Key finding: 5.5 million people
In total, more than 5.5 million people received food assistance from RDC at least once between 2018 and 2025.
Extreme Poverty in Wealthy Nations: Evidence from France
Non-technical Summary
Original paper
Extreme poverty remains, to this day, a blind spot in economic studies on income and wealth inequality. Recent research has focused more on the other end of the distribution—understanding the economic mechanisms underlying great wealth. This relative lack of large-scale academic work on poverty stems from the fact that the data available to researchers do not adequately cover the most disadvantaged populations, particularly the homeless. Yet these issues are particularly relevant today: numerous indicators document a rapid increase in extreme precariousness in several wealthy countries—including France, the United Kingdom, and the United States—over the past decade.
Data from the Restos du Cœur, henceforth RDC, offers us a unique opportunity to shed new light on this segment of the population. This untapped (and anonymized) data is both extremely rich—it covers a very significant portion of the most vulnerable people across the entire country—and available “in real time”—twice a year, just a few months after the data is collected. Furthermore, the RDC’s network is particularly dense: there are 1,928 RDC distribution centers, or an average of one center for every 35,000 residents. We estimate that nearly the entire French population lives within a 20-kilometer radius of a drop-in center, with an average distance of between 5 and 10 kilometers.
This data allows us to track the individuals assisted by RDC throughout their time with the organization. For each individual, we examine, among other things, their demographic and family characteristics, their ZIP code (which allows us to link them to the specific economic conditions of their local area, such as unemployment or inflation rates), their sources of income (wages and labor market status, welfare benefits, pensions), certain expenses (rent, utilities, child support, debt repayments), their housing situation (homelessness or, more generally, housing insecurity, public housing, private rental, or homeownership), and the frequency with which they visit RDC. We therefore have access to valuable and highly detailed information about the recipients, their finances, and their consumption patterns.
Our article uses this data to shed light on the determinants, dynamics, and consequences of extreme poverty in France.
1 KU Leuven 2 Brown University 3 Toulouse School of Economics 4 University of Mannheim 5 Federal Reserve Bank of Chicago
What do we mean by food insecurity, housing insecurity, and homelessness?
Let’s begin by defining the terms we use throughout the study. The households received by RDC are experiencing food insecurity. Among them, we distinguish those experiencing housing insecurity, a category that includes homelessness in the strict sense as well as any other situation of precarious or inadequate housing. These categories are schematically represented below and described in more detail.
Scope of Restos data observation
Food insecurity
People supported by RDC, eligible according to their “rest for living” (a measure of purchasing power), and belonging to households that identify themselves as needing food assistance. These individuals may experience housing insecurity, be tenants in private or social housing, or, less frequently, homeowners.
Within this category
Housing insecurity
People who are homeless, do not have their own housing, or live in precarious or inadequate accommodation. These situations are reported in a mandatory field completed by the people receiving assistance (according to the ETHOS typology).
Most severe forms of housing insecurity
Homelessness in the strict sense
People living on the street or without their own housing, including those temporarily accommodated in emergency shelters (CHU), social reintegration accommodation centers (CHRS), or hotels. These situations are reported in a mandatory field completed by the people receiving assistance (according to the ETHOS typology).
Part one
Raw data analysis
The first part of our article highlights a number of descriptive statistics regarding the people received by RDC.
Raw data analysis · Question 1
1. How many people are affected by food or housing insecurity?
RDC conduct two registration campaigns each year (summer and winter), during which those receiving assistance (1.3 million in 2025) complete a questionnaire that determines their eligibility for food assistance. On average, RDC assist nearly 2% of the population residing in France. Among them, approximately 40% are in situations of housing insecurity (homeless in the strict sense of the term, or living in precarious or inadequate housing).
Key finding: 1 in 10 people exposed
In other words, nearly one-tenth of the people living in France are, at any given time, at risk of falling into food insecurity over the next decade.
Key finding: +62% more people experiencing housing insecurity between 2018 and 2025.
In terms of trends, we observe a sharp increase (+62%) in the number of people facing housing insecurity between 2018 and 2025 (averaged across the two survey waves each year). The trend is similar for the number of people assisted by RDC, which rose by 37% during the same period.
International comparison: change in the homelessness rate in France and England
People per 100 residents
Sources: France—Restos du Cœur (RDC) data; England—Ministry of Housing, Communities and Local Government.
One of the objectives of our article is to examine the factors that have driven this trend. The inflationary pressure of recent years is obviously one of the main contributing factors. Similarly, migratory movements linked to geopolitical crises—particularly the arrival of Ukrainian refugees—have also significantly contributed to the increase in the number of people receiving assistance.
