Abstract
Objective
The middle class has historically played a decisive role in mass movements and rebellions against the status quo, including in both peaceful and violent protests that spread throughout the Middle East and North Africa following the events of the Arab Spring. Yet, we know little about why the middle class tends to be over-represented in protest movements, or why that protest takes nonviolent versus violent forms.
Methods
We develop a model in which protest constitutes a “costly” signal of discontent, where violent actions are more costly than peaceful demonstrations and where the cost of action increases with individual income. Under such assumptions, the upper-middle class will engage in nonviolent protest, the lower middle class will support violence against the government, while the poorest and richest will abstain from opposition activities.
Results
Survey data from the World Values Survey and the Gallup World Poll across 20 Arab countries supports these hypotheses. Using both parametric and semi-parametric analyses, we find that participation in nonviolent protests and general strikes rises from the first (lowest) and peaks at the fourth income quintile, then declines thereafter. Meanwhile, support for political violence rises sharply between the first and second quintiles, falling for individuals in the upper quintiles. Our findings are robust to broader measures of wealth and status, as well as to corrections for regime type, levels of regime support, and joint determination of anti-regime behaviors.
Conclusions
These findings shed light on why certain socioeconomic groups engage in anti-governmental behaviors, while others do not, and suggest that income subgroups within the middle class may choose different modalities of protest.
