India has sustained a national system of democratic governance in a country marked by substantial variation in geographic, economic, and demographic circumstances. This combination sparks the question whether democracy at the national level automatically translates into subnational democracy. Does democracy ‘trickle-down’, with states essentially mirroring national democracy? Or do we see diversity across states, with some states being distinctly more democratic than others?
A growing body of scholarship focusing primarily on large federal democracies in the Americas has demonstrated that undemocratic or even authoritarian subnational regimes often persist after national-level democratisation (for example, Gibson 2013, Giraudy 2015). In a recent paper (Harbers et al. 2019), we extend this type of analysis to India. Our findings reveal a complex panorama of subnational variation, with democracy thriving in some states even as it is under threat in others.
Conceptualising and measuring electoral democracy in India
To measure variation in electoral democracy across Indian states, we develop a conceptual model based on four constitutive dimensions: turnover, contestation, autonomy, and clean elections. In order for states to be considered democratic, they must score high on all of these dimensions. The absence of any one of them translates into low scores on the overall index. Figure 1 illustrates our conceptualisation.
Empirically, our analysis covers roughly three decades. We score the degree of electoral democracy for 28 states and the union territories of Delhi and Puducherry for each electoral period between 1985 and 2013. The indicators in our dataset are available for 171 elections.1
The first two dimensions of the index tap into how competitive elections are. Turnover captures whether the majority party in the governing coalition has changed in the previous three consecutive electoral periods. Contestation measures the degree of competition in the party system based on two indicators: (1) the effective number of parties in the legislative assembly, an indicator not only of the number of parties but also their relative size; and (2) the strength of the opposition, which reflects the percentage of seats not controlled by the governing coalition.
Our third dimension – autonomy – indicates whether subnational elections actually determine who governs the state. If President’s Rule is imposed, the national government essentially takes over the administration.2 We therefore consider President’s Rule, and the temporary suspension of elected subnational institutions that goes with it, a violation of democracy, akin to “interrupted regimes” in cross-national research (see Doorenspleet 2000).
The fourth dimension – clean elections – reflects whether citizens are able to cast their vote in a free and fair manner. Our coding of this dimension is based on news reports of electoral malpractice, focusing specifically on the number of deaths due to election-related violence, and the percentage of constituencies for which incidents of malpractice are reported.
Figure 1. Conceptual model
Note: The logical OR (denoted by the symbol +) indicates substitutability, while the logical AND (denoted by the symbol *) indicates that each dimension is necessary. Scores for turnover and contestation are thus added, and then multiplied with those for autonomy and clean elections to create the index.
Our index of electoral democracy identifies compelling variation across states. Figure 2 maps the average score of each state or union territory over the whole time period. Uttarakhand and Kerala – displayed in dark blue – emerge with the highest scores on the democracy index, confirming Kerala’s status as one of the most successful examples of subnational democracy in India (Heller 2000). Jharkhand, Nagaland, Tripura, and Arunachal Pradesh are found on the other end of the scale, and receive the lowest score on the index.
Figure 2. Average score on electoral democracy index
Note: Darker shades of blue indicate higher values on the electoral democracy index. The map reports the mean score for the period 1985-2013. For states formed during this period, the score reflects the period of statehood.
Figure 3 provides disaggregated scores across dimensions and over time. It shows that variation exists across the constitutive dimensions, indicating that threats to subnational democracy come in multiple forms. Firstly, democracy can be undermined by a lack of robust party competition. The combined score of turnover and contestation – shown in the figure through the position of the circle – points to problems with party competition, and signals that the electoral playing field is tilted systematically against challengers of the incumbent elite. Examples of states where this is the case are Arunachal Pradesh and more recently Gujarat.
A second potential threat is the lack of election security, resulting in (violent) disruptions of the electoral process. When the triangle is below the circle, this indicates that the overall score on the index is lower because of instances of electoral malpractice. Our database of news reports documents bomb explosions at polling stations, boycott calls by insurgent groups, and threats against campaign workers. These instances of anti-systemic violence constitute a ‘horizontal threat’ to subnational democracy. We find that anti-systemic violence can undermine subnational institutions both in states with an otherwise strong democratic track record, including Assam, but also worsen the situation in states that already have a low score, such as Nagaland.
Lastly, the autonomy of subnational regimes, and consequently subnational democracy, is undermined by instances of President’s Rule. The index drops to zero when the subnational regime is temporarily interrupted. President’s Rule can be seen as a ‘vertical threat’ to subnational democracy as the intervention by the national government sidesteps subnational institutions. In states such as Uttar Pradesh (UP), the use of President’s Rule interrupts an otherwise high score on the index. In other states, such as Jammu & Kashmir, President’s Rule is just one of several threats to subnational democracy.
Figure 3. Democracy index by state
Beyond a comparison across dimensions, the index also allows us to examine how different states develop over time. Observing these temporal trends, it appears that the states of Goa, Punjab, and West Bengal have become progressively more democratic between 1985 and 2013. During the same period, however, Gujarat, Meghalaya, Sikkim, and Tripura seem to be “backsliding” (Bermeo 2016) and becoming less democratic. Whilst literature on relatively new democracies in Latin America has shown that authoritarian elite sometimes remain in power at the subnational level after a national transition to democracy, these cases represent a different type of trend: subnational democratic decline.
Our findings show a diverse panorama of democracy across Indian states. In some states, parties that initially challenged the dominance of Congress have now become dominant forces in their own right. While there has been substantial progress with regard to election security under the auspices of the Election Commission of India, violence and malpractice remain a concern in states such as Bihar and UP. In the new states of Chhattisgarh and Jharkhand, subnational democracy is struggling. Overall, our analysis paints a nuanced picture of democracy across states, which in turn indicates the need to take into consideration the variation in subnational circumstances when crafting policies to safeguard democracy and empower voters.
- The data collection was supported by a H2020 grant from the European Research Council [Grant #323899].
- President’s Rule is authorised in the case of “a breakdown of the constitutional machinery”, as stated in Article 356 of the Constitution. However, scholars have noted that Article 356 has often been declared in circumstances where a breakdown had not taken place, and was rather used to intimidate state governments or resolve intra-party conflicts.
This article was originally published in Ideas for India
Stay updated with all the insights.
Navigate news, 1 email day.
Subscribe to Qrius