I believe I know why California has fewer costal vacation homes than Florida and the East coast. The state’s costal commission (the same one that Elon Musk got into a legal tiff with) is extremely NIMBY and makes all sorts of construction hard, even by CA standards. Pair that with a complete ban on the construction of houses on the beach that was put in place back in the 70s, and there’s a cap on the most desirable vacation homes, along with a challenging environment for building houses near the ocean.
I suspect that another big reason is that in California, a higher fraction of the population is already near the coast. If you live in San Francisco or Los Angeles or San Diego, there’s no need for a beach house, because you’re already pretty close - but you might want a mountain house. Whereas if you live in Atlanta or Charlotte or Philadelphia or DC or Buffalo, a beach house might make more sense.
San Franciscans definitely want houses at Tahoe, in wine country, and/or in the beautiful forested areas, either coast-adjacent redwoods or Sierra mountain pines. Each has its own supply restrictions: in Tahoe the watershed protection authority imposes a de facto cap similar to the coast; wine country is zoned to protect viticulture IIRC; and the forests are so fire prone as to make homes there very difficult to insure.
the "vacation homes" in Manhattan are 8- and 9-figure condos that serve as pied a terres (or straight up investment vehicles) for the ultra wealthy who don't have them as primary residences to avoid NYC income income tax
The number would be higher if it did, this looks like a pretty straight occupancy analysis. But there's also a long history of rich New Yorkers basically only spending part of the year in the city. The Roosevelts spent summer in Canada, Fall in Hyde Park, winter in the city, etc. etc.
I'm sitting in a cabin on the shore of a relatively small lake in northern Minnesota, so naturally I found this post very interesting. Two points: first, the "Arrowhead" region of Minnesota, where I am, is also where the Boundary Waters are located, so most of the land is off limits to any development at all. Second, you're absolutely correct about building restrictions. The state and county have VERY severe regulations designed to protect the lakes, and they rigorously enforce them. The big surprise I had when I first moved here was that the St. Louis county government is in many ways the Home Owners Association from hell. Woe betide anyone who builds near a lake without first getting the (expensive) green light from the folks down in Duluth.
I think this probably says less about elite consumption and impact on housing then it seems at first. The challenge with this is the definition of vacation home is a bit squishy. The census.gov link says their count includes fishing cabins and migrant housing. A "Vacation home" sounds like a rich person luxury, but I am guessing a lot of these are not places you would want to live long term. This is why there is a high percentage in low density states (Montana, Wisconsin, Maine, Alaska, etc). Land is cheap and a middle income family can have a rustic cabin on a lake a couple hours out of town for very little money. Lots of them shared by families, inherited and shared between adult siblings, etc. Your family can spend a couple weeks/weekends at the lake each summer for way less than other types of vacations. But this is a step above camping, and you couldn't really live at these locations.
A more interesting analysis would filter out these low value properties. Maybe look at the number of "vacation homes" that are at or above the median home value for the state.
Does a house used as an AirBNB count as a "vacation home"? I mean, to me it seems like basically the same thing, whether you own a house and rent it out 80% of the year, versus it's just a property management company that owns it and rents it out 100% of the year, it's basically the same. But I don't know how you'd count it in your data.
Is the definition here of vacation home meaning "owned by one person" or can a vacation home be a time share/VRBO/AirBnB? There is "roughly one home for every 2.4 people in the country". Sounds real but I would think home ownership is not at that rate, or at least 1 in every 2.4 people I know don't own their home. It would be interesting to see the data on what % of people own more than one vacation home, and how many does one person own if they have more than one.
That would depend upon someone's definition of vacation. Mine might be different from yours, or my wife's. My idea of a vacation house would be a small three or four bedroom house fifty miles from a major city, and somewhere ten miles away from a small town. Someplace that I can still access the net and interact, but someplace where I can see the stars and there's not a soul around.
My wife's idea of a vacation home would be fifteen miles outside a major city, where she could go shopping and enjoy new things.
I wrote this for VtDigger, our online newspaper here in Vermont, although they chose not to publish it. Here in Vermont housing prices are just completely out of control and I think vacation, or at least, second homes are part of the problem. And I'm not sure this gets reflected in the census data, at least as for as I could tell.
About thirty years ago, shortly after I became a property owner, I lost an argument with the zoning board in my town, East Montpelier.
The issue pertained to setbacks for building additions and the regulation stated that these needed to be 75 feet for back and side property lines in my district. Since neither of the two adjoining property owners objected to my building plans, I thought a variance would be easy, but I was wrong and denied a permit.
