Identity politics — a phrase that defines our times. The rise of right-wing populists in countries as different as India, the US and Brazil, has increased the urgency with which we seek to understand how race or ethnicity matter to political behavior. The steady stream of articles on “what do Trump voters want” in the New York Times are a testament to that.
Worryingly, the tenure of leaders like Donald Trump or Narendra Modi has been associated with a rise in hate crimes –- be it against Jews and Asians in the US, Muslims in India, or the January 6th attacks on the Capitol in Washington D.C.. These events have raised pressing questions — what is the role of ethnic or racial propaganda in shaping politics and violence and how do we study it?
Studying the impact of propaganda can be tricky. First, we have to devise a way to measure people’s exposure to an idea at a fine-grained geographic level. Second, we need to ensure that the exposure was not a consequence of picking places where people particularly susceptible to the idea could be found i.e the dreaded endogeneity problem. Third, given the explosion in communication, transportation and media technology, it is hard to study the marginal effect of an idea on people’s behavior given all the media people have already consumed in their lifetimes. And finally, even if we were to solve these challenges, we have the ultimate one – how do we convincingly show a link between propaganda and specific political outcomes?
In a previous post, I discussed the challenges we face in studying the legacies of historic racial/ethnic cleavages on contemporary politics. But historic studies of identity are useful not just to understand the enduring fractures of the past, they also give us clues about how to study the present.
Three working papers in historical political economy that probe the link between propaganda, identity and politics have come to our rescue.
The Birth of a Nation and the KKK
The first, by Desmond Ang reads like a taught thriller and studies the effects of a cinematic blockbuster called the “The Birth of a Nation” that opened in 1915. The movie told a fictional tale about the founding of the Ku Klux Klan during Reconstruction.
Around 10 million Americans – roughly one-fifth of the adult white population – watched the movie in its first two years. It was in such high demand that movie goers paid as much as forty times the going rate for a movie ticket. The Birth of a Nation remains one of the highest grossing movies of all time.
Ang puts to use several peculiarities of this era. First, only 60% of American counties had a movie theater at the time of the movie’s release. Second, because the movie was disseminated through an extended 5-year roadshow, counties often had to wait for months if not years to see the movie. Third, the KKK was largely a spent force and almost non-existent at the time of the movie’s release
Ang constructs a dataset on the date and location of the roadshow using local newspapers and trade reports. He shows that the release of the film led to sharp spikes in racial violence in the short run (see Figure 1). Counties were five times more likely to experience a lynching during the month of the movie’s arrival. He also finds increases in race riots in following the movie’s showing.
He uses the presence of a movie theater in 1914, the year prior to the movie’s opening in 1915, as an instrument for the county receiving the movie. He finds that the screening of the film increased a county’s likelihood of having a KKK chapter (“Klavern”) in 1930 by a whopping 60 percentage points! These effects persisted over a century – roadshow counties were more likely to contain a Klavern in the 1960s and 2000s and greater numbers of white supremacist groups and anti-minority hate crimes today.
Father Coughlin and Antisemitism
The second paper by Tianyi Wang studies the impact of America’s first populist radio personality, Father Charles Coughlin. A Roman Catholic priest, Father Coughlin began his radio career in the 1920s and grew to amass a radio audience of 30 million by the mid-1930s. Known as the “Radio Messiah”, his radio broadcasts blended economic populism, antisemitism, and fascist sympathies.
Wang studies the impact of exposure to Father Coughlin’s radio program on voting outcomes in the 1936 presidential election. To do so he collects unique data on the location and technical details of Coughlin’s radio transmitters in 1936. He does this so he can predict the strength of signals intercepted by radios in different US counties – a measure of the intensity of listeners.
To counter concerns of endogeneity, he utilizes the variation in Coughlin’s signal strength resulting from topographic factors such as hills. By regressing the voting outcomes in the 1936 Presidential elections on actual signal strength while controlling for a hypothetical signal strength that would exist if there were no geographic features obstacles, he is able to identify the residual variation in signal strength. This, he argues, is likely to be uncorrelated with past voting outcomes.
He finds that a one standard deviation increase in exposure to Father Coughlin’s radio program was associated with a reduction in the Democratic Party’s vote share by about 3.8 percentage points (Figure 2).
He also examines the effects of Father Coughlin’s antisemitism on the proliferation of pro-Nazi attitudes in the US. Using FBI records, he traces all cities with a local branch of the pro-Nazi German-American Bund in 1940. He finds that cities with a one standard deviation increase in radio exposure to Father Coughlin in the late 1930s were 9 percentage points more likely to have a branch of the pro-Nazi German-American Bund.
Rath Yatra and the rise of the BJP
In a previous post, I argued that the rise of the hindu nationalist Bharatiya Janata Party in India in the early 1990s should be attributed to historic caste inequalities at the local level. The third paper by David Blakeslee argues that the pro-Hindu propaganda efforts of the party boosted its electoral fortunes.
The paper focused on the Ram Rath Yatra campaign of 1990 where the party leader, L.K. Advani traveled close to 10,00 kilometers in a truck decorated to look like the chariot of the hindu god Ram. The principle cause of the campaign was to demand the demolition of a mosque and the construction of a temple in what the party claimed was the birth place of Ram in Ayodhya in the North of India. The campaign content was rich in religious symbolism, Hindu nationalist ideology and anti-Muslim demagoguery. The photo above shows a young Narendra Modi in the truck.
Blakeslee utilizes two techniques to study the impact of the campaign on the Indian parliamentary elections of 1991. First, the Yatra aimed to travel through big cities in order to make the most impact, suggesting the cities were chosen not for their close competitiveness but rather for their urban density and the BJP’s traditional urban bases. Second, in traveling between cities, the Yatra passed through a number of electoral constituencies that were incidental to the campaign’s main focus on the cities.
Blakeslee finds that electoral constituencies through which the yatra passed experienced a 5 percentage point increase in the BJP’s vote share. The results are robust to restricting the sample to only the constituencies that were on the route between cities. He argues that the exposure of these constituencies to the propaganda was largely devoid of the strategic thinking that might have shaped the yatra’s route.
Together these authors exhort us to take seriously the power of speech and the consequential ways in which they shape populist and racist politics. All three papers, while incredibly rich in the data they deploy and the empirical strategies they use, leave us wanting more to understand the psychological, sociological and political mechanisms through which propaganda spreads and sticks on such a large scale. History as they say is destined to repeat itself and we need strategies to meet it when it predictably does.