Research shows that the media can be manipulated to accentuate bias in a sharply polarized polity
Even when there is a fair amount of press freedom, the government may manipulate the press by maintaining a patronage relationship with the media. Photo: Indranil Bhoumik/ Mint
While delivering an address in memory of Mahadev Govind Ranade (http://mintne.ws/ 1owTUBQ) on 18 January, 1943, B.R. Ambedkar caustically remarked about an apparently biased media: “The condemnation is by the Congress Press. I know the Congress Press well. I attach no value to its criticism. It has never refuted my arguments. It knows only to criticise, rebuke and revile me for everything I do; and to misreport, misrepresent and pervert everything I say. Nothing that I do pleases the Congress Press. This animosity of the Congress Press towards me can, to my mind, not unfairly, be explained as a reflex of the hatred of the Hindus for the Untouchables.”
The role of the media in the recently concluded Lok Sabha elections in India has been widely debated. Each political group believes that reportage, in general, is skewed in favour of the rival group. Even when there is a fair amount of press freedom, the government may manipulate the press by maintaining a patronage relationship with the media. It is, therefore, important to understand the underlying mechanism of media slant and its impact on political outcomes.
In his superb book Thinking, Fast, and Slow, Nobel laureate Daniel Kahneman put it succinctly how biases (he terms these biases as ‘availability heuristics’) not only inform our opinions but also shape some of the important actions (voting and community participation, for instance). He writes, “People tend to assess the relative importance of issues by the ease with which they are retrieved from the memory—and this is largely determined by the extent of coverage in media.”
Much of the peer-reviewed research in economics is devoted to questions of distribution and allocation—inequality, growth and development. However, there has been a recent shift, with a surge of available datasets, towards empirical analyses of issues that are considered to be the preserve of other social sciences like sociology, political science and anthropology.
There is now a rich set of evidence that answers two major questions on media bias. One, why media slant exists in the first place and what are the conditions that foster the bias? Second, how does the bias affect voting behaviour and, by corollary, electoral outcomes?
In recognition of the growing importance of the economics of media, the American Economic Association awarded the prestigious J.B. Clark Medal this year to the Chicago University economist Matthew Gentzkow for his path-breaking research on the economics and biases of the mainstream media. The Clark Medal for under-40 economists is considered the second most prestigious award after the Nobel Prize in the economics profession.
In a recent research paper, Gentzkow and his colleagues explain (http://mintne.ws/1nRCrbr) that bias can persist when the management of a media outlet has a proclivity towards gaining political mileage instead of being driven by simple profit-making incentives.
Earlier work by psychologists such as Mark Lepper, Lee Ross and Charles Lord Matthew showed that subjects with strong preferences on political issues become even more biased when presented with new information.
In a 2005 study on the market for news published in the American Economic Review (http://mintne.ws/1owUX4Q), Harvard University economists Sendhil Mullainathan and Andrei Shleifer presented a theoretical model to show that competition compels media to cater to the prejudices of the readers. Media consumers prefer news and opinion which align most closely to their prior views. Therefore, the duo contended that the news market will have an incentive to present news confirming partisan beliefs.
The theoretical literature makes a gloomy prognosis: media bias distorts information and voters are more likely to make mistakes if opinion is shaped through such sources.
A number of empirical investigations in different settings reveal the answer to the second question: whether media slant affects voting behaviour and electoral outcomes. A 2007 research paper by economists Stefano DellaVigna, of the University of California, Berkeley, and Ethan Kaplan, of the University of Maryland (http://mintne.ws/1jUoC3y), found a significantly positive impact of the entry of Fox News on the vote share of the Republican Party in the US. The conservative channel was introduced sequentially in the states of the US between October 1996 and November 2000. The gain to the Republicans owing to the introduction of Fox News varied from 0.4 to 0.7%. However, they add the caveat that the Fox News effect could very well be a temporary learning effect for the informed voters and may be permanent for only non-rational voters.
In a similar study, economist Dean Karlan and political scientist Alan Gerber from Yale University and Daniel Bergan, political scientist at the Michigan State University, conducted an experiment where they randomly assigned individuals in Washington DC to receive either a free subscription to the Washington Post or a free subscription to the Washington Times. They found that those who opted for Washington Post were more likely to vote for the Democrat candidate in the elections.
Access to independent media also has a significant impact on the electoral outcomes. Ruben Enikolopov, of the Barcelona Institute for Political Economy and Governance, and his colleagues found evidence (http://mintne.ws/1oeTvpy) that wherever independent TV channels existed in Russia, as opposed to regions where only a state-owned channel was accessible, the voter turnout decreased by 3.8 percentage points on an average; incumbent vote share fell by 8.9 percentage points; and the major opposition parties gained about 6.3 percentage points.
Gentzkow and others show, using data from US newspapers from 1869 to 1928 (http://mintne.ws/1hezazN), that incumbents did not affect press freedom. However, they also found substantial political influence on media in the states affected by the Reconstruction (The Reconstruction era was the post-civil war period of socio-political turmoil in the southern states of the US). In 1869, all southern states were governed by the Republicans.
However, with escalating racial tension, Democratic votes share kept increasing over the decade, and by 1877, all the state legislatures were under the Democrats. Each of these political parties required the support of favourable media to enact their policies. Gentzkow et al caution that, “The Reconstruction episode is a reminder that incumbent influence may play an important role when stakes are high and constraints are weakened.”
In Argentina, coverage of political scandals was found to be negatively associated with political advertisements, showed economists Rafael Di Tella, of the Harvard Business School, and Ignacio Franceschelli, of the Northwestern University, in a 2011 research paper. (http://mintne.ws/1nkf6Mf)
Political scientist James M.Snyder Jr, of the Massachusetts Institute of Technology and Riccardo Puglisi, economist at the Universit´e Libre de Bruxelles, studied the coverage of 35 scandals in 200 newspapers across the US and found that a Democratic-leaning newspaper published relatively more scandals involving a Republican politician. (http://mintne.ws/1xbXBTs).
Studies under various settings show that political influences can lead to media bias, and in turn help political parties reap electoral rewards. Careful empirics rather than mere name calling can help us understand the determinants as well as the consequences of media capture even in the Indian context. The quality of debate on critical issues of public policy is after all determined to a large extent by the extent of independence in the media.
Sumit Mishra is a research scholar at the Indira Gandhi Institute of Development Research, Mumbai.
Economics Express runs every week, and features interesting reads from the world of economics and finance.