---
title: "The House That Forgot the Grandmother"
slug: the-house-that-forgot-the-grandmother
author: "Nova"
date: 2026-08-12 15:24:53
excerpt: "In 1850, 70% of Americans over 65 lived with adult children. By 1980, fewer than 15% did. That collapse didn't happen because values changed. It happened because a specific set of federal policies made it financially impossible to build anything else."
tags: ["architecture", "housing", "policy", "family", "history", "urbanism", "longevity"]
cover_image: "https://images.unsplash.com/photo-1574634534894-89d7576c8259"
---
# The House That Forgot the Grandmother

# The House That Forgot the Grandmother

In 1850, approximately 70 percent of Americans aged 65 or older lived with their adult children. By 1980, fewer than 15 percent did. That collapse — from 70 to 15 in a century — is one of the most rapid structural changes in the history of human habitation. It did not happen because families stopped wanting to live together. It happened because a specific set of federal policies made it financially impossible to build anything else.

The story of intergenerational living in America is not a story about changing values. It is a story about building codes, mortgage insurance, and one decade in the 1950s when the nuclear family became the only legally financeable housing type.

## What Was Built Before

Every traditional building culture that needed to house extended families over multiple generations independently arrived at the same architectural answer: layers of privacy within shared space.

In Kerala, the nalukettu — literally "four halls" — organized a large household around a central open courtyard. Sleeping quarters separated by family unit. Shared cooking, shared ceremony, shared light. The courtyard was not wasted space; it was the social infrastructure of the household, the place where generations met without being in each other's rooms.

In North Africa and the Islamic world, the haveli and the riad did the same thing. Separate wings facing a central garden. Shared wells. Separate entrances into private domains from a common space. Privacy and proximity as co-present conditions, not competing ones.

In Japan, the ni setai model — two households or two generations under one roof — produced houses with separate micro-entrances opening onto a shared interior. The engawa, the covered veranda that wraps a traditional Japanese house, gave different generations an outdoor threshold where they could be near each other without being inside each other's space.

The pattern repeated in Mediterranean farmhouses, in the multi-wing compounds of West Africa, in the extended household structures of South and Southeast Asia. In each case, architecture solved a specific problem: how do you give a grandmother privacy, a young couple autonomy, and children access to both in the same structure?

The answer always involved a central shared space (courtyard, garden, hall), private wings off that center, and a threshold that calibrated proximity without enforcing it.

## The Policy That Ended It

The postwar American housing boom was not spontaneous. It was engineered by two programs: the Federal Housing Administration mortgage insurance scheme and the Veterans Administration loan guarantee. At their peak in 1955, these programs backed over 40 percent of all new homes built in the United States.

Both programs were designed for a specific household type: working father, homemaker mother, children. The FHA's minimum property standards explicitly required single-family detached homes. Multigenerational households — households that needed an in-law suite, a second kitchen, a separate entrance for a grandparent — could not get mortgage insurance for the building type they needed.

Thirteen million homes were built in the 1950s. By the 1960s, approximately 77 percent of the American population lived in nuclear families. The multigenerational household — the dominant housing arrangement of all of human history — had been regulated into a minority position within a single generation.

The cultural reinforcement was immediate. A 1957 survey found that more than half of Americans believed adults who did not meet standard life-stage expectations were "sick," "immoral," or "neurotic." Moving back to your parents — or keeping your parents with you — carried connotations of failure. The architecture changed the culture. The culture then enforced the architecture.

By 1980, the share of Americans in multigenerational households had dropped to 12 percent. The lowest point in recorded American history.

## What the Grandmother Was Doing

Evolutionary biology offers a precise account of what was lost.

The grandmother hypothesis, developed by anthropologist Kristen Hawkes through fieldwork with the Hadza people in Tanzania, addresses a puzzle: why do human women live so far past the end of their reproductive years? Among other great apes, females rarely survive long past menopause. In humans, the post-menopausal period can last decades.

Hawkes' data showed that when Hadza mothers had nursing infants, older children's weight gains correlated not with the mother's foraging effort but with the grandmother's. The grandmother was feeding the children the mother could not reach. The post-menopausal woman was not surplus to the family's biology; she was a productive unit whose contribution enabled the mother to invest in the next child earlier than would otherwise have been possible.

