Homelessness & Empty Homes - Trends & COVID-19 Impact

Homelessness and vacant homes are a serious problem in the US, costing local and federal governments and agencies billions of dollars each year in lost tax revenues and social spending.

The White House reports that as of 2019, over half a million Americans don’t have a home to sleep in on any given night, while almost 17 million potential homes were standing empty. If the overall numbers of homeless citizens weren’t shocking enough, between 2017 and 2019, there was an increase of over 34,000 unsheltered homeless people nationally - even before a global pandemic and expected recession.

All this, while the number of empty properties around the country has increased by over 1.1 million since 2010, leaving over 12% of all housing units in the US vacant as of the latest figures in 2018.


Key findings from our research

  • There are 59 empty properties for each homeless person in the US, an increase of 43% since 2010
  • Since 2010, the number of empty properties per homeless person has increased 24%
  • Nationally, 12.3% of housing units in 2018 were vacant
  • In 2019, 555,672 US citizens were confirmed homeless
  • Since 2010 there has been a 7.2% increase in the vacant housing units
  • Latest figures reveal there are 16,987,623 vacant homes in the US
  • In 2019, California had a homeless population exceeding 151,000, that’s more than 18 times the homeless average across all other states (7,995 per state)
  • California also had 1.2m empty properties in 2018. That’s more than 9 empty properties per homeless person.
  • 63% across the country believe they are at risk of being homeless if they lose their income, with 49% having less than one months worth of savings to pay their rent/mortgagee
  • Millions at risk of being homeless with 51% thinking they’ll lose their jobs as a result of the pandemic

What is an ‘empty home’?

An empty home is what the government refers to as a ‘vacant housing unit’, which, for whatever reason, is not occupied by a citizen. This can include vacation homes but also properties which have sold but are not occupied. This is according to the United States Census Bureau, where all property data for this analysis was collected. You can see all exceptions and inclusions to property types in the official government definition.

Our research process

We used data taken from official government sources between 2010 and 2019, to understand whether how much of the US’ housing stock was currently vacant and how this compared to the number of homeless citizens across the country. The map below shows:

  • Number of homeless people per state
  • Number of empty properties per state
  • The ratio of homelessness and vacant properties per state
  • How this has changed per state since 2010

In addition to our research into the levels of homelessness and vacant housing units since 2010, we have also taken a look at how the COVID-19 pandemic could affect these numbers and the fears and worries of US citizens in relation to affording their homes in these trying times.

Use the map below to click through each year to see how each state compares, as well as how the actual numbers of empty homes and homelessness have changed since 2010.


2018 - Top 5 states with highest number of vacant properties:

  1. Florida - Over 1.73m
  2. Texas - Over 1.32m
  3. California - Over 1.2m
  4. New York - 996,832
  5. North Carolina - 673,500

2010 - Top 5 states with highest number of vacant properties:

  1. Florida - Over 1.71m
  2. Texas - Over 1.17m
  3. California - Over 1.15m
  4. New York - 845,095
  5. Michigan - 685,683

2018 - States with highest ratio of homeless people to empty properties:

  1. Mississippi has an average of 166 empty homes per 1 homeless person
  2. Indiana has an average of 151 empty homes per 1 homeless person
  3. West Virginia has an average of 127 empty homes per 1 homeless person
  4. Alabama has an average of 122 empty homes per 1 homeless person
  5. Louisiana has an average of 110 empty homes per 1 homeless person

2010 - States with highest ratio of homeless people to empty properties:

  1. Indiana had an average of 133 empty homes per 1 homeless person
  2. West Virginia had an average of 98 empty homes per 1 homeless person
  3. South Carolina had an average of 77 empty homes per 1 homeless person
  4. Maine had an average of 68.6 empty homes per 1 homeless person
  5. Delaware had an average of 68.5 empty homes per 1 homeless person

2018 - States with highest number of homeless people:

