hindu, Chennai, April 18, 1997.

31 K Karunakaran, who had acted as a liaison man between Kesri and Gowda remarked at a press conference in Trissur, Kerala, “In the Congress party the Congress parliamentary party leader is subordinate to the party president, and whoever wants to know the position of the party on any major issue must consult the president”, The Hindu, Chennai, October 17, 1996. Gowda had met Rao on March 29, 1997 and the Congress announced

withdrawal of support on March 30, 1997.

32 The Hindu, Chennai, April 21, 1997.

33 One of the first major decisions by Gujral was to transfer the Central Bureau of Investigation (CBI) chief Joginder Singh, a Deve Gowda appointee, who had been actively pursuing leads in the Ashok Jain Foreign Exchange Regulation Act (FERA) violation case. The Congress fund collection from abroad and Kesri, the long time treasurer of the Congress and his connections with Jain was purportedly linked with this case. Other instances include the transfer of the Income Tax (IT) officer who had fined the Congress for not filing its IT returns. Gujral’s handling of the Uttar Pradesh crisis in October 1997 was more in tune with the Kesri’s thinking than the UF stand. The press reported that Kesri and Gujral met often. Others also testified to this close and cordial relationship shared between them. A CWC member Vijaya Bhaskar Reddy noted that the excellent equation between the two was a reflection of the coordination between the UF and the Congress, The Hindu, Chennai, October 4, 1997.

34 At the state level the Congress was the main competitor to many of the UF constituents. The Uttar Pradesh Congress Committee (UPCC) was one of the first to appeal to the central leadership to withdraw support to the UF, The Hindu, Chennai, October 21 and November 4, 1996. The Karnataka PCC called for withdrawal in November 1996, The Hindu, Chennai, November 3, 1996. Mamata Banerjee who broke off to form the Trinamool Congress was against the policy of supporting the UF when the Left Front was the principal competitor in West Bengal.

35 Kesri attended the SC meeting on October 15, 1996 to enlist support arguing that a move to support a BSP-led coalition would signal both social justice and gender equality. He wanted some compromise as the Congress-BSP alliance predated the UF-Congress alliance. The Samajwadi Party (SP) threatened to walk out of the alliance if this was done. Mulayam Singh Yadav (SP) incidentally did not take part in this meeting, The Hindu, Chennai, October 16, 1996.

36 Sharad Pawar and Rajesh Pilot were supposed to be prominent challengers. Dissident voices included Mamta Banerjee, Jitendra Prasada and A R Antulay. The shadow of the “natural” claimants to the leadership, the Nehru-Gandhi family always existed. Old “family” loyalists like Arjun Singh, K Karunakaran and N D Tiwari among others were also waiting for slip-ups. Kesri was accused of fixing the voter list for the party elections. He also lost control over many of the state units. While Mamta Banerjee and Bangarappa moved away in West Bengal and Karnataka respectively, Jagannath Mishra, R L Yadav and Krishna Sahi decided to form a new party in Bihar.

37 The Jain Commission report on the Rajiv Gandhi assassination ceased to be an issue for the Congress beyond the few weeks after the leaks of it appeared in the India Today, New Delhi. The Congress had no qualms in asking the same UF constituents to support its claim after the fall of the UF and also after the 1998 elections.

38 The abbreviation NDA is used to denote the BJP-led coalition experiments between 1998-2004.

39 For instance, it unfailingly met before each parliamentary session.

40 The Manipur crisis in May 2001 between the BJP and the Samata Party (SAP) was not discussed as a matter of the NDA but purely as an interparty issue between the two parties, The Hindu, Chennai, May 22, 2001; The Economic Times, New Delhi, May 21 and 23, 2001. Similar problems in Jharkhand in May 2002 and Tamil Nadu in September 2002 did not affect the national level.

41 Joint Statement issued after the Coordination Committee meeting held on February 2, 1999, The Hindu, Chennai, February 3, 1999

42 Sanjay Nirupam, then Shiv Sena member connected the PMO with the Unit Trust of India (UTI) muddle in a debate in the Rajya Sabha. At an emergency meeting of the coordination committee the prime minister is reported to have told those who cannot maintain discipline could go. The convenor George Fernandes reiterated that those in the government should not go outside and criticise the government

Green Left

decisions “and if they do, they are free to go”, The Hindu, Chennai, August 2, 2001.