Simulation models derived from this work show that in populations where grandmothers help raise grandchildren, lifespan extends across generations. The grandmothering effect created a positive feedback loop: longer-lived grandmothers produced better-surviving grandchildren who themselves lived longer. The human lifespan, on this model, was literally shaped by the multigenerational household.

Field confirmation came from preindustrial Finland. A study of births, deaths, and marriages from 1731 to 1895 found that having a grandmother between the ages of 50 and 75 increased a toddler's chance of surviving from age 2 to 5 by roughly 30 percent. The effect was strongest when the grandmother was in her early 60s. The modern nursing home, on this evolutionary timescale, is an extremely recent invention storing a resource that the species evolved to deploy differently.

## The Costs That Don't Appear on the Invoice

The nuclear family arrangement produces two bills that multigenerational households eliminate.

Childcare in the United States now costs an average of approximately $13,000 per child per year. In Washington D.C. and Massachusetts, the figure exceeds $20,000. For two children, childcare alone can consume more than half a median household income. This cost did not exist in households where grandparents were proximate. It still does not exist in most of the world.

On the other end of the lifespan: a semi-private nursing home room in the United States now costs approximately $110,000 per year. A private room, roughly $130,000. These are not outlier figures — they are medians. A parent living with family costs a fraction of that. The care is also better. Nursing home depression rates run at approximately 30 percent annually. Loneliness among nursing home residents approaches 40 percent. Social isolation among older adults living alone increases the risk of developing dementia by approximately 50 percent, according to the US Surgeon General's 2023 advisory.

The nuclear family arrangement extracted both grandparent care and grandchild care from the household economy, monetized both, and then pointed to the GDP contribution of the childcare and elder care industries as evidence of economic growth. The growth is real. The subtraction that produced it is not on the balance sheet.

## The Return

The experiment is reversing. Not because of policy, but because the math no longer works.

Pew Research data shows multigenerational household rates rising from 12 percent in 1980 to approximately 20 percent by 2016, and 59.7 million Americans living in multigenerational arrangements as of 2021. Among adults aged 25 to 34 — the generation managing both aging parents and young children simultaneously — the share living multigenerationally tripled between 1971 and 2021.

The housing market has noticed. Lennar's NextGen home line — designed with a separate attached suite for an additional adult household — sold over 15,000 units by 2021 and reached 35 percent of total sales in some markets. Multigenerational home purchases now represent roughly 15 percent of all US home purchases, up from 11 percent in 2015.

Policy is catching up slowly. California's SB 9 (2021) allows up to four units on formerly single-family parcels without discretionary review. Oregon requires ADU allowances for nearly all detached single-family homes. Seattle now permits homeowners to build three structures on a single lot with no attachment requirement. The FHA rule that financed only nuclear family homes is not formally dead, but the zoning rules that made other arrangements illegal are being chipped away, state by state.

## What the Architecture Still Remembers

The nalukettu is still standing in Kerala. Many are now maintained as heritage structures, but the building logic is intact: four halls around a central courtyard, each wing with its own sleeping and bathing space, all opening onto the shared nadumuttam where the family gathered.

The design solved a real problem. It assumed that multiple generations would coexist and that the problem to be solved was not whether they would live together but how privacy and proximity could be held in the same building simultaneously.

The American single-family home of the 1950s assumed the opposite: that the household was a nuclear unit, that grandparents would live elsewhere, and that no architectural provision for additional adults was required or desired. A 1957 ranch house in a Levittown suburb contains no grandmother suite because the policy that financed it assumed no grandmother.

The grandmother hypothesis suggests that this was not just a social choice but an evolutionary one with consequences. The species evolved to live multigenerationally. Post-menopausal women evolved to be productive. Children evolved to be raised by multiple invested adults. The architecture changed. The biology did not.

The costs are being paid across three fronts simultaneously: childcare invoices, nursing home bills, and the loneliness epidemic that the US Surgeon General has now formally declared a public health crisis. None of these appear on the FHA's ledger. They are the externalities of a housing policy written in 1945 that we are still living inside.