The ruling dispensation in the last 5 years or so has on multiple occasions advocated simultaneous holding of state and national elections with the ostensible objective of reducing administrative and financial cost incurred on holding elections at staggered intervals. Recently the proposal has once again surfaced after the new BJP government has come to power.
Reasons for Supporting Simultaneity
There are substantial number or arguments that have been put forward in support of the same. A report on “Feasibility of Holding Simultaneous Elections Lok Sabha and State Legislative Assemblies’ advocated it on three grounds which included policy paralysis that results from the imposition of the Model Code of Conduct during election time, impact on delivery of essential services and burden on crucial manpower that is deployed during election time. 
The other major argument that is often quoted to support simultaneity is the sheer volume of arrangements and resources that is needed to conduct an election in India. In 1952, the size of the electorate constituted 173 million voters, by 2014 it had gone up to 863 million and in 2019, there were 900 million voters which included an estimated 130 million first time voters.  The expenditure incurred by the Election Commission of India (ECI) on conducting national elections however went up from Rs 10.52 crores in 1952, to Rs 3870 crores in 2014, that is 367 times, which roughly translates into a YOY growth rate of 6%. Figures spent on the 2019 elections by the ECI could not be accessed by the author. The funds spent by ECI is however only a fraction of the total money that is spent by political parties. As per a study by the Center of Media Studies, political parties and candidates spent nearly Rs 60,000 crore in India’s recently concluded general election of 2019, making it twice the amount spent in 2014. It is argued that enforcing simultaneity would substantially reduce the expenditure by political parties since it would reduce duplication on state and national elections. 
Implications of enforcing Simultaneity
However there are certain implications that might result in the enforcement of simultaneity. Experts suggest that it may well open the backdoor to situations like the 42nd Amendment Act. Enforcing it may also require increasing or decreasing the term of Parliament and State Assemblies which detracts from the constitutional provision of a full five- year term of legislatures. The Natchiappan Committee made the suggestion of holding elections in two phases, one synchronized with the end of the Lok Sabha and the other sometime during the midterm of Lok Sabha to resolve this practical problem.
The fact of the matter is that lengthy election schedules appear to be unavoidable even during the recently concluded parliament election which was conducted in 7 phases from 11th of April to 19th of May, 2019. Even during this more than a month long jamboree, there were concerns voiced, that it gave enough time for manipulation of later phases of voting. Simultaneous election of state governments will simply imply that the elections will be one long one year instead of being spread over 5 years. There will therefore be continued practical challenges of ensuring the required administrative and logistical arrangements which however was not considered as an insurmountable problem as per a discussion paper by Niti Aayog. 
The Impact on Federalism
However, there are far more powerful implications as regards the democratic federal structure of the country that could be impacted by this move. In the last couple of years, the landscape of Indian politics has considerably changed. The dominance of the Indian National Congress (INC) has been considerably diluted to see the emergence of another powerful alternative, the Bhartiya Janata Party (BJP). This has also simultaneously coincided with the emergence of many regional parties and satraps that require all national parties to frequently align with them either explicitly or through other non-tacit agreements especially during the state and local elections. There has also been a phenomenal increase in money muscle power and use of carefully orchestrated media coverage through all mediums to influence perception and image of parties and their leaders. Winning elections in India therefore, not only requires a poll manifesto that aligns with the aspirations of the Indian voter but also needs in great measure, ingredients of substantial political funding, organizational acumen, well-crafted strategic alliances, careful selection of political aspirants with the accompaniment of a well-designed intensive and intrusive media campaign. In a bid to gain power, political parties apart from the ostensible development agenda also factor in calculations and equations based on caste, creed, religion, language, economic status, nationalistic rhetoric and topographical features while topping it with generous infusions of money and media power.
Simultaneity in national and state elections with attendant powerful political campaigns can therefore put the voter in a decision conundrum of choosing between state and national issues for which he or she may not be sufficiently informed or even empowered. The partisan role played by the media can sufficiently throw the voter off-balance. There is thus a high possibility of state elections mirroring national elections or vice -versa. This can throw up fractured mandates in national elections if state issues are more dominant in the minds of the voter or more powerful national parties that evict local and regional parties should voters vote more on national considerations.
