Cohousing is an architectural and governance model that treats friendship formation as a design objective. Its central insight is simple: friendship requires repeated contact under conditions of low stakes, and the built environment can be intentionally structured to produce those conditions at residential scale. Rather than treating the house as a self-contained unit and the neighborhood as incidental, cohousing arranges private dwellings around shared spaces—common dining halls, shared gardens, workshops, children's play areas—that draw residents into regular, voluntary contact. The repeated encounter is engineered into the site plan. What might happen by accident in a dense urban neighborhood is made likely by design.
The model originated in Denmark in the late 1960s, where architect Jan Gudmand-Høyer published a manifesto in 1968 arguing that existing housing forms were producing social isolation that neither the nuclear family nor the traditional village community was any longer able to address. His proposal—medium-density clusters of private homes arranged around shared facilities—was implemented in several Danish projects in the early 1970s and spread across Scandinavia over the following decade. The term bofællesskab (living community) was coined, and the English translation "cohousing" entered the vocabulary when architects Kathryn McCamant and Charles Durrett brought the model to the United States in the late 1980s after studying Danish projects extensively.
What distinguishes cohousing from other forms of intentional community—communes, intentional communities organized around shared ideology, religious residential communities—is that it does not require ideological homogeneity or communal ownership. Residents own their individual units. Participation in shared life is structured but not mandatory. The design creates conditions for friendship; it does not compel it. This distinction matters for the model's scalability and for its social outcomes: cohousing produces higher rates of neighborly interaction, mutual aid, and reported social wellbeing than conventional residential development, but without the attrition rates and interpersonal intensity of more ideologically demanding communal models.
The research evidence on cohousing's social outcomes is consistent. Residents report significantly higher levels of neighbor interaction, more mutual aid (childcare, meal sharing, informal support during illness), lower loneliness, and stronger sense of community than matched comparison populations in conventional housing. Studies from Denmark, the Netherlands, the United Kingdom, and the United States find these effects robust across age groups, tenure types, and design variations. The effects are not magic—they are the predictable consequence of designing residential environments to produce the proximal, repeated, low-stakes encounters that friendship formation requires.
The barriers to cohousing at scale are primarily political and financial, not technical. The model requires land acquisition and development processes that are friendlier to resident-led or non-profit development than to conventional market developers, who have little commercial incentive to build something their buyers could in principle develop themselves. Planning frameworks that favor conventional subdivision development over the cluster arrangements cohousing requires add process friction. Financing structures that treat cohousing projects as non-standard add cost. In most jurisdictions, cohousing projects are developed by small groups of committed future residents who navigate a development process designed for large commercial developers—a mismatch that keeps production volumes small relative to demand.
The policy implication is clear. If cohousing produces measurably better social outcomes than conventional residential development, and if those outcomes have measurable public health and civic value, then the barriers to its production should be treated as policy problems requiring active correction. Several European countries—Denmark, the Netherlands, Germany—have modified planning and financing frameworks to enable cohousing development at larger scale. The results suggest that treating friendship formation as a housing policy goal, and removing the structural impediments to design models that serve it, is a tractable policy intervention with significant social benefit.