The Samajwadi Party won the two by-elections in Gorakhpur and Phulpur, with large margins. Inevitably, these are seen as strong signals ahead of the upcoming general elections. They also constitute a warning for UP Chief Minister Yogi Adityanath and his deputy, Keshav Prasad Maurya, who had held those seats.
These results matter for obvious symbolic reasons. For the BJP, retaining the chief minister’s seat was a matter of pride and safety. For the SP and the BSP, the victories vindicate the idea that only an alliance of Opposition parties can defeat the BJP in the state where it won its largest victories in 2014 and 2017. The alliance presented these by-elections as the first real popularity test of Yogi Adityanath and as a rehearsal for the general elections. The BJP has held the Gorakhpur seat continuously since 1991 (1989, if one clubs it with the Hindu Mahasabha). In politics, symbols matter.
However, it would be a mistake to attribute any predictive value to the two by-elections.
First, as is usual for by-elections, the turnout was low (47.5 per cent in Gorakhpur and 37.4 per cent in Phulpur). Since 1964, the average turnout in by-elections in UP is 41 per cent, against the 51 per cent in the same seats in general elections. This factor played in favour of the alliance. Second, it is not unusual for Lok Sabha by-polls to have an elite character or strong symbolic overtones, given the profile of the individuals who relinquish their seat, either by switching House, seat, state, or due to their demise. Rajiv Gandhi succeeded his brother in Amethi in 1980 in a by-election. Neeraj Shekhar, son of former Prime Minister Chandra Shekhar, retained his father’s seat in Ballia after his demise in 2008.
The symbolic charge of the Gorakhpur and Phulpur by-polls is compounded by their respective histories. The Gorakhpur seat has been associated with the head of the Gorakhnath temple since the early 1960s. Yogi Adityanath’s mentor, Yogi Avaidyanath, won the seat in 1970 after the death of his predecessor, Hindu Mahasabha leader and head priest of the Gorakhnath Math, Digvijaynath. The BJP has retained its hold on this seat since 1991.
Situated on the outskirts of Allahabad, Phulpur has been a site of contest between political giants practically since Independence. Jawaharlal Nehru contested and won the seat twice, in 1957 and 1962 (the second time against Ram Manohar Lohia). His sister, Vijaya Lakshmi Pandit succeeded him in a by-election in 1964. She prevailed over the socialist Janeshwar Mishra, who won the seat in the next election, defeating K.D. Malviya, petroleum minister in Indira Gandhi’s cabinet. V P Singh occupied the seat in 1971. His successor, Kamala Bahuguna, turned the constituency into a socialist stronghold, which remained so until Keshav Prasad Maurya’s victory in 2014. It is common that Lok Sabha by-elections involve high-profile seats.
This has meant that in most cases, there has been a strong incumbency advantage for whoever previously held the seat. Since 1962, the incumbent party has retained the seat in 68.4 per cent of the cases (26 out of 38).
Third, Lok Sabha by-elections are not good predictors of performance to come, nor a good measure of popularity of the government in place at the state level. Among the 35 by-elections that have taken place between 1962 and 2014 in UP, only 11 results matched the party that won the following general election. By-election winners also matched the party in power at the state level at the time in 58 per cent of the cases.
An analysis of 153 Vidhan Sabha by-elections in UP confirms their poor predictive value. By-election results have matched the result of the subsequent election in only 46 occasions (31 per cent of the cases). If we isolate the 54 by-elections that have taken place in the two years preceding a general election, or the 77 by-elections that have taken place within two years after a state election, the ratio remains roughly the same (29 per cent and 34 per cent, respectively). Seventy per cent of the seats that went to by-election within a year after a Lok Sabha poll changed hands, against 63 per cent for the entire period.
There is much evidence pointing that by-elections are fought and won over local factors and, above all, over the personality of the contestants. In the present case, the BJP thought it could sweep these two seats by fielding relatively low-profile candidates, banking on the popularity of the party and of the chief minister. They were mistaken. Local personality, caste combinations and the alliance factor prevailed.
Do these by-elections prefigure a successful SP-BSP alliance in 2019? Undoubtedly, such an alliance would be a force to reckon with. But this alliance remains a far-fetched idea for three reasons. First, there is a history of acrimony between the leaders of both parties. Akhilesh may have distanced himself from his father, but Mayawati never forgot her tussles with Mulayam. Given that the next election might be a question of survival for her and her party, she might, however, change her position.
Second, there are internal factors working against this alliance. Both parties have historically depended on three types of electoral bases. The first base comes from their core electoral groups (Jatav Dalits and Yadavs). The second base is provided by Muslim voters, who, by and large, have supported candidates from either parties. The third source comes from the candidates themselves, recruited among locally dominant groups, and whose electoral base may or may not coincide with the two parties’ core support bases. It is the third type of support that is likely to derail an SP-BSP alliance, since a seat-sharing agreement between the two parties means losing half the tickets they may aspire to. Many of them might contest on third-party tickets or even try to get BJP tickets.
And finally, should it happen, this alliance will last as long as there is no power-sharing consideration between the two parties, besides seat-sharing agreements. They may collectively defeat the BJP but they will not be able to govern together. Be that as it may, the race for 2019 remains wide open.