Raw data analysis · Question 2
2. How does France rank among wealthy countries?
Comparing extreme poverty across countries is challenging, particularly when it comes to homelessness. Different countries do not use a common methodology to measure its extent—for example, some include all forms of housing insecurity, while others limit their definition to people sleeping on the streets. Furthermore, households in situations of housing insecurity are, by their very nature, difficult to identify and are therefore often underestimated. While keeping these caveats in mind, data published by the OECD allows us to draw up a comparative picture across countries in 2022.
Homelessness rate
The bar labeled “RDC” (0.42%) shows the percentage of individuals who are homeless in the strict sense, and the bar labeled “RDC+” (0.62%) shows the percentage of people in housing insecurity (including all those without a personal residence—for example, those living with relatives or in precarious or inadequate housing).
The figures provided by the OECD should be interpreted with caution: they only give a rough estimate of the total number of homeless people and are not directly comparable with those from RDC.
Key finding: France, represented in bright red, has one of the highest shares of homeless people (in the strict sense) in the OECD, estimated at 0.49% of the population.
Key finding: Using a methodology as close as possible to that of the OECD, we estimate that RDC welcome a significant portion of the number of homeless people reported by the OECD (0.42% of the population). When we include all forms of housing insecurity, we observe a much higher figure (0.62%).
Unless otherwise specified, the remainder of this note focuses on the subpopulation of French residents. This choice is based primarily on their uniform access to various social transfer payments (RSA, housing allowances, etc.) as well as to public housing.
Raw data analysis · Question 3
3. Which age groups are most at risk of extreme poverty?
Key finding: A key finding emerges from our dataset: children and teenagers are vastly overrepresented among those in extreme poverty (and even more so among those in housing insecurity), while older adults are, conversely, underrepresented. 25% of those experiencing housing insecurity are children aged 0 to 9, whereas this age group accounts for only 10% of the general population.
Age distribution
France
People assisted by RDC
Housing insecurity
All ages
Share of the population or group
Children and adolescents
detail for ages 0–18Share of the population or group
A closer look at the child and adolescent segment of the age pyramid shown on the left.
How to read this: This chart shows the age pyramid for our dataset (in red), for our subsample of people experiencing housing insecurity (in blue), and for France as a whole (in light gray). For example, just over 20% of children aged 0 to 9 are assisted by RDC, compared to 20–25% among those experiencing housing insecurity, even though this age group represents 10% of the French population.
Raw data analysis · Question 4
4. How long does an episode of extreme poverty last?
One of the most unexpected findings of our raw data analysis is that food insecurity (i.e., people assisted by RDC) and housing insecurity (which includes homelessness as well as all other forms of precarious housing) are both transitory phenomena.
Key finding: More specifically, the average duration of a stay at RDC is approximately 18 months, and the duration of an experience of housing insecurity is even shorter (17 months). Only 14% of individuals assisted by RDC remain beyond 3 years, and it is rare (about 10% of cases) for individuals to return after having been out for at least one year.
Food insecurity and homelessness.
In both cases, the largest share of spells is concentrated in the first campaign.
Food insecurity
Homelessness
Number of RDC campaigns
How to read this: Nearly 40% of food-insecurity episodes and just over 35% of homelessness episodes last only one campaign.
18 monthsaverage duration at RDC, measured by number of campaigns
17 monthsaverage duration of housing insecurity
Duration varies with age.
People under 25 spend fewer campaigns at RDC.
Under 25
Ages 25–64
Ages 65 and over
Number of RDC campaigns
How to read this: Nearly 45% of young people under 25, nearly 40% of adults between 25 and 64, and more than 35% of those over 65 stay at RDC for only one campaign.
21 monthsaverage duration after age 65, measured by number of campaigns
It is important to understand that these two statistics—a short-term but very widespread risk—are two sides of the same coin. The fact that people who use RDC stay for relatively short periods of time means that the sample is constantly changing, and that, consequently, many inhabitants will request food assistance sooner or later.
Raw data analysis · Question 5
5. What are the employment situations, incomes, and expenses of the households receiving assistance?
Data from RDC provides a very detailed overview of the employment and financial situations of those receiving assistance.
Income
90%of people assisted are not employed
Among all people assisted, and included in this 90%:
3%students
10%people with disabilities
11%retirees
First, most individuals (90%) are unemployed (note that this figure includes 3% students, 10% people with disabilities, and 11% retirees). This is, of course, partly due to the fact that it is precisely when they find themselves in this situation that households turn to RDC for assistance. As a result of this exclusion from the labor market, the primary source of income for those of working age who turn to RDC for help is social assistance, notably through the RSA (“Revenu de Solidarité Active”, a lump-sum cash payment akin to a UBI except that recipients must be older than 25), unemployment insurance, family and housing allowances, and disability benefits.