43 The committee to draw up the code of conduct comprised of Sikander Bakht (BJP), Mursoli Maran (DMK), Arjun Charan Sethi (BJD), and George Fernandes (SAP), The Tribune (Chandigarh), August 2, 2001

44 Then BJP president Jana Krishnamurthy remarked that the NDA is not a railway compartment with passengers going in and out at every station. Cited in Neena Vyas, “A Revolving Door Called the NDA”, The Hindu, Chennai, September 2, 2001.

45 A number of all party meetings were held on the issue of reservation of seats for women on which there was no internal consensus in the NDA but had support in other parts of the political spectrum.

46 Its membership was not restricted to the cabinet as at times the deputy chairman of the Planning Commission also headed such groups. Besides the empowered committee members, other departmental ministers could/ were also invited to share their views.

47 The SAD’s (Badal) concerns regarding the Uddam Singh district was taken up by a GoM. Rajasthan, a Congress-ruled state had been crying for drought relief probably every single year. However, the government took care only when it came to dealing with a similar problem in Andhra Pradesh by setting up a GoM. This GoM was constituted in the year 2001 much before the country as a whole was affected in 2002. Significantly in 2002 there was no GoM to deal with the crisis indicating the possibilities of using the GoM mechanism to pander to the allies.

48 When the NAG was released, the NDA had observed, that theirs was not a minimum programme like that of the UF. The UPA almost as if to deflect criticism stated that the CMP was the foundation for another CMP-collective maximum performance, National Common Minimum Programme, source (accessed October 14, 2004).

49 Guided by six governing principles the CMP reflects the shared concerns of the different partners on various issues. Following a target-based approach it does not go into the details but only sketches their concerns, policies and priorities in certain areas. The left parties endorsed the CMP through a signed statement. 05272004_cmp_left.htm (accessed October 14, 2004)

50 Press release prime minister’s office (PMO), May 28, 2004 release/release.asp?relid=1829&kwd (accessed May 29, 2004)

51 DMK president, K Karunanidhi displayed the portfolio distribution agreement to show that the accord was not followed at the time of allocation, The Hindu, Chennai, May 25, 2004.

52 For example, the GoM on affirmative action includes, Sharad Pawar (NCP) who is the chairman, Lalu Prasad (RJD), Ram Vilas Paswan (LJP), P Chidambaram (INC), Kamal Nath (INC), Meira Kumar (INC), and Dayanidhi Maran (DMK). The UPA has also included chief ministers in the GoMs, for example the GoM on the 350th anniversary celebrations of the Taj Mahal includes the chief minister of Uttar Pradesh.

53 The GoM mechanism is being examined in a more detailed study that is underway.

54 CPM Politburo member Jyoti Basu speaking on the decision to form such a committee, The Hindu, Chennai, June 7, 2004

55 The National Advisory Council (NAC) constituted in May 2004 is headed by a chairperson with the rank and status of a union cabinet minister and it could have a maximum of 20 members. It is funded through the prime ministers office and is authorised to engage the services of experts and academics to assist in its working and is expected to meet once in every quarter. (accessed December 30, 2005).

56 The Left Front’s blow-hot blow-cold stand is an example of this future anticipation. Their multiple objectives included preparation for assembly elections in Kerala and West Bengal where the INC is the principal opponent, prevent the BJP from cashing in on the UPA weakness and ensure that the opposition space is not captured wholly by the BJP.

57 M Karunanidhi remarked that “there are lessons to be learnt from experiences of coalition governments in the recent past. If these lessons are properly learnt, the coalition governments at the centre can not only complete their full term, but also achieve their broad objectives, in terms of meeting the aspirations of the people successfully”, M Karunanidhi, ‘Inaugural Address’ in Lakshmi Krishanmurthi, R Hariharan and Gert W Kueck (eds), Making a Success of Coalitions, East West, Chennai, 2000. He was delivering the inaugural address at a national workshop ‘Making a Success of Coalitions’, December 7 and 8, 1999 at Chennai.