  1. California has 129,972 homeless people on any given night (5.3% increase)
  2. New York has 91,897 homeless people on any given night (40.1% increase)
  3. Florida has 31,030 homeless people on any given night (85.5% increase)
  4. Texas has 25,310 homeless people on any given night (38.8% increase)
  5. Massachusetts has 20,068 homeless people on any given night (20.6% increase)

2010 - States with lowest number of homeless people:

  1. California had 123,480 homeless people on any given night
  2. New York had 65,606 homeless people on any given night
  3. Florida had 57,551 homeless people on any given night
  4. Texas had 35,121 homeless people on any given night
  5. Georgia had 19,836 homeless people on any given night

Top 5 states where homelessness has increased since 2010

  1. South Dakota - increase of 58.6%
  2. New York - increase of 40.1%
  3. Massachusetts - increase of 20.6%
  4. Hawaii - increase of 12%
  5. Wyoming - increase of 10.4%

Top 5 states to reduce homelessness since 2010

  1. Louisiana - reduction of 75.5%
  2. Georgia - reduction of 52.1%
  3. Mississippi - reduction of 50.7%
  4. Florida - reduction of 46.1%
  5. West Virginia - reduction of 45.1%

Top 5 states where number of vacant properties has increased since 2010

  1. North Dakota - increase of 61%
  2. Hawaii - increase of 30.1%
  3. Wyoming - increase of 29.3%
  4. Alabama - increase of 29%
  5. Mississippi - increase of 28.7%

Top 5 states where number of vacant properties has reduced since 2010

  1. West Virginia - reduction of 28.5%
  2. Nevada - reduction of 13.7%
  3. Arizona - reduction of 6.2%
  4. Georgia - reduction of 5.2%
  5. Michigan - reduction of 4.2%
Show Top Stats Per State

If these figures in our map aren’t concerning enough, 2020 and the Coronavirus has created a whole new set of worries when it comes to housing, affordability and homelessness.

Potential COVID-19 Impact On Vacant Homes And Homelessness

Our research showed that between 2010 and 2018, there had been a 13.6% decrease in the number of homeless people in the US. However, in 2020 the economy and people's finances have been rocked by the COVID-19 pandemic.

To get an idea on how the pandemic might affect the number of vacant homes and homeless in the US, we conducted a survey of 1,090 US citizens, asking them how fearful they were about being able to afford their homes and whether a deterioration in their incomes could lead to them becoming homeless.

“More than half of Americans fear they will lose their job due to COVID-19 pandemic”

Many businesses across the US have seen huge drops in their revenue during 2020, with almost $900 billion being offered to businesses by the US government, in an attempt to keep them afloat and their staff in employment; in addition to the over $600 billion budget for the individual stimulus packages - view our insights into how people have been spending their checks.

However, with the vast majority of funds made available via the CARES Act being used up, many businesses face having to make some hard decisions when it comes to their employee’s contracts and millions of citizens fear not being able to keep a roof over their head.

According to our survey, over half (50.8%) of working-age Americans in employment are concerned they might lose their job through the pandemic; meaning 78 million (based on 2019 numbers) workers in the US fear they could lose their job due to effects of the Coronavirus pandemic.

However, our data found that this fear is more prevalent in those earning under $49,764 (the US median annual salary), with 56.3% of respondents earning under the nation’s median wage revealing that they are worried they might lose their income.

In comparison, this fear is shared with just 45.8% of Americans that make over $49,764 per year and fewer than a third (31%) of those that make more than double the nation's median annual wage ($100,000+). Showing a clear trend that those in lower pay brackets are the most likely to be worried about being laid off.

It’s no wonder people are fearful of losing their jobs in the US. According to our data, over 76 million (48.8%) American workers have less than a month’s worth of savings should they lose their job, with 1 in 10 (9.4%) people saying they would struggle after just a week.

