Imposing curfews during episodes of civil unrest reduces serious violence more than it merely displaces disorder to other times or places

What's this about?

People disagree about whether curfews cut serious violence during unrest, or just move trouble to other times and places.

What supporters say

• Curfews keep people off the streets at night, so fewer chances for fights, looting, or clashes may arise.

What critics say

• Curfews may anger people, harm trust in leaders, and cause a strong public or police backlash.
• Research from many studies finds no clear drop in crime during curfew hours or across the whole day.
• Studies often fail to check if trouble moves to nearby places or happens later.
• Police checks during a curfew can bring officers and crowds into direct conflict.

How to read this

The number of points on each side does not show who is right; the strength of the proof matters more.

The bottom line

Curfews may make streets calmer for a short time, but strong proof that they cut total serious violence is lacking. The evidence leans against the claim, though these studies do not fully test short curfews during riots.

The claim that emergency curfews reduce serious violence during civil unrest more than they simply move it elsewhere is not strongly supported by current evidence. Curfews may calm streets in the short term, but the broader effect depends on enforcement, policing and how communities respond.

The case for

The main argument for curfews is straightforward: removing people from public spaces can reduce opportunities for violence. Fewer people on the streets may mean fewer spontaneous clashes, acts of looting and encounters between protesters, opponents and police during the hours when unrest is most intense. 1

That mechanism is plausible, particularly when violence is concentrated at night. But plausibility is not the same as proof that a curfew prevents more serious violence than it displaces. The National Institute of Justice rates juvenile curfew laws as ineffective at reducing youth crime. Although those laws are not the same as emergency curfews imposed during riots, the underlying idea is similar: limiting people’s access to public spaces should reduce offending.

The available evidence therefore gives the pro-curfew case a reasonable theory but little direct support. There is no strong evaluation in the material reviewed showing that curfews during episodes of civil unrest reduce total serious violence after accounting for what happens later or in nearby areas.

The case against

The strongest systematic evidence points the other way. A Campbell review of 12 quantitative studies found that juvenile curfews were ineffective overall. During restricted hours, the average effect was slightly positive rather than a reduction in crime; across all hours, the effect was approximately zero. That pattern is important because it suggests that a quieter curfew period may not mean less total offending. 2 (see Figure 1)

This research does not directly test short-term emergency curfews during riots, so it cannot settle the question by itself. Still, it is the strongest systematic evidence available here, and it argues against treating curfews as a dependable way to reduce violence rather than shift it.

Enforcement can also create new confrontations. Reports from the Netherlands described clashes, looting and arrests in several cities after a COVID-19 nighttime curfew was introduced, including confrontations involving officers enforcing the restriction. These reports do not prove that the curfew caused the unrest, but they show how enforcement can become another focus for disorder instead of simply removing opportunities for it. 3

Curfews may also trigger political and security backlash, especially where people already distrust authorities. A peer-reviewed study of curfews in Turkey linked their introduction to adverse political reactions and increased rebel attacks. The setting differs substantially from a short, targeted curfew in a city experiencing unrest, so the findings do not show that backlash is inevitable. They do, however, provide evidence that restrictions can sometimes worsen the wider conflict. 4 (see Figure 2)

A further problem is that displacement is rarely measured properly. Studies may record what happens during curfew hours, but they often do not track violence on following days or in neighboring locations with equal care. The systematic review compares restricted hours with all hours, but emergency-unrest reporting generally lacks this broader comparison. 5

It is also difficult to isolate the curfew’s independent effect. Authorities usually impose one after violence has already escalated, often alongside heavier policing and community intervention. Research shows that governments with recent violence or human-rights abuses are more likely to impose curfews, making simple before-and-after comparisons unreliable. In Amsterdam, for example, de-escalation was attributed partly to informal guardians and community actors who intervened directly with potential rioters. Curfews may therefore be one part of a larger operation, rather than the cause of any improvement.

The bottom line

The evidence favours the case against the claim, with high confidence in the overall direction but lower confidence about any particular emergency curfew. The strongest systematic research does not show a reliable reduction in total violence, while reports and studies identify possible enforcement confrontations and political backlash.

The evidence is not a direct, definitive test of emergency curfews during civil unrest. No strong evaluation measures serious violence during restricted hours, later periods and nearby places against a credible counterfactual. But the pro side is mainly supported by a plausible mechanism, while the opposing case has stronger systematic and observational backing. Curfews may sometimes help, but current evidence does not show that they generally reduce serious violence more than they displace it.