Election Commission of India (ECI), Elections, Government of India, India, LS Elections, NOTA, Stories

Data doesn’t indicate any correlation between use of EVMs & Anti-Incumbency


Few Psephologists have suggested that there was 90% pro-incumbency in Lok Sabha & Assembly elections when India used ballot papers. They went onto say that the use of EVMs has reversed this trend and now, more than 90% of the results are anti-incumbent. But the data doesn’t support this theory. 


Data from the Election Commission of India (ECI) website was considered for this analysis. The methodology is explained towards the end of the article.

A total of more than 350 election results (State Assembly & Lok Sabha) were considered and each of the results were categorized as ‘Pro’ or ‘Anti’.

Now much difference between Pre-EVM & Post-EVM results

The consensus on the use of Electronic Voting Machines (EVMs) was reached in 1998 and their use was progressively increased since then. EVMs were used for all the constituencies starting 2001.

Before EVMs were introduced in 1998, more than 220 elections were held. Out of these, the incumbent party lost in 121 (53.5%) where as the incumbent won in the remaining 105 (46.5%).

Till date, after the introduction of EVMs in 1998, more than 120 elections were held. Out of these, the incumbent lost in 70 (56%) while the incumbent won in 55 (44%).

In other words, the pro & anti-incumbent trends were not very different before & after the introduction of EVMs.

No effect of EVMs on the incumbency trends

The pro or anti incumbency trends across the years were more based on the political climate during that time than the use of ballots or EVMs.

For instance, the first big anti-incumbency trend was in the 1977 elections (immediately after emergency) when out of the 18 results, 16 were anti-incumbent and only two were pro. Similarly in 1980, post the failure of the Janata experiment, 13 out of the 15 results were anti-incumbent and only 2 were pro. In states like Tamil Nadu & Kerala, the anti-incumbency trend started in the late 1960s, after which the incumbent lost in almost every election.

So, it can be concluded that there is no impact of the use of EVMs on the incumbency trends in Indian elections.

For the purpose of this analysis, the following methodology was followed.

  1. All election results in Independent India were considered except the first election in a respective state or a province.
  2. A result is categorized as ‘Pro-incumbent’ when the a party retained power singly or in alliance with another party. Cases where a party lost seats compared to the previous election, but still retained power, are categorized as ‘pro-incumbent’.
  3. A result is categorized as ‘Anti-incumbent’ if the ruling party loses election and a different party or grouping came into power.
  4. In cases where different parties ruled during the same tenure, the party/coalition that ruled the longest is considered for categorization as pro or anti.

About Author

Rakesh has been working on issues related to Right to Information (RTI) for a decade. He is a Data/Information enthusiast & passionate about Governance/Policy issues.

Comments are closed.