Beyond Mano Dura: Can Information about Democratic Costs Shift Voter Preferences on Security Policies

The intervention consists of a short 30–50 second pre-recorded audio message delivered to respondents during a phone survey. There are two active treatments. Treatment 1 provides an abstract warning that mano dura security policies can concentrate executive power, erode checks and balances, and restrict civil liberties, and that such institutional losses are difficult to reverse. Treatment 2 delivers the same warning anchored in El Salvador's documented experience under President Bukele, including mass incarceration without due process, removal of Constitutional Court justices, enablement of indefinite reelection, and restrictions on press freedom. Both treatments acknowledge mano dura's potential short-term crime reduction before presenting democratic costs, maintaining a balanced framing. A control group receives neutral civic information encouraging informed voting.
The target group is voting-age Peruvians (18+) drawn from a nationally representative sample recruited via random digit dialing, with quotas across sex, income level, macroregion, and age.
The intended purpose is to test whether providing information about the democratic costs of mano dura security policies reduces willingness to vote for a mano dura candidate and shifts acceptance of security-democracy tradeoffs among Peruvian voters.

2026-05-20

2026-06-06

Willingness to vote for a mano dura candidate; acceptance of security-democracy tradeoffs.

Willingness to vote for a mano dura candidate is measured through a stated-preference scale in which respondents rate how likely they would be to support a presidential candidate who promises to fight crime with mano dura, states of emergency, and forceful police or military actions, even if those measures could weaken democratic controls such as judicial processes or freedom of the press. Acceptance of security-democracy tradeoffs is measured through a scale capturing respondents' agreement with the statement that they are willing to accept that some democratic institutions be weakened during a security crisis if that helps reduce crime.

Two exploratory outcomes are examined: belief updating on security policies; and minimum crime reduction threshold required to accept democracy-weakening measures.

Belief updating is measured through a categorical question asking respondents whether the information heard during the survey changed their opinion about security policies. The minimum crime reduction threshold is measured through a question asking respondents how many out of every 100 serious crimes a mano dura policy would need to prevent for them to support measures that weaken democratic institutions, on a 0–100 scale.

The study uses a three-arm randomized controlled trial embedded in a nationally representative computer-assisted telephone interview (CATI) survey. The three arms are: control (neutral civic information), Treatment 1 (abstract democratic warning), and Treatment 2 (concrete case anchored in El Salvador's experience under President Bukele).

The target sample is voting-age Peruvians (18+). The total sample size is 1,200 respondents, with equal allocation of 400 per arm. Randomization is conducted at the individual level, stratified by sex, income level, macroregion, and age. Participants are recruited via random digit dialing by Escucha Perú, a professional survey firm, with quotas ensuring national representativeness across the four stratification variables. The survey is fielded between Peru's first presidential round (April 2026) and the June 2026 runoff.

The full text (original in Spanish also available) of the three treatment scripts is as follows:
Control — Neutral electoral information:
""The runoff election is an opportunity for Peruvians to choose the country's direction. It is important to learn about candidates' proposals, compare their governing plans, and vote responsibly. An informed vote strengthens our democracy and ensures our decisions reflect the needs and values of all Peruvians. Remember: your vote is your voice.""
Treatment 1 — Abstract democratic warning:
""Ahead of the runoff election, it is important to be well informed. Candidates propose mano dura policies to address insecurity. These policies can reduce crime in the short term, but they frequently weaken democratic institutions: they concentrate power in the executive, erode checks and balances, and restrict civil liberties. Once these democratic guarantees are lost, they are very difficult to recover. When evaluating security proposals, it is important to consider not only crime reduction, but also the protection of our freedoms and the rule of law over the long term.""
Treatment 2 — Concrete case (El Salvador / Bukele):
""Ahead of the runoff election, it is important to be well informed. Candidates propose following El Salvador's 'Bukele model' to address insecurity. While Bukele's mano dura policies reduced crime rates, they also severely weakened Salvadoran democracy: his government imprisoned tens of thousands without due process, removed justices from the Constitutional Court, enabled his own indefinite reelection, and restricted press freedom. Although crime declined, democratic institutions were significantly eroded. When evaluating security proposals, it is important to consider not only crime reduction, but also these costs to our democracy.""

Not available

Randomization done in office by a computer

Individuals

No

none

1200 individuals, Peruvians over 18

Control: 400; Treatment 1 (abstract democratic warning): 400; Treatment 2 (concrete case): 400.