The Lok Sabha elections earlier this year witnessed extensive use of internet by political parties and candidates in the campaign phase.
By: Prasanna Tantri
Many political analysts believe that internet and social media played an important role in shaping the electorate’s choice. With assembly elections for the states of Jammu and Kashmir, Maharashtra, Jharkhand and Haryana round the corner, the role played by internet-based media in the political process is, once again, under the spotlight.
Politicians of all hues of the political spectrum, even the ones with no exposure to the internet hitherto, are trying their hand at using social media. Careful academic work in this area has thrown up surprising findings: For a start, availability of internet significantly reduces voter turnout. More importantly, there is no evidence to show that internet systematically benefits any particular party; in fact, internet-based media just leads to ideological segregation.
While there has been a lot of work on the influence of mainstream media on the political process, there is little yet on the role played by the internet.
Robert D Putnam, in his research paper published in the Journal of Democracy in 2000, stated that “some of the allegedly greater democracy in the cyberspace is based more on hope and hype than on careful research”.
It is possible to come out with arguments, both in support of and in opposition to the thesis that internet has a positive impact on voter participation in the political process. On the positive side, the internet significantly reduces the cost of acquiring information and, at the same time, it also removes many filters applicable to the traditional media. Internet-users are exposed to much wider points of view.
However, on the flipside (strictly from the point of view of political participation), internet availability also leads to what economists term the “crowding out” effect. In other words, the internet greatly enhances the opportunities available for entertainment. Therefore, it is possible that additional entertainment avenues created by the internet may eat away the time people spend in comprehending political developments. In that case, the internet may lead to reduced political participation.
Thus, it is extremely hard to assess the impact of internet on political participation using theory. The only way left in such a situation is empirical testing of the competing hypothesis. Establishing causality in an empirical phenomenon is easier said than done.
Let’s take the case of influence of internet: the fact that leaders who extensively used internet-based platforms for campaigning happen to win the election is not sufficient to prove the positive role played by the internet. It is possible that such leaders might have held more traditional rallies and their party workers also worked extra hard.
A mere existence of a negative correlation between availability of internet and voter turnout is again not sufficient to show that internet leads to reduced political participation. This is because areas with high internet connectivity could be systematically different when compared to areas with low connectivity. People in such areas may differ with respect to education, preferences, wealth, etc, all of which may influence their political participation.
In such a scenario, researchers look for an event which leads to an increase in internet availability but does not have any effect on other factors which determine political participation. Such an event is termed as exogenous. Now, any change in political participation in the presence of such exogenously-determined treatment can safely be considered as causal.
Oliver Falak, Robert Gold and Stephan Heblich examine one such instance, of an exogenous change, in internet availability in Germany.
In their recent research paper, published in the American Economic Review, they use the fact that the roll-out of high-speed internet was delayed in some municipalities purely due to technological errors. Using this exogenous variation in internet availability, they identity areas with high-speed internet as “treatment” areas and areas where the roll-out was delayed as “control” areas. They study the differential voting behaviour in the control and treatment areas.
They document an economically as well as statistically significant 5% drop in voter turnout in areas highly exposed to internet. They also show that internet does not alter the ideological preference of the users. Users show a tendency to self-segregate: left-leaning citizens expose themselves to more left-leaning material on the internet and right-leaning ones to more right-leaning material. Their study does not find evidence of the internet disproportionately benefiting any particular party or ideology. Interestingly, they also find that the internet ends up crowding out TV-viewing while the newspapers largely remain unscathed.
Even in India it is quite evident that voter turnout in areas with high internet availability is much lower than areas with lower internet availability. However, as the availability of internet is not exogenous, it is hard to make causal claims on Indian data. We are in the process of designing causal tests in this regard.
The above research findings raise a note of caution for those politicians who may be overtly relying on the internet-based media for their campaigns. There does not seem to be any substitute for maintaining direct contact with the electorate. Modern methods can, at best, work as compliments to the traditional campaigning methods. Just because successful campaigners used new methods does not mean that new methods led to success. Even in the age of internet, the political karyakarta remains extremely relevant for mobilising support for political parties.
The author is the associate director, Center For Analytical Finance, Indian School of Business
