Do Public Protests Change Anything?

Public protests are an essential feature of democracy – or an occupational hazard, depending on your perspective. There is evidence that they have become more common over the last two decades and we are now living in a period of major upheaval, similar to the 1960s. But do they change anything? When people get out in the streets and blast airhorns and hold up placards, does public opinion shift in the protesters’ favour?

That’s what two economists sought to find out. Amory Gethin and Vincent Pons analysed data covering 110,000 protests in the U.S. since 2017. These data were taken from the crowd counting consortium – a volunteer project that collates publicly available figures on the size of political crowds. Of course, most protests are small, attracting only a few hundred participants. But some are much larger, with crowds in the tens or hundreds of thousands.

Table from ‘Social Movements and Public Opinion in the United States’.

Interestingly, a single issue – racism – accounted for 27% of all protests. And in fact, 84% of protests were for ‘liberal’ or Left-wing causes, while only 14% were for conservative or Right-wing causes. This huge disparity is presumably due in part to factors like conservatives being older and more likely to live in rural areas.

To assess the political impact of protests, the authors analysed data from Twitter, Google and several nationwide surveys. The surveys included questions on things like vote intention, presidential approval and the most important issue facing the country – allowing Gethin and Pons to see whether protests are associated with shifts in public opinion.

The authors began by comparing tweets, Google searches and measures of public opinion before and after the start of each protest movement. They then used a ‘difference-in-difference’ approach – comparing the change in tweets, Google searches and measures of public opinion in counties that saw more versus less protest activity. What did they find?

In short, the only movement that had a discernible political impact was Black Lives Matter. All the others led to spikes in tweets and Google searches but left no real mark on public opinion. This can be seen in the chart below, which shows results for vote intention (Trump versus someone else) and presidential approval. Only the George Floyd protests are reliably associated with shifts in public opinion.

Chart from ‘Social Movements and Public Opinion in the United States’.

It seems that if the goal is to affect some kind of political change, most public protests are a waste of time. And note: the argument that “if we don’t protest then only the other side’s message will get heard” doesn’t work. As mentioned, 84% of protests were for ‘liberal’ or Left-wing causes; one side is doing almost all the protesting.

Of course, individual protesters might have other goals like achieving a sense of belonging or taking part in a collective ritual – and for them, attending protests would not be a waste of time.

Gethin and Pons’s findings indicate that, with one notable exception, even large protests have surprisingly little impact on public opinion. Rather than attending protests, budding activists might be better off writing to their political representative or trying to win round friends and family.