Adding light to public space is a foundational idea in situational crime prevention, which works on the environment rather than on the offender, resting on the premise that a brighter place is a place where people can see one another and where an offender runs a greater risk of being seen. Chile ran a national programme replacing public street lighting, but judging whether it worked is difficult, because no one randomly assigned which municipalities received new lights. The authors therefore matched municipalities that received the upgrade with untreated municipalities of similar socio-economic profile, simulating a randomised trial from observational data, comparing 101 treated municipalities against 244 untreated ones between 2015 and 2019. The results do not all point the same way: several crimes and many incivilities fell, but violent robbery did not.
Research Objectives
- Evaluate the impact of the Public Lighting Replacement Programme on the reduction of crimes and incivilities across Chilean municipalities.
- Separate the effect of the lighting upgrade from socio-economic differences between municipalities by simulating a randomised trial from observational data.
- Test whether the effect of the programme varies by type of crime and type of incivility.
- Assess whether improving urban infrastructure alone is sufficient, or whether complementary measures are required.
Methodology
- Data came from five government sources: the Centre for the Study and Analysis of Crime (CEAD), the Undersecretariat of Energy, the National Statistics Institute (INE), the National Congress Library (BCN) and the National Municipal Information System (SINIM), after a process of standardisation and cross-validation.
- The municipality was the spatial unit of analysis, being the country's political and administrative division. The study period ran from 2015 to 2019, with 2015 as the pre-treatment year, streetlight replacement carried out through 2018, and 2019 as the post-treatment year. Later years were excluded because COVID-19 restrictions distorted normal crime patterns.
- 101 municipalities received the upgrade, 12 in 2016, 72 in 2017 and 17 in 2018, while 244 municipalities that did not upgrade served as the comparison group.
- Four crime variables were measured: theft, violent robbery, surprise robbery and vehicle theft. Six incivility variables were measured: street vending, alcohol consumption, property damage, public disorder, public drunkenness and public brawls. All refer to activity in public space and may therefore be influenced by lighting conditions.
- Propensity score methods were used, calculating the probability that a municipality would enter the programme given its observed characteristics, then matching treated municipalities with untreated ones of similar score, controlling for more than twenty socio-economic variables including population density, the poverty index, school dropout rates and the rural share of the population.
- Effects were estimated by two methods, stratification into 5 and into 10 strata, and matching, working with rates per 10,000 inhabitants so that municipality size would not distort the results.
Key Findings
- Estimates were similar in sign and magnitude across every method used, which the authors take as a sign of coherence in the findings.
- Effects were negative, meaning a decrease, for all crimes and incivilities except violent robbery, which rose slightly under every technique employed.
- Among crimes, only surprise robbery fell significantly, and aggregating all crime categories together produced no significant effect.
- Among incivilities, two showed no change, property damage and public drunkenness, while the other four fell; aggregated, incivilities showed a significant reduction of between 9.52 and 14.76 per cent depending on the technique.
- The size of the reductions varied widely between categories, from 10.98 per cent for public brawls under the 5-strata estimate to 58.12 per cent for public disorder under optimal matching across the whole sample. Street vending showed the largest overall reductions, with no technique returning less than 40 per cent.
Recommendations
- Improved lighting helps but is not sufficient for violent crime: the authors report that violent robbery showed no significant reduction, so interventions beyond infrastructure are needed.
- The complementary measures the authors name are targeted police patrolling and investment in community programmes.
- Urban infrastructure improvement should be treated as one tool within situational crime prevention, under an integrated urban security strategy rather than as a standalone measure.
- Future research should examine whether factors such as targeted patrols and community programmes can enhance the deterrent effect of public lighting.
- The authors set out three limitations of their own: the observational design prevents establishing a definitive causal relationship; reliance on police records and citizen reports carries a risk of underreporting that may vary across municipalities and crime types; and the analysis does not distinguish between installing new lighting and improving existing systems.
Key Takeaways
- Lighting does not reduce every crime type equally. The clearest effects appear on public-space incivilities, while among crimes only surprise robbery fell significantly and violent robbery did not fall at all.
- The authors describe their evidence as correlational rather than causal, since an observational study cannot control for every unobserved variable.
- For Thailand the point worth carrying forward is that adding street lighting may do more for public-space disorder than for violent crime. This connection is our own; the study covers Chile only and uses Chilean offence categories.
References
Cadena-Urzúa, P., Guardiola, J., & Montes, F. (2025). The impact of public lighting improvement on crimes and incivilities reduction in Chile. Frontiers in Sociology, 10, Article 1541359. https://doi.org/10.3389/fsoc.2025.1541359
Full text (Open Access): https://www.frontiersin.org/articles/10.3389/fsoc.2025.1541359/full