58 See list of the subjects on which it has communicated to the government. Source: (accessed February 2, 2006)

59 March and Olsen, 1989, op cit, p 50.

60 A discussion paper entitled Government Subsidies in India was issued to generate debate. The Targeted Public Distribution System based on a two-tier subsidised pricing system was introduced in June 1997, Source: Economic Survey 1997-98, Ministry of Finance, Government of India, New Delhi, 1998. On the issue of petroleum products the government set in motion the phased withdrawal of the Administered Pricing Mechanism (APM) and the prices of major petro-products were gradually linked with market prices. To check the deficit in the oil pool account oil bonds were issued, Source: Economic Survey 1998-99, Ministry of Finance, Government of India, New Delhi, 1999.

61 The decision of the Left Front, AIADMK, and Trinamool Congress who have been/are members of some coalition or the other to “boycott” coordination meetings at times and yet remain in the alliance should be seen in this light.


Arora, Balveer (2002): ‘Negotiating Differences: Federal Coalitions and National Cohesion’ in Francine R Frankel et al (eds), Transforming India: Social and Political Dynamics of Democracy, Oxford, New Delhi.

Browne, Eric, John Frendreis and Dennis W Gleiber (1984): ‘An “Events” Approach to the Problem of Cabinet Stability’, Comparative Political Studies, 17(2), July, 167-97.

Chakrabarthy, Bidyut (2006): Forging Power: Coalition Politics in India, Oxford University Press, New Delhi.

Kashyap, Subash (1997): Coalition Government and Politics in India, Uppal, New Delhi

Krishanmurthi, Lakshmi, R Hariharan and Gert W Kueck (ed) (2000): Making a Success of Coalitions, East West, Chennai.

Laver, Michael, Ian Budge (eds) (1992): Party, Policy and Government Coalitions, St Martin’s Press, London.

Laver, Michael and Kenneth A Shepsle (1996): Making and Breaking Governments: Cabinets and Legislatures in Parliamentary Democracies, Cambridge University Press, Cambridge.

Laver, Michael and Normal Schofield (1998): Multiparty Government: The Politics of Coalition in Europe, Oxford University Press, Oxford, 1990.

Levitt, Barbara and James G March (1988): ‘Organisational Learning’, Annual Review of Sociology, 14:319-40.

Mantzavinos, C, Douglas C North and Syed Shariq (2004): ‘Learning, Institutions and Economic Performance’, Perspectives on Politics, 2(1), March, 75-85.

March, James G (ed) (1999): The Pursuit of Organisational Intelligence, Blackwell, Malden, Massachusetts.

March, James G and Johan P Olsen (1989): Rediscovering Institutions: The Organisational Basis of Politics, Free Press, New York.

Neisser, Ulric (2003): ‘Cognitive Psychology’ in Margaret P Munger (ed), The History of Psychology: Fundamental Questions, Oxford, New York.

Peters, Guy B (1999): Institutional Theory in Political Science: The ‘New Institutionalism’, Pinter, London.

Posner, Michael I (1998): ‘Current Research in Cognitive Neuroscience’ in Indramani Singh and Raja Parasuraman, Human Cognition: A Multidisciplinary Perspective, Sage, New Delhi, pp 41-54.

Riker, William (1962): The Theory of Political Coalitions, Yale University Press, New Haven.

Singh, M P and Anil Mishra (ed) (2004): Coalition Politics in India: Problems and Prospects, Manohar, New Delhi.

Sridharan, E (1999): ‘Principles, Power and Coalition Politics in India: Lessons from Theory, Comparison and Recent History’ in D D Khanna and Gert W Kueck (eds), Principles, Power and Politics, Macmillan, New Delhi, pp 270- 90.

Swaan, Abram de (1973): Coalition Theories and Government Formation, Elsevier, Amsterdam.

Thakurta, Paranjoy Guha and Shankar Raghuraman (2004): A Time of Coalitions, Sage, New Delhi.

Warwick, Paul (1994): Government Survival in Parliamentary Democracies, Cambridge University Press, Cambridge.

Dear Reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here

Back to Top