Not only is the possibility of the voter being influenced either way, there is also the added possibility of the dominant party using both the national and state level machinery to influence state elections especially with state funding and certain projects dependent on central largesse. As such making a choice in favor of simultaneity of holding elections sans the necessary check and balances of enforcement of transparency in political funding, expenditure limits, and models of election conduct can further harm the edifices of a vibrant and independent democratic tradition in the country
The impact on federalism has been corroborated by research studies as well. A study conducted by the IDFC institute and reported by Hindu concluded that there was “on average, a 77 per cent chance that the Indian voter will vote for the same party for both the State and Centre when elections are held simultaneously”. This analysis was done for four rounds of electoral data for 4 rounds of Lok Sabha elections of 1999, 2004, 2009 and 2014 wherein state elections coincided with that of national. They noted that the “trend of choosing the same party has gone from 68 per cent in 1999 to 77 per cent in 2004 to 76 per cent in 2009 and 86 per cent in 2014.” 
A similar study was done by Jagdeep S. Khokhar and Sanjay Kumar using figures from 1989 elections onwards found, that when simultaneous elections for the Assembly and the Lok Sabha were held, it was found that in 24 elections, the major political parties polled almost a similar proportion of votes both for the Assembly and the Lok Sabha, while only in seven instances was the choice of voters somewhat different. 
Analysis by us using only 2014 election results
A similar analysis was done by the author of this article but using only the 2014 elections which was unique at many levels as it gave rise to the hegemony of the BJP. One found that there were 9 states where state elections were held after the May 2014 general elections within the 0-9- month period constituting 1148 assembly constituencies and 151 parliamentary constituencies. For the purposes of doing this analysis, all BJP allies (Shiv Sena, Telugu Desam, Shiromani Akali Dal, Sikkim Democratic Front etc.) were taken to be part of the winning NDA party combination and similarly all the other parties besides NDA were clubbed as non-NDA.
Out of the 151 parliamentary constituencies that subsequently went for state polls, BJP and its allies had won 96 seats or approximately 64% of the contested seats in 2014. These 151 parliamentary constituencies translated into 1148 state constituencies. This implied that at the central level the NDA won 737 seats but in actual state level was down to 501 seats whereas for non-NDA, it translated into 411 seats at the central level but increased its tally to 647 state level seats. Bifurcating Telangana and Andhra Pradesh as two separate states, the NDA won 6 out of 10 states having lost Delhi, Orissa, Telangana and Arunachal Pradesh that subsequently went for elections post 2014 within a 9- month period.
Schedule of State Elections Held in the 0-9 month period post May-2014 General Elections
Percentage of Seats Won in the State Elections by NDA and non-NDA compared to National elections in the corresponding constituencies
Building the Model
To arrive at estimates for probability of winning at both national and state level, a logistic regression model was used. The data for constructing the model was taken from the Election Commission Site, whereas party affiliations were adjusted for after going through the site of www.indiavotes.com. Some of the other variables that was used to construct the model was the total voting percentage, the age of the candidate, the time duration from the period when the elections were held, and the type of the constituency. Since most of the candidates were male, sex of the candidate was not found to be a statistically significant variable in the model.
Results of the Model
The results of the model were astounding. The results seemed to suggest that other variables being constant, the party that wins at the center would have a winning log odds ratio of 8.23 : 1 for state elections. Even after adjusting for state level differences, the log odds ratio of winning the state elections would still be approximately 5.16 : 1 should the same party wins at the center. More important, the duration of the interval between the central and the state level elections was an important variable which seemed to suggest that the greater the duration between the two elections, the probability of the winning party at the center also winning elections at the state would considerably decline.
Using this model without adjusting for state differences and using a 0-9-month window period, it appears that should a ruling party win at the center, keeping other variables constant, there is an average 60% probability that it will win the corresponding state elections as well. This probability however varies with the probability being 72.2% if the elections in the state are held alongside the center elections, comes down to 56.2% for state elections held after 5 months, to 49.2% for after 7 months and 42.2% after 9 months.
It is often said that politics and politicians are both fickle. Election results too are equally uncertain. To rule the roost even for 5 years, the stakes are high, and therefore the investment too can be very high. In this process, there are innumerable factors that come into play which can range from money power, perceptions, political and development agenda, caste, religion, socio-economic status, prowess of the candidate, political alliances, percentage of women and young voters, ant-incumbency factor, to even number of parties and participants in the fray. Therefore, any statistical model would never be able to account for all factors given the high degree of variation within each constituency.
Nevertheless, there appears to be strong evidence that national elections do have a high probability of affecting results at the state level as well. For a vibrant federal and democratic set up, that may not be very healthy to have dominance of a party both at a national and state level. As such the proposal to have simultaneity in both state and national elections needs to be examined far more rigorously for its repercussions on the overall democratic foundations of the country.