Penn Calendar Penn A-Z School of Arts and Sciences University of Pennsylvania
India in Transition

Coalition Types and Economic Reform

E. Sridharan
October 9, 2012

Can coalition governments in India be stable? And if so, can they undertake economic reforms and, more generically, policies that have short-term political costs but only long-term benefits? And if they do so, can they remain stable? The withdrawal from the United Progressive Alliance (UPA) coalition government of its second-largest constituent, the Trinamul Congress of West Bengal led by Mamata Banerjee, in September 2012, over policy announcements of reduction of fuel subsidies to tackle the fiscal deficit and liberalization of foreign direct investment, sharply poses these questions.

On coalition government stability in general and in India, the literature indicates that on average, the more the number of parties and the less their ideological connectedness, the less stable the coalition. By type of coalition, single-party majority governments are more durable than minority and/or coalition governments on average (and/or since coalition governments can be minority governments too). If we disaggregate by type of minority and/or coalition government, single-party majority governments are on average the most stable, worldwide, followed by majority coalitions, more so than minority governments, whether coalition or single party. Within majority coalitions, minimal winning coalitions – those that have the minimum necessary number of parties, not necessarily legislators, for a majority – are found to be more stable than surplus majority coalitions, or those that have more parties than necessary for a majority. However, if one changes the literature’s standard definition to one in which the exit of entry of parties does not change the coalition if the leading party remains the same, then surplus majority coalitions would be more stable on average.

Surplus majority coalitions, are rational inter alia as a political insurance policy so as to reduce the pivotal power of smaller parties, as in certain coalitions where parties to the left and right of the dominant coalition partner are kept on board for this reason, and as a political insurance policy against defection or political blackmail by factions within the leading party. A reform-minded leading party in a surplus majority coalition can be relatively invulnerable to the threat of defection from the coalition by smaller partners as none of the smaller partners is pivotal, unlike in a minimal winning coalition, and cannot be sure that other smaller partners will follow it in bringing down the government. Therefore, for a leading party that plans to implement a major economic or foreign policy shift, building a surplus majority coalition can be beneficial both as a political insurance policy, and positively as a broad consensus-building policy. A surplus majority coalition is likely to make the leading party particularly invulnerable when the main opposition party that is the likely nucleus of an alternative coalition has too few seats to command credibility and induce defections towards itself, and when the next election is still distant and no party wants to precipitate an early election.

In India, there is a tension between maximizing participation and coalition stability. The coalition era since 1989 – particularly since 1996 – has seen a great increase in participation, in that as many as thirty-five parties and five independents, most of them representing a variety of state-specific caste bloc/religious/tribal social constituencies, have participated in the Council of Ministers. From the literature, one would have expected that in a rent-seeking polity minimal winning coalitions that maximize the power (cabinet) share of each coalition partner as well as give the government a majority, would be typical of non-single party majority governments. However, counter-intuitively, all governments since 1989 – all non-single party majority governments – have been minority governments and except two (Chandra Shekhar, 1990-91, and Rao, 1991-96) all have been minority coalitions (dependent on external supporting parties) including the United Front governments of Gowda and Gujral (1996-98), the 1998-99 BJP-led government, the 1999-2004 BJP-led National Democratic Alliance (NDA) government and the UPA I (2004-09) and UPA II (2009, continuing).

However, two of them, NDA and UPA I, and three if the UPA II government lasts its full term, have been stable despite being minority coalitions. What has been the source of their stability? I have tried to explain elsewhere why such minority coalition governments have been formed and have also argued that what are formally minority coalitions actually have the character of surplus majority coalitions. This is what gives them and their leading parties stability. In a nutshell, if one considers not just the parties in government (the executive coalition) but the also the external supporting parties (taken together, the legislative coalition), no party has been pivotal for the NDA, UPA I and UPA II coalitions they are stable as they are, in effect, like surplus majority coalitions. Additionally, some supporting parties are “inter-locked” electorally at the state level with the leading party and are unlikely to pull out of a Central coalition. For example, the BJP and the Akali Dal, Shiv Sena and JD(U), are electorally inter-locked, and in a peculiar way, the SP, BSP and Left with the Congress despite state-level rivalries, as none of them would want to create an opening for the BJP to form an alternative coalition due to both ideological (Left) and state-level rivalries. Hence, the NDA, UPA I and UPA II have been stable.

Regarding the question of economic liberalization, what is the capacity of such minority coalitions to undertake these types of reforms? Economic liberalization involves policies such as fiscal cuts and opening of the economy to international competition that involve short-term costs to actual or potentially organized and mobilized groups such as labor unions and protected industries, but only deferred, uncertain and diffuse benefits. Hence, there are short-term political costs with only long-term and uncertain benefits for a reforming government. In the case of a coalition government, particularly a minority coalition, the issue of stability due to the pullout of coalition partners is a difficult one. Under what type of coalition arrangements might these types of policies become politically viable?

One can tentatively make the argument that if a surplus majority legislative coalition (including governing and supporting parties) of well over 300 (out of 545) seats, is led by a reform-minded leading party in a strong position, enjoying, say, 180-220 seats, with the leading opposition party too small in numbers to be credible as the nucleus of an alternative coalition (say, below 120 seats), the leading party will then be highly autonomous and able to undertake policies that require a longer time horizon, at least in the first half of its term before the electoral cycle turns unfavorable. This is because the leading party would be allied to a number of smaller partners, none being pivotal (by definition, in a surplus majority coalition) to the majority and not able/likely to undertake collective action vis-à-vis the leading party either because they are electorally inter-locked at the state level, because they are at loggerheads, or because there are too many parties needed for collective action, each would-be pullout being very small, or if none of the key supporters are anti-reform.

Looking back, in the case of the NDA, conditions (a), (b), (c), (d) (i) (the Akali Dal, Shiv Sena, JD(U), TDP and INLD were electorally inter-locked in their states with BJP), (d) (iii) and (d) (iv) were met. Therefore, it was able to make some reformist moves, such as disinvestment. In the case of UPA I, (b) and (c) were not met nor (d)(i) or (d)(iv) (the Left being a large chunk of the legislative coalition with sixty-two seats), although (d)(ii) (SP and BSP being at loggerheads) was. The relative weakness of the leading party (Congress had 145 seats) and the size of the anti-reform, statist/populist/rent-seeking parties (Left, SP, BSP) in the coalition made reform very difficult, even if desired. UPA II was somewhat like the NDA in that (a), (b), (c), (d)(ii) (SP and BSP) were met but not (d)(iv) since several statist/populist/rent-seeking parties (Trinamul, DMK, SP, BSP)  were part of the legislative coalition and the Congress itself was not reform-minded in the first half of its term. By the time it became reform-minded in 2012, the electoral cycle had become unfavorable despite other aspects of a surplus majority coalition favoring full-term stability. One could conclude that while a surplus majority legislative coalition can lead to full-term stability for even a formal minority coalition, only a surplus majority coalition with several other critical conditions in place would enable even a reform-minded leading party to undertake policies that have short-term costs but only long-term benefits.

E. Sridharan is the Academic Director of the University of Pennsylvania Institute for the Advanced Study of India (UPIASI). He can be reached at

 India in Transition (IiT) is published by the Center for the Advanced Study of India (CASI) of the University of Pennsylvania and partially funded by the Nand and Jeet Khemka Foundation. All viewpoints, positions, and conclusions expressed in IiT are solely those of the author(s) and not specifically those of CASI and the Khemka Foundation.

© 2012 Center for the Advanced Study of India and the Trustees of the University of Pennsylvania. All rights reserved.