However, with the average job search taking 9 weeks (in normal times), our data found that just 51 million (32.9%) of the US’ 156 million workers have enough in savings to keep them afloat between losing and starting a new job.

According to our data, renters are the most vulnerable to running out of savings within a month. With more than 43 million occupied rental units in the US, our surrey found that over half (54.5%) of rental residents would run out of savings within a month should they lose their employment; with 1 in 4 (24.5%) struggling within 2 weeks

In comparison, just 29% of people who had either already paid off their mortgage or never had one said they’d run out of savings after a month.

Through the brunt of the pandemic, several protections were created federally and by states to protect renters and those paying mortgages. However, like many of the other government actions, their time frames are starting to run out and struggling citizens could soon find themselves without a roof over their heads.

Even before these protections and support systems began running out, our study found that almost 1 in 3 (30.1%) Americans were unaware of any state support on offer and fewer than a third (32.4%) had actually used any form of state support during the COVID-19 pandemic, in comparison to 38.4% using federal help.

“Over half of Americans fear they will become homeless if they lose their job”

With so few American’s having enough in savings to keep them above water in the event that they lose their job, our data found that over half (55.3%) of working Americans feel that they are at risk of becoming homeless should they experience a reduction in their income.

This fear of becoming homeless rises to close to two thirds (64.2%) within renters - equating to almost 22 million households according to the latest rental occupation figures - and just over half (51.5%) of Americans who are paying a mortgage.

If even just a fraction of these renters or mortgage payers were to see their income drop, we could see the number of homeless people and vacant homes explode across the US.

Projections from March estimated that unemployment, during and due to the COVID-19 pandemic, could reach 30% - over 50 million. If these projections are even remotely close, based on our data, more than 25 million people face losing their homes due to the drop in income and lack of financial support.

It’s unlikely that 25 million people would end up on the streets, however, with such a huge number facing the possibility, just a small fraction of this 25 million becoming reality could see the homeless numbers in the US explode from just over 555,000 nationally to the millions.

With increased numbers of people “on the streets” local, state and federal governments will see budget deficits increase exponentially through the drop in property and residential taxes. Even for other citizens, even a slight increase in the number of people losing their homes could see property values drop due to increasing vacancy, an increase in anti-social and criminal behaviours, and drop in local infrastructure repairs and projects (due to the lack of local funds and budget restrictions).


To conduct our research, we collected the number of occupied and unoccupied properties per state for each year between 2018 (most recent available data) and 2010 using data.census.gov. We also collected the historical data for homelessness statistics using Whitehouse.gov. Our ratios were then simply calculated by dividing the number of empty properties by the number of homeless people in each state for a given year.

In addition, for our survey, we surveyed 1,090 US citizens via the Amazon Mechanical Turk program. Respondents were surveyed between July 28th and August 1st 2020.

Demographics for respondents were:

Female 51.6%
Male 47.9%
Other 0.4%
Prefer not to say 0.2%
18 - 24 9.1%
25 - 34 39.9%
35 - 44 27.5%
45 - 54 13.7%
55 - 64 6.7%
65+ 3.1%


  • https://www.visualcapitalist.com/the-anatomy-of-the-2-trillion-covid-19-stimulus-bill/
  • https://www.self.inc/info/stimulus-check-survey/
  • https://www.statista.com/statistics/269959/employment-in-the-united-states/#:~:text=In%202018%2C%20around%20155.76%20million,million%20employed%20people%20is%20expected.
  • https://www.statista.com/statistics/187577/housing-units-occupied-by-renter-in-the-us-since-1975/#:~:text=In%202019%2C%20there%20were%20approximately,vacancy%20rates%20across%20the%20country.
  • https://www.indeed.com/career-advice/finding-a-job/how-long-does-it-take-to-find-a-job#:~:text=The%20average%20amount%20of%20time,be%20searching%20for%20a%20job.
  • https://www.statista.com/statistics/187577/housing-units-occupied-by-renter-in-the-us-since-1975/