Share of income spent on housing or temporary shelter
45%Private-sector tenants
40%Public-housing tenants
20%People without a home of their own
People without a permanent home also spend a significant portion (20%) of their income on temporary shelters, hotel stays, etc., which may not be free.
In terms of expenses, rent is by far the largest expense: on average, it accounts for 45% of the income of tenants in the private rental market and 40% of that of tenants in the public housing sector. People without a permanent home also spend a significant portion (20%) of their income on temporary shelters, hotel stays, etc., which may not be free. Approximately 10% of people are in debt.
Key finding: Finally, one of the key findings is the relative stability of the beneficiaries’ disposable income—that is, their income after paying rent and associated expenses. Regardless of their housing situation (private housing, public housing, or housing insecurity), their disposable income stabilizes at around 300€ per consumption unit to cover all their basic expenses.
Part two
Determinants of Entry and Exit Flows
The second part of our analysis goes beyond these raw descriptive statistics and takes a more rigorous look at the determinants of entry and exit flows from food and housing insecurity. Since RDC centers are located throughout the country, we can correlate these entry and exit flows with geographic variations in local macroeconomic conditions in order to estimate their respective roles in the formation and persistence of poverty.
To this end, we have collected a wealth of annual data at the “living basin” level: unemployment rates, average income, food and rent inflation, the number of businesses and new business start-ups, industrial sectors, the availability of public housing, the capacity of emergency shelters, weather conditions, and so on. For exit flows from poverty, we are also able to estimate the impact of individual characteristics, such as household demographic and socioeconomic variables.
Our empirical analysis takes advantage of the fact that economic variables vary differently across areas over time; for example, the unemployment map does not coincide with the rent map, and it changes differently from year to year. This allows us to isolate and independently estimate the effect of each economic variable on the proportion of households that enter or exit food insecurity or homelessness each year.
For a concrete example of how we estimate these effects, the interactive map below shows the number of people experiencing homelessness per thousand households in each living basin in 2022–2023. It also displays other aggregates, including rent levels, the unemployment rate, and more.
Hover over an area for details; search for it to zoom in.
Leaflet · IGN municipality geometry · classes by augmented living basin
Winter 2022–2023
Building the atlas…
Key finding: Extreme poverty also has a strong geographical dimension. It is more prevalent in urban areas, particularly in city centers: approximately 50% of those affected live in city centers, compared to 30% of the total French population, which points to a trend of urban impoverishment. However, poverty is not exclusively an urban phenomenon: 30% of RDC distribution centers and nearly 20% of the people they assist are located in rural areas or isolated towns.
Determinants of entry and exit flows · Question 6
6. What are the macroeconomic determinants of entry flows?
The first important empirical finding we obtain is a quantification of the impact of economic shocks and certain public policies on entry flows into food and housing insecurity. We observe that these effects are very large and statistically significant. The main economic determinants of entry flows into extreme poverty are, on the one hand (with a positive sign), the unemployment rate and rent inflation, and on the other hand (with a negative sign), the average income in the living basin, the stock of public housing, and the number of businesses (particularly in certain sectors).
Key finding: For example, we show that food and housing insecurity are very responsive to fluctuations in the unemployment rate. In practical terms, this means that a 10% increase in the unemployment rate leads, all else being equal, to an increase of approximately 10% in the number of people who are homeless or experiencing housing insecurity, and a 13% increase in food insecurity.
The effects of several economic variables on housing insecurity are shown in the figure below. The right-hand panel can be used to simulate the impact of selected economic variables on entries into housing insecurity at the living-basin level.
Estimates of entries into housing insecurity.
Elasticities and displayed intervals
How to read this: The central coefficient is the one indicated by a pink dot—for example, 0.42 for private rents. This coefficient measures the percentage impact of a 1% increase in rent levels. To calculate the impact of a 10% increase, multiply this coefficient by 10.
If this interval does not contain the value 0, we say that the estimated impact is statistically significant; this means there is a greater than 90% chance that the coefficient is different from 0.
A local shock can quickly trigger new entries.
A change in:
A 10% increase in unemployment leads, all else equal, to a change in entries into housing insecurity of
≈ +10 %
Confidence interval: [+5.8%, +13%].
Determinants of entry and exit flows · Question 7
7. What are the macroeconomic determinants of exit flows?
Let us now consider individuals who have already fallen into poverty. What macroeconomic or public policy variables are likely to help them escape poverty? One of the most striking findings of our analysis is that no macroeconomic variable has a robust, economically or statistically significant impact on exit flows from food or housing insecurity. In other words—and this is the most important point of our study—recessions push households into extreme poverty at a much faster rate than the subsequent recovery helps them escape it. These results suggest that public policies targeting the labor and housing markets can help prevent people from falling into extreme poverty, but have little effect in helping those who have already fallen into it to escape; helping these individuals instead requires targeted interventions tailored to their personal circumstances.
