Sage Journals: Discover world-class research

Abstract

Chinese

Home sharing, particularly via online platforms, is becoming a mainstream housing strategy as social processes evolve and housing costs rise. Recent research has studied shared rentals as a modality for students and kin-based households, as one strategy among diversifying pathways to housing and as a social phenomenon. However, we still know little about whether it actually creates opportunities for home seekers in unaffordable markets. Analysing online rental listings in Los Angeles, we find that shared rentals are both more affordable and more widely available across diverse neighbourhoods than traditional whole-unit rentals. Shared rentals have historically been understudied due to their limited data trail, but they offer important entryways into unaffordable markets. We argue for shared housing research to shift its traditional focus away from students and young adults and towards a broader exploration of the diverse populations that may benefit from or depend on shared housing.

Keywords

affordability Craigslist rental housing housing search shared housing

Introduction

Shared living is undergoing a renaissance. As housing costs rise and employment, education and household formation evolve, home sharing – once considered the domain of students or a kin-based safety net – is becoming a mainstream strategy around the world (Arundel and Ronald, 2016; Druta et al., 2021; Heath et al., 2017; Kenyon and Heath, 2001; Maalsen, 2020; Wiemers, 2014). Stimulated by digital platforms acting as market-making intermediaries, medium- and long-term shared renting now occurs not only between friends and family but increasingly also among strangers (Maalsen, 2020; Parkinson et al., 2021). Economic need often motivates sharing but it can also be a lifestyle choice (Harten, 2024; Heath et al., 2017; Hilder et al., 2018; McNamara and Connell, 2007; Zhang and Gurran, 2021). Private developers are also capitalising on home sharing’s growth and diversity. Branded as ‘co-living’, professionally managed shared housing has appeared in global cities like San Francisco, London, Singapore, Tokyo, Hong Kong and Mumbai (Bergan et al., 2021; Chen, 2022; Corfe, 2019; Devidayal, 2022; Druta and Ronald, 2021). In turn, local and national governments have begun providing and regulating shared housing (Green and McCarthy, 2015; Margolies, 2020; Ortega-Alcázar and Wilkinson, 2021).

The US Census Bureau’s 2023 Current Population Survey shows that there are over 9 million non-family households containing two or more members. Forty-six per cent of non-family households whose householder is under 30 years old contain two or more members, as do about 29% of those whose householder is between 30 and 39 years old. Yet, despite home sharing’s popularity, we still know little about how well it works as an affordable housing strategy. This article investigates what kinds of neighbourhoods are made available online through shared housing, and at what price points. It examines shared housing privately organised through Craigslist, one of North America’s largest websites for classified advertisements. Previous work has discussed sharing broadly as one strategy among diversifying housing pathways (Clapham, 2002; Fuster et al., 2019) or qualitatively focusing on shared household practices and sharing as social phenomena (Clark et al., 2017; Heath et al., 2017; McNamara and Connell, 2007). We advance this literature to analyse and theorise shared renting as an important housing submarket.

We collect online rental listings from Los Angeles County, California (‘Los Angeles’ hereafter), home to a diverse, majority-renter population and one of the less-affordable housing markets in the USA (Zhu et al., 2021). Then we compare the statistical and spatial distributions of both shared and full-unit small apartment listings in the context of local affordability thresholds, and analyse language about rent and affordability in listing titles. We find that rooms for rent are less expensive than small apartments for rent – thus offering more affordable accommodation than traditional rental listings – and that affordability features more prominently in shared housing advertisements. We also find that room listings are more widely available in different kinds of neighbourhoods than small apartment listings are, and that they in turn represent an important, if less formal, affordability option for accessing housing in unaffordable markets.

A better understanding of diverse housing strategies can reshape our knowledge of housing market dynamics. We argue that shared housing tends to be affordable to a more diverse range of incomes, and, importantly, may be the only affordable housing option for many single, low-income renters trying to access unaffordable markets such as that of Los Angeles. We call for a shift in researchers’ focus away from sharing as a niche strategy, and towards a broader inquiry into the diverse populations that may benefit from or depend on shared rental housing.

This article is organised as follows. The next section reviews the literature on shared housing, its digital mediation and the analysis of housing market phenomena using online listings. Then, we detail our data collection, cleaning and analysis. Next, we present the results of this analysis. Finally, we consider these findings against the broader literature, discuss insights for practitioners and lay out a research agenda to better understand this important housing phenomenon.

Background

Shared living has a long history but its meaning has changed over time. In the early 20th century, households were often flexible, expanding and contracting as needed. Housing was frequently multi-generational and often included extended family, live-in staff or boarders (Milan, 2000). Over time, social processes of industrialisation and urbanisation – including an emphasis on nuclear family housing – have produced decades of shrinking households and a proliferation of single-person households, particularly in the post-war Global North (Bradbury et al., 2014).¹ In this context, shared living came to be viewed as a niche strategy, often associated with students (Rugg, 1999) or ‘doubling up’ among low-income and immigrant communities (Myers et al., 1996; Pilkauskas et al., 2014; Wiemers, 2014).

Researchers have recently begun documenting a proliferation of sharing arrangements that go beyond existing networks of friends and family and extend past young adulthood (Aeby and Heath, 2020; Druta et al., 2021; Maalsen, 2020). Among these, privately organised sharing has received the most attention. Private sharing still includes student sharing, especially among the growing number of international students (Nasreen and Ruming, 2019, 2021), but also intergenerational sharing (Arundel and Ronald, 2016; Wiemers, 2014), intentional sharing (where sharing is part of an alternative lifestyle; Grundström, 2021; Hilder et al., 2018) and functional sharing (where sharing is purely a means to an end, typically in the context of low incomes and intense affordability pressures; Gurran et al., 2022; Harten et al., 2021; Ortega-Alcázar and Wilkinson, 2021).

These sharing arrangements exist in a complicated legal landscape with ramifications for both public regulations and private contracts. In California, for instance, the law distinguishes tenants – anyone who has an oral or written rental agreement with the landlord/owner – and occupants – anyone who resides in the rental property. Shared units can thus house multiple tenants or they can house a tenant and occupant(s), if the tenant opens the unit to others via subleasing. On the private contractual level, legal issues may arise when the terms of the rental agreement preclude subleases or require advance permission. Where home sharers are co-tenants, the vacating of one party can cause legal contentions. In addition, illegal subdivisions for unit sharing may violate health and safety codes (California Department of Real Estate, 2022).²

In addition to privately organized sharing, a growing market offers professionally managed sharing: ‘co-living’ branding commodifies sharing and markets it to single young professionals, promising high-end facilities, central locations and a community of like-minded residents in exchange for small units and above-market rents (Bergan et al., 2021; Druta and Ronald, 2021). As an attractive investment opportunity, co-living developments have sprouted in global cities around the world, frequently fostered by local governments eager to add new arrows to their quiver of efforts addressing the global housing affordability crisis (Chen, 2022; Corfe, 2019; Devidayal, 2022; Margolies, 2020).

What explains this recent surge of interest in home sharing? In the USA, just two decades ago, policymakers and researchers alike largely regarded shared renting as an inappropriate, unacceptable way of living (Ahrentzen, 2003). Sharing, especially with strangers, was at odds with strong Western cultural norms connecting the nuclear family, home and homeownership (Bate, 2018; Gorman-Murray, 2007). While shared living, especially intergenerational, kin-based arrangements, has long played an important role in other parts of the world, such as Southern Europe and Asia, the rise of shared renting in traditional homeownership societies such as the USA is a relatively recent phenomenon (Arundel and Ronald, 2016).

One strand of literature situates sharing in the context of an overall shift in housing inequality and worsening housing and labour precarity. These scholars see sharing as a coping mechanism – a ‘tenure of last resort’– for those needing affordable accommodation in unaffordable markets. Together with prolonged or intermittent parental co-residence, the rise of renting and delayed homeownership, these works frame sharing as an adaptive strategy driven more by need than by choice (Arundel and Ronald, 2016; Heath et al., 2017; Kim et al., 2020). Other scholars focus on young adults to document an overall shift in their housing careers. In this context, sharing is but one step in more complex, protracted and often precarious housing pathways that characterise contemporary urban living for this demographic (Clapham, 2002; Kenyon and Heath, 2001). Beyond investigating sharing as a housing choice, researchers have also studied the social dimension of sharing. Focusing on shared household practices, this literature emphasises the intimacy of sharing a home, which requires careful negotiations and offers a source of both tension and material and emotional support (Clark et al., 2017; Heath et al., 2017; Maalsen, 2020; McNamara & Connell, 2007; Steinführer and Haase, 2009; Waldron, 2024).

Like other housing processes, the matching for shared housing among unrelated individuals today occurs mostly online. Digital platforms, increasingly the interface for housing searches, offer a window into housing markets via their online representations. Their troves of data can unlock analyses of rental markets, which have traditionally been difficult to monitor but have recently attracted growing research attention (Boeing and Waddell, 2017; Boeing et al., 2021; Harten, 2021; Kroft and Pope, 2014; Rae, 2015; Schachter and Besbris 2017). Leveraging online listings’ novel data opportunity, recent research has shed light on enduring housing market questions such as tenant discrimination (Murchie and Pang, 2018) and new questions regarding how online mediation introduces new – and exacerbates existing – bias and inequality (Boeing, 2020; Harten, 2021). Others have revealed how the portrayal of information can affect a range of real-world rental housing dynamics (Besbris et al., 2021; Harten et al., 2021; Maalsen and Gurran, 2022; Zhang and Gurran, 2021).

In sum, this literature suggests that shared housing offers low-income renters a pathway into unaffordable markets (by trading off privacy/space for lower rent), and that online information exchanges both shape housing search outcomes and reveal neighbourhood-scale housing opportunities. However, fundamental questions about sharing’s trade-offs and spatial distribution remain unanswered. What are the actual rent vs privacy trade-offs of shared housing? In what kinds of neighbourhoods does sharing create affordable housing opportunities compared to other small housing types? Neighbourhoods shape life outcomes (Buck, 2001; Friedrichs et al., 2003; Galster, 2011): if home sharing offers an affordability strategy, then a better understanding of its locations and costs is crucial in the present era of housing crises and planners’ search for solutions.

Methods

In this study, we advance this growing literature by using online listings to investigate shared living as a housing submarket to evaluate its role (in terms of locations and costs) as an affordable housing option in Los Angeles county, one of the most expensive US housing markets. The county has roughly 10 million residents spread across its coastal plains and inland valleys. Belying its reputation for sprawl, the Los Angeles urban area is the USA’s densest and exhibits the high racial, income and wealth diversity and segregation that characterises most large US cities.

We examine full unit and shared unit rental listings. This allows us to investigate both the availability of shared offerings as well as their rents³– and to contrast both with those of small apartments (which we define as studio and one bedroom). Small apartments represent the closest ‘whole unit alternative’ to a room in shared housing, as these can accommodate a single person or a couple in a single sleeping space. Accordingly, the small apartments represent the ‘private’ version and the shared units represent the ‘shared’ version. Thus, we can examine their trade-offs (e.g. cost, privacy, location).

Does home sharing augment the supply of affordable housing? We address this question in two parts. First, are rooms for rent less expensive than small apartments for rent, and if so, to what extent? We hypothesise that they are – and that they in turn offer more affordable accommodation than traditional rental listings. Second, are room listings and small apartments equally available across space? We hypothesise that rooms for rent are more widely available than small apartments – due to their flexibility, the constraints of zoning and Los Angeles’ single-family home-dominated urban landscape – and that they in turn represent an essential, privately and sometimes informally organised affordability strategy.

Data

To answer these questions, we collected rental listings from Craigslist. Craigslist is not the only online platform for rental housing listings but it has been a dominant one for years (Boeing and Waddell, 2017; Hangen and O’Brien, 2023; Kroft and Pope, 2014; Schachter et al., 2023). Such data offer a digital representation of the rental market, including the sharing market, though with possible sampling biases. Using a web scraper, we collected housing listings posted to Craigslist in Los Angeles County from June 2020 to March 2021.⁴ Specifically, each day we collected 500 listings apiece posted in the ‘apts/housing’ and ‘rooms/shared’ categories. The former represents whole housing units for rent. The latter represents listings for individual rooms and shared housing: that is, anything other than whole (and thus private) housing units for rent. This can include rooms for rent in a house (akin to a boarding house), apartment or accessory dwelling unit, and existing tenants looking to add a new tenant, among other possible arrangements.⁵ To generate a random sample, the scraper ran at a random time each day to avoid possible time-of-day biases. The initial collection resulted in 255,436 listings nearly evenly split between apartments for rent and rooms for rent. These listings include the latitude/longitude coordinates of each rental, provided by the lister.

As is always the case with such crowdsourced and volunteered geographic information, the initial data were messy and required substantial cleaning and deduplication. First, we discarded any miscategorised listings that were not actually located in Los Angeles County, leaving us with 251,341 listings still roughly evenly split between apartments and rooms. Next, we dropped duplicate post IDs (indicating that a listing for the same housing unit was reposted multiple times, which is common), leaving us with 175,556 listings. However, it is noteworthy that substantially more room listings than apartment listings were dropped in this phase, suggesting different dynamics in how listers in each category list and relist rental opportunities. Then we dropped listings with duplicate body text (indicating that a listing for the same housing unit was reposted as a separate new listing), leaving us with 128,260 listings. Finally, we retained only observations with valid data in key variables (i.e. we dropped observations with obviously bogus numbers of bedrooms or implausible rents), yielding a final data set for analysis containing 124,332 listings comprising 47,926 rooms for rent and 76,406 apartments for rent, more than half of which (40,796) were listings for studios or one-bedroom apartments.

We supplemented these listings with tract-level sociodemographic data from the 2020 5-year American Community Survey (ACS). We also derived tracts’ distances to employment subcentres, the ocean shoreline, transit, large university campuses and major roadways using public data as described in Table 1.

Table 1.

Tract-level variables and their units and sources.

Variable	Source and units
Population	ACS, 1000s of residents
Area	ACS, km²
Per cent White	ACS, per cent non-Hispanic or Latino White alone
Per cent Latino	ACS, per cent Hispanic or Latino of any race
Per cent Black	ACS, per cent non-Hispanic or Latino Black or African-American alone
Per cent female	ACS, per cent
Per cent with degree	ACS, bachelor’s degree or higher
Per cent age 20–24	ACS, per cent
Per cent age 25–34	ACS, per cent
Median household income	ACS, 100,000s of 2020 inflation-adjusted US$
Median home value	ACS, millions of 2020 inflation-adjusted US$
Housing units	ACS, count of total units
Per cent single unit detached	ACS, per cent
Per cent vacant units	ACS, per cent
Distance to subcentre	SCAG 2016, distance in km to nearest subcentres with employment >100,000
Distance to ocean	Derived, distance in km to nearest ocean shoreline
Distance to transit	OpenStreetMap and LACMTA, distance in km to nearest transit line excluding local/limited lines
Distance to campus	Integrated Post-Secondary Education System, distance in km to nearest university campus with enrolment >10,000
Distance to highway	OpenStreetMap, distance in km to nearest highway with OSM classification ‘motorway’

Analysis

First, we examined whether room rentals are less expensive than small apartment rentals, and if so, to what extent. To do so, we estimate the statistical distribution of asking rents for room and for small apartment listings, calculate the difference in means and conduct Welch’s unequal variance t-tests to determine if asking rents for rooms differ significantly.

Then, we calculate the percentage of rooms and small apartments for rent considered affordable to various income groups. We define ‘affordable’ as a 30% rent burden at Los Angeles County’s Area Median Household Income (AMHI) thresholds for a one-person household in 2021. In addition to household size-adjusted AMHI, California’s Department of Housing and Community Development also reports income bands defined in relation to AMHI. These identify households as acutely low income (0–15% of household size-adjusted AMHI), extremely low income (16–30%), very low income (31–50%), low income (51–60%) and moderate income (61–110%) (HCD, 2021). We use these income thresholds to calculate maximum affordable rents for each income group of one-person households – at a rent burden of 30% of income – and in turn the number of listings affordable to each.

This reveals if room rentals are more affordable than small apartments to different income groups, but it does not tell us if listers use social cues and language to advertise as such. Previous research suggests that listers use rental listings to communicate to desirable and probable target populations (Harten, 2021; Harten et al., 2021; Maalsen and Gurran, 2022). Understanding the degree to which references to affordability feature in these listings reveals its relative importance in each submarket. Accordingly, we analyse listing titles across both rental types for references to rent and keywords related to affordability. The title is the first point of contact between a lister and potential tenant. On the Craigslist results page, a listing is represented by its title, hyperlinked to the full listing, as well as tags for key information: a locational marker, the post time, size and rent. We search listing titles for mentions of (i) rent, (ii) affordability keywords and (iii) additional affordability words, generated via a natural language-processing tool, word2vec, to ensure coverage of relevant terminology (Rehurek and Sojka, 2010).⁶

Next, we examine whether listings of each type are equally available across space by mapping the spatial distribution of room listings and small apartment listings across Los Angeles County. Then we calculate Gini coefficients and cumulative distribution functions (CDFs) of listings’ distributions across tracts to measure how evenly or unevenly they are distributed spatially. A Gini coefficient of 0 would mean the listings are evenly distributed among the tracts, whereas a coefficient of 1 would mean that just one tract contains all the listings. To unpack these spatial distributions, we estimate two negative binomial regression models to explain tract-level listing counts as a function of socio-demographics, location attributes, transport features and other controls including housing units and vacancy rates. Model I predicts room listing counts and Model II predicts small apartment listing counts, using the same set of predictors. See Table 1 for variables, sources and descriptions.

Findings

Affordability of room rentals

Room rents are on average cheaper than rents for small apartments. This is unsurprising but important to quantify in order to understand the extent of the discount as renters trade off privacy for lower rents. Across our listings, the average room rents for US$1054 while the average small apartment rents for US$1907. The rent difference in means is US$853 and is statistically significant (p < 0.001).

Next, we evaluate both rental types against maximum affordable rents by income groups. In 2021, the monthly AMHI for one-person households in Los Angeles County was US$4667 (HCD, 2021). Applying the thresholds for each income group, the maximum affordable monthly rent is US$210 for acutely low-income earners, US$420 for extremely low-income earners, US$700 for very low-income earners, US$840 for low-income earners and US$1540 for those with moderate incomes.

Table 2 breaks down how many listings exist at rent levels affordable to each income group by rental type. Rooms are affordable to single-person households from a much wider spectrum of incomes: 38% of room listings were affordable to low-income earners but only 2% of small apartments were listed at a rent affordable to them. Even very low-income renters – with a maximum affordable rent of US$700 – would find more than 10,000 affordable room listings, compared to fewer than 400 affordable small apartment listings. Interestingly, the moderate-income threshold demarcates a remarkable split: the vast majority of room listings (88%) feature rents affordable to moderate-income earners – but the picture is flipped for small apartments, where 78% list rents affordable only to those with high incomes (>110% of AMHI).

Table 2.

Number and share of listings affordable to each income group of one-person households, by rental type.

	Room	Small apartment
Below 30% of acutely low income	676 (1%)	39 (0%)
Below 30% of extremely low income	2368 (5%)	70 (0%)
Below 30% of very low income	10,623 (22%)	337 (1%)
Below 30% of low income	18,072 (38%)	622 (2%)
Below 30% of moderate income	42,306 (88%)	8845 (22%)
Above 30% of moderate income	5621 (12%)	31,951 (78%)
Listings	47,926 (100%)	40,796 (100%)

The ratio of tract-level median room rents to median small apartment rents underscores the rent savings achieved through sharing. On average, rooms offer a discount of 44.7% over small apartments. Deep discounts exceeding 60% and small discounts below 10% are rare. When we differentiate employment-rich tracts – defined as those located within one mile of an employment subcentre with at least 100,000 jobs – from other tracts, we find that rent discounts are available in both. Median rents in the other tracts are lower for both rooms and small apartments, but rooms still provide an average discount of US$875, comparable to a median discount of US$870 in the employment-rich tracts.

Finally, we also analyse the relative importance of references to rent and affordability in listing titles. Although rent information is typically available via a tag next to the title, we find that both rent values and references to affordability regularly appear in the title text. Six per cent of small apartment listing titles include a reference to either cost or affordability. In contrast, room listing titles include a reference to cost or affordability nearly twice as often, just over 10% of the time.

Availability of room rentals

Figure 1 illustrates the spatial distribution of room and small apartment rental listings, aggregated at the tract level and normalised by population. Room listings are more widely available in more tracts across Los Angeles County compared to small apartment listings. They are also more evenly distributed across Los Angeles County than small apartment listings. The tract-level distribution of room listings has a Gini coefficient of 0.71 but the distribution of small apartment listings has a Gini coefficient of 0.78. Hence, both distributions exhibit some spatial concentration at the tract level, in line with the findings of Boeing (2020), but small apartment listings are more spatially concentrated than room listings.

Figure 1.

Tract-level room (left) and small apartment (right) rental listing counts per thousand residents.

Another way to measure the availability of these rental types across space is through CDFs. As Figure 2 illustrates, 93% of tracts contain at least one room listing but only 70% contain at least one small apartment listing. Similarly, 41% of tracts contain at least 10 room listings but only 32% contain at least 10 small apartment listings. Nearly half of the tracts (46%) contain at least one listing per 500 residents but just over one-third of the tracts (36%) contain at least one small apartment listing per 500 residents. Across these various measures, rooms for rent are consistently more available than small apartments for rent.

Figure 2.

CDFs for room and small apartment listings across tracts, with outliers truncated.

Table 3 presents the parameter estimates for Model I (predicting room listing counts) and Model II (predicting small apartment listing counts). The models explain the number of listings of each type per tract as a function of neighbourhood characteristics. Clear differences exist in their respective relationships with sociodemographic and built environment covariates. The same set of predictors explains nearly three-quarters of the variation in small apartment listing counts (R² = 0.728) but just over one-third of the variation in room listing counts (R² = 0.347). In other words, tracts’ sociodemographic and built environment characteristics are moderately correlated with their small apartment listing counts but only slightly correlated with their room listing counts. This has important implications for neighbourhood diversity, which we discuss later.

Table 3.

Tract-level negative binomial regression model estimates. Model I’s response is room listing counts and Model II’s is small apartment listing counts.

	Model I		Model II
	Coeff	S.E.	Coeff	S.E.
Intercept	3.5808*	0.3377	2.8660*	0.3512
Population	−0.1915*	0.0338	−0.0887*	0.035
Area	0.0025*	0.0006	0.0009	0.0006
Per cent White	0.0005	0.0017	0.0119*	0.0018
Per cent Latino	−0.0009	0.0019	−0.0011	0.0020
Per cent Black	−0.0087*	0.0023	−0.0110*	0.0025
Per cent female	−0.0228*	0.0050	−0.0102*	0.0051
Per cent with degree	0.0033	0.0025	0.0058*	0.0026
Per cent aged 20–24	0.0196*	0.0052	0.0115*	0.0052
Per cent aged 25–34	0.0087*	0.0039	0.0151*	0.0040
Median household income	−0.0001*	0.0001	0.0000	0.0001
Median home value	0.0003	0.0002	−0.0005*	0.0002
Housing units	0.0008*	0.0001	0.0007*	0.0001
Per cent single unit detached	−0.0014	0.0011	−0.0242*	0.0012
Per cent vacant units	0.0053	0.0029	−0.0082*	0.0031
Distance to subcentre	−0.0340*	0.0049	−0.0282*	0.0051
Distance to ocean	−0.0002	0.0024	0.0085*	0.0025
Distance to transit	0.0208*	0.0056	0.0169*	0.0059
Distance to campus	−0.0606*	0.0082	−0.0214*	0.0086
Distance to highway	0.0417*	0.0143	−0.0301	0.0155
Pseudo-R²	0.347		0.728
N	2301		2301

Notes: * Indicates statistical significance at p < 0.05. See Table 1 for variable descriptions.

Figures 1 and 2 and the Gini coefficients together provide evidence that room listings are more evenly distributed spatially than small apartment listings are. These regression results offer further evidence that room listings are more evenly distributed than small apartment listings with respect to socio-demographics. For example, all else equal, a one percentage point increase in a tract’s white population share is associated with a 0.01 increase in small apartment listings but a statistically insignificant increase in room listings. A one percentage point increase in college degree attainment is associated with a 0.01 increase in small apartment listings but a statistically insignificant increase in room listings.

Similarly, measures of location desirability, including median home value and distance to the ocean, significantly predict small apartment listing counts – but not room listing counts: lower home values and longer distance to the ocean are associated with more small apartment listings but these coefficients lack statistical association with room listing counts. Unsurprisingly, single-family detached housing stock is negatively associated with small apartment listing counts only. Other locational characteristics, however, such as a greater share of young residents (between ages 20 and 34) and a closer proximity to a university campus, significantly predict both higher small apartment and room listing counts.

Finally, all else equal, measures of racial diversity show more consistent trends across the two models. For example, the Latino population percentage is insignificant in both models. The Black population percentage on the other hand is significant in both: a one percentage point increase in a tract’s Black population share is associated with a 0.01 decrease in both room and small apartment listings.

Discussion

This study asked if and where home sharing offers affordable housing. We approached this question from two perspectives. First, we found that rooms for rent are less expensive than small apartments for rent, on average. They are therefore more attainable for one-person households from a variety of incomes and offer more affordable accommodation than traditional, small apartment listings. Second, we found that there are more listings for rooms for rent than for small apartments, and that they in turn represent an important affordability option for accessing housing in a wide range of neighbourhoods in unaffordable cities and exclusive local housing submarkets.

On average, small apartment rents are nearly double those of rooms, suggesting that economising on space and privacy can indeed provide a pathway to affordability. The literature has documented that growing numbers of single renters are willing – or needing – to make this choice. Our findings help explain the trend towards sharing by quantifying the cost savings achieved in Los Angeles County. Importantly, rooms are not only more affordable generally but are more affordable to people with a variety of incomes. Our analysis shows that for lower-income renters, rooms may be the only affordable option when looking to access an expensive housing market such as that of Los Angeles through private renting. On the flipside, in the absence of effective policy solutions, this means that in tight markets low-income earners not willing or able to share may either be excluded from moving to the city or, if displaced from their current accommodation, have to move away.

Sharing promises a pathway to carving out affordability in unaffordable markets. However, the literature’s focus on sharing as a housing option merely for students and young professionals leaves unanswered whether home sharing is suitable or even realistic for people outside this narrow demographic. Our analysis focuses broadly on one-person households, but we do not observe sharer demographics. Assumptions regarding who may want or could reasonably be expected to share housing need to be interrogated to avoid misjudging the role of home sharing in the fight against unaffordability. For example, the detrimental effects on vulnerable populations – when sharing is assumed to be a fair and appropriate housing choice – have already been documented in the literature discussing the UK’s decision to raise the age for sharing among single housing welfare recipients (Green and McCarthy, 2015; Ortega-Alcázar and Wilkinson, 2021). Future research into who is sharing and why is needed to fully unpack the share sector for policy and planning.

Our results also reveal that listers, typically either landlords or existing tenants, emphasise affordability in room listing titles more so than in the titles for small apartment listings. However, our research design cannot tell us whether these are savvy landlords strategising to advertise desirability or existing tenants reflecting a well-known need among the target population. Regardless, including rents or affordability keywords in listing titles signals the relative importance of affordability in the housing search for these submarkets. Future research into rental listings as communication channels could shed further light on how each rental type is advertised and who the desirable target populations are.

Another major promise of sharing as an affordable housing strategy lies in its inherent flexibility. In principle, any type of structure or unit can be shared, allowing for bottom-up densification even in neighbourhoods that have long opposed it. Our findings provide evidence that this promise is realistic. Room listings are indeed more evenly distributed across Los Angeles than small apartment listings are. Our analysis suggests that small apartment listings are more prevalent in whiter and better educated neighbourhoods; all else equal, they also tend to be in places with lower home values and less single-family housing. The same is not true for room listings. Lack of statistical association suggests that shared housing offers access to a greater variety of neighbourhoods in more locations, thus providing a pathway into a challenging housing market and access to a variety of spatial opportunity structures.

Notably, however, certain barriers to entry persist. Small apartment and room listings are negatively associated with greater percentages of Black residents and positively associated with greater percentages of young residents aged 20–34. The former suggests that longstanding obstacles to Black home seekers may continue via room sharing, even while other communities can benefit from their affordability and ubiquity. The latter implies the importance of small rental units for younger adults but also suggests that fewer such units may be available in older communities. Older home seekers have long faced housing market barriers such as limited access to information and reduced search radii, which may not be attenuated by room shares.

In sum, these findings suggest that sharing can be an effective affordability strategy, potentially creating access to more housing markets for people with a diversity of incomes – albeit at the cost of space and privacy. Documenting sharing as a sizeable, privately organised housing submarket may be seen either as promising, given the ongoing, global affordability crisis, or as alarming, if interpreted as a symptom of the same. Either way, our analysis quantifies a housing phenomenon not well captured by traditional housing data and makes the case that it should be measured better. Accordingly, fair housing advocates and planning practitioners, too, must critically interrogate the role of shared housing in their communities. Whether home sharing reflects the savvy of a new generation of city dwellers or is a reflection of grim economic realities matters for how policy responds and demands further inquiry.

This study creates several other opportunities for future research. First, our analysis relies on a single listings platform. Although Craigslist is a popular platform in North America with the fewest barriers to entry, future research should look at competing sites to understand self-selection biases, such as those mediated by language or race. Additionally, data collection occurred between 2020 and 2021, in the midst of the global Covid-19 pandemic. As other scholars have suggested, the pandemic disrupted housing processes, and may have affected the information supply, unit characteristics and listing language. However, our collection captured both the initial plunge and the subsequent skyrocketing of housing costs. While we do not expect a pandemic upending of market fundamentals for the submarkets considered, future research should replicate this study with post-pandemic data. Finally, our research design is correlational, but future causal research can further unpack these relationships for policy making.

There are also several opportunities to expand our analysis. In many ways, Los Angeles exemplifies the modern American metropolis and its crises, but studying smaller towns or cities in other parts of the world would enrich these findings. Focusing on rental listings to study shared housing only captures home sharing between strangers, not between friends and family. Our results thus only partially capture the dynamics of the larger phenomenon. Future research could look at both these privately negotiated shared agreements as well as differences between professionally managed versus peer-managed sharing. Additionally, our analysis cannot speak to the quality of shared rentals. The significant rent discounts offered by the shared housing market could at least partially be achieved via substandard living conditions. Overcrowding, for instance, has been documented in home sharing in other geographical contexts (Harten et al., 2021; Nasreen and Ruming, 2019). Finally, although the large rent differentials between rooms and small apartments for rent suggest its appeal as an affordable housing strategy, our analysis ultimately does not reveal what motivates home sharers.

These open questions sketch out a nascent research agenda for shared rentals. Does the more even spatial distribution of rooms for rent translate into better access to opportunities? Measuring access to transit and employment, from room listings’ vs small apartment listings’ locations, could further illuminate sharing’s role in creating affordable access. In principle sharing can happen anywhere, but in practice certain neighbourhoods may be more or less conducive to sharing. As preliminary accounts of protests against home sharing make the news (e.g. Brasuell, 2022), it is critical to understand the relationship between shared housing and neighbourhood change. Future research must unpack what type of structures are hosting shared living and who is receiving the rent. Is this a market where tenants organise, from the bottom up, to meet rising rents, or are shared rentals managed by owners and property managers that recognise the potential to extract more rent by increasing occupancy?

Finally, the differences in title text content across the rental types suggest the importance of exploiting the rich data contained in listings’ body text to better understand how each rental type is advertised and what language reveals about target demographics. A major concern with the rise of shared housing is that it may not actually be as accessible as its affordability and wide availability promise. In particular, an analysis of text bodies could shed light on who can effectively access shared living. Given the intimacy of sharing a home, the dynamics of sex and other identity markers such as class, race, language or origin may significantly shape this market.

Conclusion

As cities grow more unaffordable, alternative pathways into the housing market become more important. Past research has shown that shared rentals offer one such pathway, but it has typically focused on only certain kinds of home seekers. In this study, we examined shared rentals as a housing submarket that can offer housing opportunities to income-diverse populations of one-person households. We analysed thousands of online rental listings in Los Angeles and found that shared rentals were both more widely available and more affordable than listings for whole housing units. Our findings suggested that they may in fact be the only pathway to housing for many low-income renters in unaffordable markets. Nevertheless, shared rental housing has historically been understudied. We call for shared housing researchers to shift the traditional focus away from young adults and towards a much broader exploration of home sharing as a housing strategy.

Footnotes

The authors wish to thank Madison Lore and Jaehyun Ha for research assistance. The authors also wish to thank the participants of the Workshop on Digitisation,Neighbourhood Change and Social Integration in May 2022 in Cologne,Germany as well as the special issue editors and three anonymous reviewers for their helpful suggestions.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research,authorship,and/or publication of this article: Julia Harten gratefully acknowledges support from the Lusk Center for Real Estate and the Social Sciences and Humanities Research Council Canada Research Chair Program.

ORCID iDs

Julia Gabriele Harten

Geoff Boeing

References

Aeby

Heath

(2020) Post break-up housing pathways of young adults in England in light of family and friendship-based support. Journal of Youth Studies 23(10): 1381–1395.

Ahrentzen

(2003) Double indemnity or double delight? The health consequences of shared housing and “doubling up”. Journal of Social Issues 59(3): 547–568.

Arundel

Ronald

(2016) Parental co-residence, shared living and emerging adulthood in Europe: Semi-dependent housing across welfare regime and housing system contexts. Journal of Youth Studies 19(7): 885–905.

Bate

(2018) Understanding the influence tenure has on meanings of home and homemaking practices. Geography Compass 12(1): e12354.

Bergan

Gorman-Murray

Power

(2021) Coliving housing: Home cultures of precarity for the new creative class. Social & Cultural Geography 22(9): 1204–1222.

Besbris

Schachter

Kuk

(2021) The unequal availability of rental housing information across neighborhoods. Demography 58(4): 1197–1221.

Boeing

(2020) Online rental housing market representation and the digital reproduction of urban inequality. Environment and Planning A: Economy and Space 52(2): 449–468.

Boeing

Waddell

(2017) New insights into rental housing markets across the United States: Web scraping and analyzing Craigslist rental listings. Journal of Planning Education and Research 37(4): 457–476.

Boeing

Besbris

Schachter

, et al. (2021) Housing search in the age of big data: Smarter cities or the same old blind spots? Housing Policy Debate 31(1): 112–126.

10.

Bradbury

Peterson

Liu

(2014) Long-term dynamics of household size and their environmental implications. Population and Environment 36(1): 73–84.

11.

Brasuell

(2022) Co-living banned in Shawnee, Kansas. Planetizen, 1 May. Available at: https://www.planetizen.com/news/2022/05/117028-co-living-banned-shawnee-kansas (accessed 5 May 2022).

12.

Buck

(2001) Identifying neighbourhood effects on social exclusion. Urban Studies 38(12): 2251–2275.

13.

Buckle

Gurran

Harris

, et al. (2023) Intersections between housing and health vulnerabilities: Share housing in Sydney and the health risks of COVID-19. Urban Policy and Research 41(1): 22–37.

14.

California Department of Real Estate (2022) California tenants: A guide to residential tenants’ and landlords’ rights and responsibilities. Available at: https://www.courts.ca.gov/documents/California-Tenants-Guide.pdf (accessed 1 December 2023).

15.

Chen

SCJ

(2022) Millennials drive co-living trend in Hong Kong and Singapore. Forbes, 12 December. Available at: https://www.forbes.com/sites/shuchingjeanchen/2022/12/12/millennials-drive-co-living-trend-in-hong-kong-and-singapore/?sh=19ab0a0d7547 (accessed 1 October 2022).

16.

Clapham

(2002) Housing pathways: A post modern analytical framework. Housing, Theory and Society 19(2): 57–68.

17.

Clark

Tuffin

Frewin

, et al. (2017) Shared housing among young adults: Avoiding complications in domestic relationships. Journal of Youth Studies 20(9): 1191–1207.

18.

Corfe

(2019) Co-living: A Solution to the Housing Crisis. London: The Social Market Foundation. Available at: https://thinkhouse.org.uk/site/assets/files/1885/smf0219.pdf (accessed 1 October 2022).

19.

Department of Housing and Community Development (HCD) (2021) Revised State Income Limits for 2021. Memorandum, State of California Business, Consumer Services and Housing Agency, December 2021. Available at: https://www.hcd.ca.gov/grants-funding/income-limits/state-and-federal-income-limits/docs/income-limits-2021.pdf (accessed 1 October 2022).

20.

Devidayal

(2022) Co-living with meals, gyms, housekeeping trending among youth. Times of India, 28 November. Available at: http://timesofindia.indiatimes.com/articleshow/95816885.cms?utm_source=contentofinterest&utm_medium=text&utm_campaign=cppst (accessed 1 October 2022).

21.

Druta

Ronald

(2021) Living alone together in Tokyo share houses. Social & Cultural Geography 22(9): 1223–1240.

22.

Druta

Ronald

Heath

(2021) Urban singles and shared housing. Social & Cultural Geography 22(9): 1195–1203.

23.

Friedrichs

Galster

Musterd

(2003) Neighbourhood effects on social opportunities: The European and American research and policy context. Housing Studies 18(6): 797–806.

24.

Fuster

Arundel

Susino

(2019) From a culture of homeownership to generation rent: Housing discourses of young adults in Spain. Journal of Youth Studies 22(5): 585–603.

25.

Galster

(2011) The mechanism(s) of neighbourhood effects: Theory, evidence, and policy implications. In: van Ham

Manley

Bailey

, et al. (eds) Neighbourhood Effects Research: New Perspectives. Dordrecht: Springer, (pp). 23–56.

26.

Gorman-Murray

(2007) Contesting domestic ideals: Queering the Australian home. Australian Geographer 38(2): 195–213.

27.

Green

McCarthy

(2015) Is sharing the solution? Exploring the opportunities and challenges of privately rented shared accommodation for single people in housing need. People, Place and Policy 9(3): 159–178.

28.

Grundström

(2021) Dwelling on-the-move together in Sweden: Sharing exclusive housing in times of marketization. Social & Cultural Geography 22(9): 1241–1255.

29.

Gurran

Maalsen

Shrestha

(2022) Is ‘informal’ housing an affordability solution for expensive cities? Evidence from Sydney, Australia. International Journal of Housing Policy 22(1): 10–33.

30.

Hangen

O’Brien

(2023) The choice to discriminate: How source of income discrimination constrains opportunity for housing choice voucher holders. Urban Affairs Review 59(5): 1601–1625.

31.

Harten

(2021) Housing single women: Gender in China’s shared rental housing market. Journal of the American Planning Association 87(1): 85–100.

32.

Harten

(2024) Housing vulnerability, shared housing, informality, and crowding: The housing strategies of China’s recent college graduates. Housing, Theory and Society. Epub ahead of print 5 April 2024. DOI: 10.1080/14036096.2024.2331170.

33.

Harten

Kim

Brazier

(2021) Real and fake data in Shanghai’s informal rental housing market: Groundtruthing data scraped from the internet. Urban Studies 58(9): 1831–1845.

34.

Heath

Davies

Edwards

, et al. (2017) Shared Housing, Shared Lives: Everyday Experiences across the Lifecourse. London: Routledge.

35.

Hilder

Charles-Edwards

Sigler

, et al. (2018) Housemates, inmates and living mates: Communal living in Australia. Australian Planner 55(1): 12–27.

36.

Huang

Clark

(2022) Families in transition: Living arrangements and intergenerational support in 21st century China. Transactions in Planning and Urban Research 1(1–2): 115–134.

37.

Kenyon

Heath

(2001) Choosing this life: Narratives of choice amongst house sharers. Housing Studies 16(5): 619–635.

38.

Kim

Woo

Cho

(2020) Is shared housing a viable economic and social housing option for young adults? Willingness to pay for shared housing in Seoul. Cities 102: 102732.

39.

Kroft

Pope

(2014) Does online search crowd out traditional search and improve matching efficiency? Journal of Labor Economics 32(2): 259–303.

40.

Los Angeles Housing Department (2018) Home-sharing ordinance, Ordinance 185931, 2 December 2018. Available at: https://planning.lacity.gov/ordinances/docs/HomeSharing/adopted/Final%20Ordinance.pdf (accessed 15 April 2022).

41.

Maalsen

(2020) ‘Generation share’: Digitalized geographies of shared housing. Social & Cultural Geography 21(1): 105–113.

42.

Maalsen

Gurran

(2022) Finding home online? The digitalization of share housing and the making of home through absence. Housing, Theory and Society 39(4): 401–419.

43.

McNamara

Connell

(2007) Homeward bound? Searching for home in Inner Sydney’s share houses. Australian Geographer 38(1): 71–91.

44.

Margolies

(2020) Co-living grows up. The New York Times, 14 January. Available at: https://www.nytimes.com/2020/01/14/realestate/co-living-grows-up.html (accessed 1 October 2022).

45.

Milan

(2000) One hundred years of families. Canadian Social Trends 56: 2–12.

46.

Murchie

Pang

(2018) Rental housing discrimination across protected classes. Regional Science and Urban Economics 73: 170–179.

47.

Myers

Baer

Choi

(1996) The changing problem of overcrowded housing. Journal of the American Planning Association 62(1): 66–84.

48.

Nasreen

Ruming

(2019) Room sharing in Sydney: A complex mix of affordability, overcrowding and profit maximisation. Urban Policy and Research 37(2): 151–169.

49.

Nasreen

Ruming

(2021) Informality, the marginalised and regulatory inadequacies: A case study of tenants’ experiences of shared room housing in Sydney, Australia. International Journal of Housing Policy 21(2): 220–246.

50.

Ortega-Alcázar

Wilkinson

(2021) ‘I felt trapped’: Young women’s experiences of shared housing in austerity Britain. Social & Cultural Geography 22(9): 1291–1306.

51.

Parkinson

James

Liu

(2021) Luck and leaps of faith: How the digital informal economy transforms the geographies of shared renting in Australia. Social & Cultural Geography 22(9): 1274–1290.

52.

Pilkauskas

Garfinkel

McLanahan

(2014) The prevalence and economic value of doubling up. Demography 51(5): 1667–1676.

53.

Rae

(2015) Online housing search and the geography of submarkets. Housing Studies 30(3): 453–472.

54.

Rehurek

Sojka

(2010) Software framework for topic modelling with large corpora. In: Proceedings of LREC 2010 workshop new challenges for NLP frameworks, Valletta, Malta, 22 May 2010, pp.46–50. Valletta: ELRA

55.

Rugg

(1999) Young People, Housing and Social Policy. London: Routledge.

56.

Schachter

Besbris

(2017) Immigration and neighborhood change: Methodological possibilities for future research. City & Community 16(3): 244–251.

57.

Schachter

Kuk

Besbris

, et al. (2023) Inclusive but not integrative: Ethnoracial boundaries and the use of Spanish in the market for rental housing. Sociological Science 10: 585–612.

58.

Steinführer

Haase

(2009) Flexible–inflexible: Socio-demographic, spatial and temporal dimensions of flat sharing in Leipzig (Germany). GeoJournal 74(6): 567–587.

59.

UN-Habitat (2022) World Cities Report 2022: Envisaging the future of cities. Available at: https://unhabitat.org/sites/default/files/2022/06/wcr_2022.pdf (accessed 1 December 2022).

60.

Waldron

(2024) Responding to housing precarity: The coping strategies of generation rent. Housing Studies 39(1): 124–145.

61.

Wiemers

(2014) The effect of unemployment on household composition and doubling up. Demography 51(6): 2155–2178.

62.

Zhang

Gurran

(2021) Understanding the share housing sector: A geography of group housing supply in metropolitan Sydney. Urban Policy and Research 39(1): 16–32.

63.

Zhu

Burinskiy

De la Roca

, et al. (2021) Los Angeles’ housing crisis and local planning responses. Cityscape 23(1): 133–160.