What four independent studies — and hundreds of screened papers — actually show about housing type, ownership, and neighborhood safety.
Real Housing Reform | realhousingreform.org | March 2026
Here's a claim you've heard in some form, probably from a city council presentation or a housing advocacy report: dense, walkable, mixed-use neighborhoods are safer. More people on the street. More eyes are watching. More community.
It's a compelling story. It's also not what the research says.
I spent time digging into the peer-reviewed literature on housing type, density, tenure, and property crime — pulling from four separate Consensus research reviews that screened over a thousand papers across the U.S., U.K., China, Sweden, Norway, Poland, South Korea, Australia, and Malaysia. What came back was consistent, cross-cultural, and pretty direct.
Neighborhoods dominated by detached, owner-occupied homes consistently record lower property crime than denser, renter-heavy areas. This finding holds across countries, methodologies, and decades of research.
That doesn't mean density causes crime. It's more complicated than that — and the complications are actually more damning for the pro-density crime argument than the headline number.
Let me walk you through what the evidence actually shows.
The Basic Finding: Density and Renting Are Both Crime Risk Factors
Start with the numbers. Research in Wuhan, China found that each unit increase in a neighborhood density index was associated with an 11.9% increase in residential burglary. Stockholm research using remote-sensing data found building density explained about a quarter of variation in burglary and theft rates. Similar patterns showed up in Poland and Southeast Asia.
Yue et al. (2022), Journal of Housing and the Built Environment; Ioannidis et al. (2024), Cartography and Geographic Information Science; Sikorski et al. (2024), Cities
On tenure, the pattern is equally consistent. British data found that renting was associated with at least a 40% increase in household crimes compared to owner-occupancy. Nordic building-level research found that privately owned buildings had lower crime rates than rentals. American research, controlling for other neighborhood factors, found that renter occupancy was a reliable, independent predictor of higher crime. Chinese city research found that more rental units meant more theft.
Farrall et al. (2016), British Journal of Criminology; Ioannidis et al. (2025), American Journal of Criminal Justice; Raleigh & Galster (2015), Journal of Urban Affairs; Xu et al. (2022), Habitat International
The Consensus research meter — drawing on 10 reviewed studies specifically on this question — found that 70% answered yes: higher density and lower homeownership correlate with increased property crime. 10% mixed. 20%, no significant effect.
That's not a fringe finding. That's a preponderance.
"But What About Social Cohesion, Active Management, and Mixed Use?"
Here's where the pro-density argument usually pivots. Yes, they say, density correlates with crime — but that's because those neighborhoods lack social cohesion, good management, and the right mix of uses. Get those right, and density is fine.
It's a reasonable theoretical argument. The problem is what happens when you actually look at where those three conditions show up in the real world.
I asked that question directly: are social cohesion, active management, and mixed use more favorable outside single-family neighborhoods — in the dense, renter-heavy environments where crime risk is elevated?
The research answer: only partially, and in the two most important cases, the opposite is true.
Social Cohesion: It's Stronger in Stable, LOWER-DENSITY NEIGHBORHOODS!
Research in Oslo found that higher neighborhood density and greater land-use mix were directly associated with lower social cohesion — even as they increased urban vitality and liveliness. Those two things moved in opposite directions. You can have a vibrant, busy neighborhood with lots of foot traffic and still have neighbors who don't know each other, don't trust each other, and don't intervene when something goes wrong.
Mouratidis & Poortinga (2020), Landscape and Urban Planning
Research in Shanghai found higher cohesion in lower-income but tight-knit, stable neighborhoods. Large UK cohort studies confirmed it's perceived cohesion — neighbors knowing, trusting, and helping each other — that matters for health and safety outcomes, and that's tied to residential stability, not housing form.
Miao et al. (2019), Social Science & Medicine; Rowley-Abel et al. (2025), Health & Place
Bottom line: Social cohesion is not automatically higher in dense or renter-heavy areas. It is often stronger in stable, lower-density, owner-occupied neighborhoods. The key variable is how long people have lived there and whether they intend to stay — not how tall the buildings are.
Place Management: Informal Control Is Just as Effective
The research distinguishes between formal management — superintendents, building staff, city services — and informal management: watchful neighbors, local norms, mutual accountability. Dense environments tend to have more formal management infrastructure. Single-family neighborhoods rely on informal systems.
Here's the thing: informal management works. Often very well. The research consistently shows that stable, owner-occupied areas with engaged neighbors produce fewer incidents than rental properties with absentee landlords, even when the rental properties have professional management on paper.
Eck (2019), Oxford Research Encyclopedia of Criminology; Rephann (2009), Annals of Regional Science
Bottom line: Management quality depends on ownership stability and engagement, not density. An absentee landlord managing 200 units is not equivalent to 200 homeowners each watching out for their own block.
Mixed Use: Present, But Not Reliably Protective
Mixed use is the one condition that genuinely does occur more outside single-family areas — by definition, single-family zones have low land-use mix. That part of the pro-density argument is accurate.
But the research is equally clear that mixed use can increase or decrease crime depending on context. When mixed use combines with concentrated social disadvantage, it correlates with more crime, not less. The commercial activity that adds vitality also adds foot traffic and opportunity. The direction of effect is not predictable without knowing what specific uses are present, how they're managed, and what the surrounding social conditions look like.
Wo (2019), Social Science Research; Zahnow (2018), City & Community; Kavaarpuo et al. (2024), Cities
Bottom line: Mixed use is present in dense areas but is not a reliable crime reducer. It requires the right composition of uses, active management, and social stability to function protectively — which brings us back to the same conditions that dense, high-turnover environments structurally undermine.
The Circular Problem No One Talks About
Here's what strikes me about this entire body of research. The three conditions most often cited as potential offsets to density-related crime risk — social cohesion, place management, and mixed use — are not reliably present in the environments where that offset is supposedly needed most.
Residential instability reduces social cohesion. High-turnover renter populations don't build the long-term neighbor relationships that produce collective efficacy. Absentee landlords reduce place management quality — and dense rental developments are structurally more likely to have absentee ownership. And mixed use, the one condition that does show up in dense areas, produces unpredictable crime effects that depend on exactly the social and management conditions that are already compromised.
The escape hatch from the density-crime finding requires conditions that the density-and-renting model itself makes harder to achieve.
That's not a policy argument. It's a structural observation from the data. And it's the kind of thing that should be front and center in any honest discussion about what we're actually building when we upzone and densify existing neighborhoods.
What the Research Does — and Doesn't — Establish
I want to be precise here, because this matters for how you use this information.
What the evidence supports:
What the evidence does not establish:
The point isn't that we should never build anything other than single-family homes. The point is that the research doesn't support the claim that density and renting are crime-neutral, and the evidence is consistent enough across enough different places that it deserves serious weight in policy discussions.
Why This Matters for Housing Policy
Washington State just passed rent control, mandated upzoning across urban growth areas, and pushed middle housing into every corner of its urban footprint. These policies will increase density and rental activity. That's their design.
None of the research I've reviewed suggests that it automatically makes neighborhoods more dangerous. But it does suggest that the social and management infrastructure required to keep crime low in those environments doesn't appear by default — and that the stable, owner-occupied neighborhoods being displaced in this process were quietly providing that infrastructure all along.
That's worth knowing. That's worth saying.
Sources: This post draws on four Consensus AI-powered research reviews and the peer-reviewed studies they reference. The full case study with complete citations is available at realhousingreform.org.
Real Housing Reform | realhousingreform.org | © 2026