Key finding: The correct interpretation of our results is that exit rates from extreme poverty are only very slightly responsive to fluctuations in economic activity; variations in food insecurity and housing insecurity over the course of the economic cycle (recessions and periods of growth) are largely determined by the inflow margin, even though their average level depends heavily on exit rates.
The effects of several economic variables on exits from housing insecurity are shown in the figure below on the left. The right-hand panel can be used to simulate the impact of selected economic variables on exits from housing insecurity at the living-basin level.
Estimates of exits from housing insecurity.
Elasticities and displayed intervals
How to read this: in this specification, if private rents rise by 1%, the number of people exiting housing insecurity rises by about 0.14%, all else equal.
An interval that does not cross zero is presented as statistically significant at a confidence level above 90%.
The same local changes have little effect on exits.
A change in:
A 10% increase in unemployment leads, all else equal, to a change in exits from housing insecurity of
≈ −1,2 %
Confidence interval: [−2.3%, −0.2%].
Determinants of entry and exit flows · Question 8
8. What are the individual determinants of exit flows?
If economic variables and policies do not promote exit from extreme poverty, what factors determine these flows? The key determinants are individual or household characteristics. First, an individual’s age has a significant effect on their chances of getting back on their feet. All other things being equal, individuals aged 65 and older have a probability of exiting the sample or residential insecurity that is 10 percentage points lower than that of young people under 25 (i.e., the percentage of those leaving food banks during a given campaign drops from 32% to 22%). Second, household income clearly plays an important role: an increase from 0€ to 500€ per month per consumption unit raises the chances of recovery by 10 percentage points, equivalent to the difference between ages 65 and 25. Finally, the most important determinant of the rate of exit from extreme poverty is the number of campaigns a household has already spent at RDC. Economists call this duration dependence: Having been a client of RDC for at least three years reduces exit rates by about 10 percentage points relative to a baseline of approximately 32% per season. Whatever the reason for this duration dependence, it means that public policies must seek to provide early and targeted assistance to households that have just fallen into poverty.
−10 points in the exit rate per campaign
The percentage of those leaving food banks during a given campaign drops from 32% to 22%.
+10 points in the exit rate per campaign
An increase from 0€ to 500€ per month per consumption unit raises the chances of recovery by 10 percentage points.
32% → 22% exit rate per campaign
Having been a client of RDC for at least three years reduces exit rates by about 10 percentage points relative to a baseline of approximately 32% per season.
Determinants of entry and exit flows · Question 9
9. What economic shocks account for the recent increase in housing insecurity?
Finally, we use our empirical results to answer the following question: To what extent did changes in inflation (prices and rents) and unemployment contribute to the rise in extreme poverty we have experienced in recent years? To this end, we propose the following exercise. We calculate the entry and exit flows into homelessness that would have occurred if one of these variables (for example, the price level) had remained at its 2018 level, while allowing the others to follow their actual trajectories.
Key finding: We conclude that, all else being equal, inflation is responsible for more than half (52%) of the dramatic increase in housing insecurity since 2018.
Inflation
Increase in housing insecurity since 2018
If prices had remained at their 2018 level, the model estimates that the increase in housing insecurity would have been 29%, instead of the 61% observed.
Share attributed to housing and food inflation
52%
Unemployment
Lower unemployment slowed the increase in housing insecurity
If unemployment had remained at its 2018 level, we would have 12% more homeless people today.
Additional increase avoided thanks to lower unemployment
12%
Conclusion
Prevent early. Provide sustained support.
Data from RDC allows us to provide new empirical evidence on extreme poverty in a developed market economy such as France’s. Prevention is the most powerful lever of public policy, particularly through labor market stability, the availability of affordable housing, and price stability. For households that have already fallen into extreme poverty, aggregate economic conditions offer little help; what matters is providing early and targeted support before duration dependence makes it increasingly difficult to escape food and housing insecurity.
Let us conclude with a subtle but important point regarding the measurement of poverty. Beneficiaries of the Restos du Cœur are not admitted based on income criteria but according to a “rest for living” criterion—that is, a measure of consumption. By definition, they belong to households that identify themselves as requiring food assistance. This criterion allows us to study poverty in a much more relevant way than by relying solely on income data, which can be misleading (for example, the same income may be sufficient to cover housing costs in a rural area but not in Paris).
Lessons for public policy:
Prevent entries
Stabilize employment, contain unavoidable expenses, and expand the supply of affordable housing.
Act without delay
Duration dependence reduces the chances of exit. The analysis supports early intervention.
Target individual situations
Exits depend more on age, income, and the duration of the spell than on local economic conditions.