Auditing Business Safety Inspectors

Last registered on July 18, 2022


Trial Information

General Information

Auditing Business Safety Inspectors
Initial registration date
July 15, 2022

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
July 18, 2022, 9:23 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.



Primary Investigator

Universidad del Pacifico

Other Primary Investigator(s)

PI Affiliation
World Bank
PI Affiliation
World Bank
PI Affiliation
UC Berkeley
PI Affiliation
World Bank

Additional Trial Information

On going
Start date
End date
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
In collaboration with the Ministry of Housing and up to five municipalities in Peru, we will conduct an experiment to evaluate whether variations in the implementation of the building safety-regulation can improve state's capacity and safety and reduce leakages. Most firms require multiple visits to obtain the safety certificate and start operations, creating inefficiencies and opportunities for rent-seeking. Constraints identified in this system include high discretion in the application of the regulation and low capacity to enforce the regulation by municipalities in charge of the implementation. We are rolling out an electronic inspection system and will assess the impact of increasing inspector accountability through random audits.
External Link(s)

Registration Citation

Barron, Manuel et al. 2022. "Auditing Business Safety Inspectors." AEA RCT Registry. July 18.
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details


Business safety inspections in Peru are conducted by municipal inspectors who have little or no oversight. Together with a vague regulation, this creates room for rent-seeking and gives inspectors little incentives to exert effort. In collaboration with up to three municipalities in Lima, we are setting up an auditing system where a highly trained third-party auditor visits recently inspected firms and double-checks a random subset of inspection items. One treatment arm has a high audit rate (~70%) and the other a low audit rate (~20%).
Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
Our primary outcomes are discrepancies between the inspector and the auditor: total, and split between number of NA, comply, and does not comply; discrepancy in inspection outcome (pass/fail); duration of the inspection.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Inspectors are randomly assigned to High or Low audit intensity arms, which will determine if 70 or 20% of their inspections will be audited. The trial will run for 6 months.
Experimental Design Details
Randomization Method
Randomization done in office by a computer.
Randomization Unit
The inspector.
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
We expect to have 100 inspectors. This number depends on the municipalities that agree to participate in the study. Some are larger, with 60 inspectors, and others are small, with fewer than 10 inspectors.
Sample size: planned number of observations
The unit of observation is the inspection. If each inspector conducts 4 inspections per day, there would be 120 visits on average over a 6-week period. Thus, with 100 inspectors we would have 12,000 observations.
Sample size (or number of clusters) by treatment arms
50 inspectors in treatment, 50 inspectors in control arm.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
The minimum detectable effect size for our main outcomes is 0.15 standard deviations. Assumptions: Confidence level = 0.95 Power = 0.80 Number of inspectors per treatment arm = 41 In case we do not reach our expected value of 50 inspectors per treatment arm, we would be powered with 82 inspectors. Correlation between outcomes of the same inspector = 0.05 We consider that the correlation between outcomes of the same inspector will be low given the diversity of businesses each inspector usually visits. Stata code local r1 = 0.05 local mean1 = 0 local mean2 = 0.15 local sd1 = 1 local sd2 = 1 local ratio = 1 local pre = 0 local post = 120 sampsi `mean1' `mean2', sd1(`sd1') sd2(`sd2') ratio(`ratio') pre(`pre') post(`post') r1(`r1') power(0.80)

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials