This chapter examines the challenges of measuring homelessness and housing exclusion, focussing on two forms that are often underrepresented in official statistics: unsheltered homelessness and temporary stays with family or friends due to a lack of housing. It discusses how differences in definitions and data collection methods can affect the coverage and comparability of homelessness estimates, and why these gaps matter for policymaking. The chapter introduces Germany’s current homelessness measurement framework and outlines the report’s objectives, methodology and scope. It also situates Germany’s experience within broader OECD efforts to strengthen the measurement of less visible forms of homelessness.
Measuring Unsheltered and Hidden Homelessness in Germany
1. Introduction
Copy link to 1. IntroductionAbstract
Measuring the scale and diverse forms of homelessness and housing exclusion
Copy link to Measuring the scale and diverse forms of homelessness and housing exclusionHomelessness and housing exclusion are among the most severe forms of poverty in OECD and EU countries. In 2026, more than 2.6 million people were recorded as experiencing homelessness in OECD countries for which data are available (OECD, 2026[1]). Nevertheless, the current headline estimate of homelessness should be understood as a lower-bound estimate.
Many governments lack a comprehensive understanding of both the scale of homelessness and the diverse ways in which people experience homelessness and housing exclusion. This is partly because the statistical definitions used by governments as a basis for measuring homelessness differ across countries and do not systematically cover the same set of living situations. For instance, some statistical definitions cover only people sleeping rough or staying in shelters, while others also include people in institutions, non-conventional dwellings, or temporary arrangements with family or friends. In addition, the headline estimate is affected by the parameters and limitations of the range of collection method(s) from which homelessness data are derived, which include street counts, population censuses and household surveys, survey-based methods, administrative data and information management systems, among others. Depending on their design and implementation, these data collection approaches may fail to take into account individuals who are not in contact with services, are not eligible for support, avoid visible public spaces, or live in insecure arrangements outside official records (OECD, 2024[2]; OECD, 2025[3]).
Homelessness measurement gaps are not evenly distributed across the population, as research has shown that some socio-demographic groups are more likely to be missed in standard data collection exercises (OECD, 2025[3]). For example, women tend to be underrepresented in homelessness data based on street counts because they often avoid sleeping in visible public spaces for safety reasons. Young people are more likely than others to move between informal stays and family networks, making them hard to capture in standard data collection exercises. In some cases, LGBTI people may be less likely to access homelessness shelters and services due to concerns about discrimination or personal safety and thus may be underreported in service-based methods. Migrants and asylum seekers may be less likely to be captured in official counts due to barriers to accessing services, administrative requirements, or concerns regarding engagement with authorities. Victims and survivors of domestic violence, as well as people in rural areas or smaller communities where services and street counts are less common, may also be missed in standard data collection exercises.
These data gaps matter for policy. Reliable information on who experiences homelessness, where they are, and what forms of housing exclusion they face provides governments with a stronger basis for designing effective responses. Better measurement can help policymakers assess the scale and profile of homelessness, monitor trends over time, identify groups that are more likely to be missed, target prevention and support measures, allocate resources more effectively, and assess whether reforms are reducing homelessness. Measurement can also have a recognition function: groups that are counted are more likely to be reflected in policy design, service planning and funding decisions (Dyb et al., 2021[4]).
Strengthening the measurement of two forms of homelessness and housing exclusion: Unsheltered homelessness and temporary stays with family or friends
Copy link to Strengthening the measurement of two forms of homelessness and housing exclusion: Unsheltered homelessness and temporary stays with family or friendsMany governments in the OECD and EU are working to improve the measurement of different forms of homelessness and housing exclusion. Some have expanded street counts to better estimate unsheltered homelessness. Others have strengthened service-based surveys, used administrative data more systematically, included homelessness-related questions in population censuses or household surveys, or developed information management systems that bring together client-level data from different services. For governments seeking to draw lessons from these experiences, the approaches involve different trade‑offs in terms of coverage, frequency, cost, data quality and ease of implementation. No single method provides a comprehensive assessment of homelessness and housing exclusion on its own. However, each can improve understanding of specific forms of homelessness when its scope, limitations and intended policy use are clear.
Purpose of this report
This report aims to identify opportunities to refine and strengthen Germany’s existing approach to homelessness measurement. It has been prepared at the request of the German Federal Ministry for Housing, Urban Development and Building (BMWSB).
In recent years, Germany has developed a dual approach to homelessness measurement, mandated by the 2020 Homelessness Reporting Act (Wohnungslosenberichterstattungsgesetz). Germany’s approach combines an annual, federal-level point-in-time count of sheltered homelessness with a biennial supplementary report that also covers unsheltered homelessness and people temporarily staying with family or friends. The Act also required the first supplementary report to assess the feasibility of extending data collection to further, not-yet-captured forms of homelessness, to the extent this is a reasonable undertaking.
Separately, the Act requires the third supplementary report, due in 2026, to assess whether adjustments to the current statistical methodology could be considered, including in relation to forms of homelessness currently not covered by the annual survey. Adjustments should be drawn from findings from the annual counts and the two preceding supplementary reports.
Taken together, the extension beyond sheltered homelessness through the biennial survey, and its built-in review mechanism, signal the government’s commitment to strengthening measurement of homelessness and housing exclusion, including its less visible forms. Despite these efforts, an unquantifiable degree of underreporting persists (Bundesministerium für Wohnen, Stadtentwicklung und Bauwesen, 2024[5]).
Focus on two specific forms of homelessness
This report reviews measurement approaches to assess two forms of homelessness and housing exclusion that present specific challenges: unsheltered homelessness – also known as rough sleeping (ETHOS Light 1) – and people staying temporarily with family, friends or acquaintances due to lack of housing (ETHOS Light 6).1 Unsheltered homelessness is defined as people living in public or external spaces without shelter that can be defined as living quarters. People staying temporarily with family, friends or acquaintances due to lack of housing is often referred to as “hidden homelessness”, even though the concept can be understood more broadly (Box 1.1).
These two forms of homelessness are less readily captured through routine reporting by homelessness services, compared to other forms of homelessness – namely, people staying in emergency or temporary accommodation (ETHOS Light 2 and 3), who are generally assessed through reporting from service providers, administrative records or case‑management systems, even though gaps in coverage and data quality may remain. People sleeping rough or staying temporarily with family or friends may have less regular contact with services and are therefore more likely to fall outside administrative records. Accordingly, measuring unsheltered homelessness and temporary stays with family or friends often requires additional, complementary approaches, such as street counts, outreach-based data collection, targeted surveys, sampling or estimation.
Box 1.1. What is “hidden homelessness”?
Copy link to Box 1.1. What is “hidden homelessness”?There is no formal definition of “hidden homelessness,” and definitions differ across countries. The term can be used i) broadly, to describe people experiencing homelessness or housing exclusion who are not captured in official statistics; and/or ii) more narrowly, to refer to specific living arrangements, such as staying temporarily with family or friends because of lack of housing.
The OECD Monitoring Framework to Measure Homelessness (OECD, 2025[3]) adopts a broad definition of hidden homelessness to refer to people whose living situation corresponds to one of the categories outlined in the ETHOS Light typology (see Table 2.1), but who do not appear in official statistics on homelessness. This may include, for instance:
people who are not in contact with any administrative support services, and are thus not registered in any service database;
people who are not eligible for support services, or are not considered a priority case to access limited public support services, and thus do not seek out support;
(linked to the above) people who avoid visible public spaces whilst experiencing unsheltered homelessness (ETHOS Light 1), and are thus not counted as part of data collection approaches based on street counts; and/or
people living in unsustainable or inadequate shelter (ETHOS Light 5) or temporarily staying with friends or family (ETHOS Light 6). In these cases, the term “hidden homelessness” may refer to individuals whose living situations are excluded by the national statistical definition, or are not captured by the data collection methods.
In practice, countries interpret and operationalise hidden homelessness in different ways, depending on their legal and statistical definitions of homelessness and data collection methods. Some apply the term broadly to refer to diverse forms of homelessness that remain outside official statistics, while others apply it to a more narrowly defined living situation, most commonly temporary stays with family, friends or acquaintances. These differences are well illustrated by the examples of Germany and Canada.
Germany, for instance, adopts a relatively narrow concept of verdeckt wohnungslose Personen (“concealed homeless persons”). This refers to people who do not have their own tenancy agreement and who are living with relatives, friends or acquaintances. Germany’s broader legal definition of homelessness is based on the absence of a rental or tenancy agreement, property rights or other access to a dwelling. The distinction matters for measurement. In Germany, the 2024 federal homelessness report focusses on three groups: sheltered homeless persons, people without accommodation, and concealed homeless persons. Sheltered homelessness is covered through official statistics collected by the Federal Statistical Office, while people without accommodation and concealed homeless persons are measured through an empirical survey that is carried out separately every two years.
By contrast, Canada provides an example of a relatively broad approach to hidden homelessness in data collection. Housing, Infrastructure and Communities Canada includes people temporarily staying with others without a guarantee of continued residence, such as people couch surfing, as well as people staying in short-term rental accommodation without security of tenure. At the same time, Canada’s Point-in-Time Count standards recognise the limits of existing count methods: hidden homelessness cannot be enumerated through the core PiT count methodology, although people experiencing hidden homelessness can be included in the survey component.
Note: ETHOS Light refers to the European Typology of Homelessness and Housing Exclusion (FEANTSA, 2007[6]). ETHOS Light 6 covers people living temporarily in conventional housing with family and friends due to lack of housing.
Structure of this report
This report is structured as follows:
Chapter 2 provides an overview of data collection approaches in OECD and EU countries to assess unsheltered homelessness and temporary stays with family or friends.
Chapter 3 outlines the strengths and limitations of different approaches to measure unsheltered and hidden homelessness (e.g. street counts, service-based approaches, administrative data, population censuses, household surveys and information management systems), illustrated by selected country practices.
Chapter 4 presents cross-cutting lessons from the case studies in Chapter 3, as well as the potential trade‑offs when considering different methodological approaches to measure unsheltered homelessness and temporary stays with family or friends. The chapter concludes with a set of proposed policy recommendations for the German Government, identifying both short-term actions and medium- to long-term reforms.
The methodological approach underpinning the analysis in this report is summarised in Box 1.2.
A more robust evidence base covering different forms of homelessness and housing exclusion can help governments identify gaps in policy responses, understand how needs and barriers differ across groups, target resources more effectively and assess whether interventions are effective in reducing homelessness over time. Further, improving the understanding of less visible forms of homelessness can challenge assumptions about who experiences homelessness and the situations in which it occurs.
Box 1.2. Methodological approach of this report
Copy link to Box 1.2. Methodological approach of this reportThe analysis in this report draws on several complementary sources to assess approaches to measuring homelessness across OECD and EU countries to identify relevant practices for Germany:
Responses to the 2025 OECD Questionnaire on Affordable and Social Housing (QuASH), which was circulated to all OECD non-EU countries in November 2025. The QuASH included a dedicated module on homelessness data, covering national data-collection systems, the methods used to measure different categories of homelessness, including relevant disaggregation, where available. Responses were reviewed and verified by the OECD, with follow-up exchanges with country officials where additional clarification was required. Selected information (including the latest data and data collection approaches) is reported in a set of detailed OECD Country Notes: Data on Homelessness and Housing Exclusion; cross-national comparative indicators on national homelessness estimates (HC3.1) and national strategies (HC3.2) are available in the OECD Affordable Housing Database.
Targeted desk research, to complement and validate the information reported through the questionnaire. This included reviewing methodological documentation, official publications and other relevant sources to better understand how national data-collection systems operate in practice. Where useful, the OECD also consulted governments officials, academic experts and other specialists to clarify methodological issues and verify the interpretation of available information. Case studies were selected to cover the main data collection approaches used across countries.
A fact-finding mission to Germany, involving discussions between the OECD and the public authorities responsible for homelessness data collection, institutes that carried out the data collection, as well as researchers, non-governmental organisations and other relevant stakeholders. These consultations provided additional insights into the implementation of Germany’s existing measurement approaches, their strengths and limitations, and potential areas for further development.
Taken together, these sources provide the basis for the assessment of Germany’s approach and its comparison with measurement practices on unsheltered and “hidden” homelessness in other OECD countries.
References
[8] Bundesministerium für Arbeit und Soziales (2022), Ausmaß und Struktur von Wohnungslosigkeit - Der Wohnungslosenbericht 2022 des Bundesministeriums für Arbeit und Soziales, Bundesministerium für Arbeit und Soziales, https://www.bmas.de/SharedDocs/Downloads/DE/Soziale-Sicherung/wohnungslosenbericht-2022.pdf?__blob=publicationFile&v=4.
[5] Bundesministerium für Wohnen, Stadtentwicklung und Bauwesen (2024), Wohnungslosenbericht der Bundesregierung - Ausmaß und Struktur von Wohnungslosigkeit, Bundesministerium für Wohnen, Stadtentwicklung und Bauwesen, https://www.bmwsb.bund.de/SharedDocs/downloads/DE/veroeffentlichungen/wohnen/wohnungslosenbericht-2024.pdf?__blob=publicationFile&v=1.
[4] Dyb, E. et al. (2021), “Governing Through Definitions and Numbers: Analysis of the Nordic Homeless Registrations as Instruments of Governing Homelessness”, European Journal of Homelessness, Vol. 15/3, pp. 187-201, https://www.feantsa.org/files/Observatory/Journals/Volume-15/V15-3/EJH_15-3_A13.pdf.
[6] FEANTSA (2007), Ethos Light - European Typology of Homelessness and Housing Exclusion, https://www.feantsa.org/ethos.
[7] Housing, Infrastructure and Communities Canada (2025), Everyone Counts - A Guide to Point-in-Time Enumerations in Canada, https://homelessnesslearninghub.ca/wp-content/uploads/2025/07/Guide-to-Point-in-Time-Enumerations-in-Canada-1st-Edition.pdf.
[1] OECD (2026), OECD Affordable Housing Database, https://www.oecd.org/en/data/datasets/oecd-affordable-housing-database.html.
[3] OECD (2025), OECD Monitoring Framework to Measure Homelessness, OECD Publishing, Paris, https://doi.org/10.1787/3e98455b-en.
[2] OECD (2024), OECD Toolkit to Combat Homelessness, OECD Publishing, Paris, https://doi.org/10.1787/0fec780e-en.
Note
Copy link to Note← 1. This report uses the ETHOS Light typology (Table 2.1) as a common framework to define and compare different living situations associated with homelessness and housing exclusion (FEANTSA, 2007[6]). The report focusses on the measurement of people living rough (ETHOS Light 1) and people living temporarily in conventional housing with family and friends due to lack of housing (ETHOS Light 6). It will refer to these categories as “unsheltered homelessness” and “(temporarily) staying with family or friends”, respectively, throughout the report. For readability, “temporarily” is not used consistently throughout.