I had to modify my building plans in ways I felt made no sense. Several years later the town changed its setback requirements to 50 feet which would have more than accommodated my old plans, but I had already gone ahead and built something conforming to the old zoning requirements. The experience embittered me and convinced me that zoning mostly consisted of pointless restrictions which make it harder for people to get a start in life.
Fast forward to today. My kids, who are a little younger than I was when I got in the skirmish with the zoning board, have both struggled to find housing. I have been blaming Vermont’s housing shortage on the zoning restrictions, from Act 250 to local zoning, which make it very difficult to build new housing, especially affordable housing.
But I am beginning to think that the housing problem is bigger and more difficult than simply lightening up on zoning. Don’t get me wrong, regulations which lack common sense are a big problem in Vermont and they make it harder to build houses here. But even if we could magically do away with all building regulations, I don’t think it would solve the housing shortage.
The problem is more about the financialization of housing than it is a shortage of houses. Here’s how I came to this conclusion.
1- I noticed that the housing crisis and affordability crisis is spread across the entire country, including places like Texas with more lenient zoning.
2- The housing problem isn’t even confined to this country but is also the case in Australia, Canada and New Zealand.
3- I talked to a woman who told me that where she lives in central Mass., 70% of the houses in the town are second homes.
4- I added up the number of houses on our fairly typical, rural Vermont road, 21, and the number of second homes which are not rented, 2, and the number of homes that cater to short term rentals, also 2. So that's roughly 10% of the houses vacant except when the second home owner is there, and another 10% of the homes used for short term rentals.
5- I asked the Vermont tax department if they could provide me with the number of second homes in Vermont and how that number has changed over time. Short answer, no, they don’t have that data.
6- I had a brief conversation with our town clerk while paying my property taxes. The empty houses I knew about were categorized as “seasonal” properties on the grand list, same category as hunting camps. In other words, even at the town level it is hard to ferret out how many of Vermont’s four-season, habitable houses are sitting empty, and, more importantly, how this number has changed over the years.
7- According to the US Census data between 2010 and 2023 (the 2023 numbers are estimates based on extrapolations from the 2020 census, I think) the Vermont population grew by 3.5% but the total housing stock grew by almost 6%.
We’ve all seen the profusion of out-of-state license plates and heard the stories about neighbor’s houses being purchased by people who viewed the seemingly high prices as reasonable and were willing to pay cash.
The problem is that houses have become another financial asset. As the country and world were flooded with money after the Great Financial Crisis of ‘08/09, and then again during Covid, a lot of that money inevitably ended up in the hands of wealthy people looking to invest it. Obviously some of it went to the stock market where, by some measures, prices are higher than they were before the crash of 1929, but people want to diversify their investments and second houses are an obvious choice. I don’t blame them. I might buy another house if I had enough extra money.
But this means that no matter how many houses we build, if we are just creating investable “assets” for people in the rest of the world, we will never be able to build enough houses to increase supply enough to bring prices down to a level that regular people can afford. The world is too vast and the amount of money created is too vast and Vermont is way, way too small to solve this problem by increasing the supply of houses here. Even building “affordable” houses, while helpful to the people who can qualify for them, doesn’t solve the problem because it simply makes more other houses available for people to buy as investments.
Six years ago I drove across the country with my son (you can read about that in Digger archives here). We drove up to Crested Butte, Colorado which he wanted to see because he had friends from Vermont who were working there. The hillsides as we neared Crested Butte were covered with huge ski houses, rows and stacks of them, mostly brown and fitting into the landscape pretty well considering how many there were.
We spent the night in a makeshift campground set up below the ski area to accommodate the workers catering to the summer people. There wasn’t enough affordable housing for the workers in town. Shop owners in town told us that almost all of the houses we saw lining the hillsides were empty and owned by rich people, mostly from California or Texas, who stayed in them a week or two a year.
Crested Butte seems like a cautionary tale for Vermont. We can’t build our way out of our housing crisis. This is not to say that we don’t need to build more housing, but simply increasing the supply of housing won’t be all it takes to solve this problem.
Nice Work... one observation is that vacation homes are "solar" of real estate. That is, they are highly seasonal and quite dependent on the fluctuations of the economy. This combination makes them very difficult from an infrastructure planning perspective. Let me give you an example.... 30A along the Florida's panhandle has a massive concentration of vacation homes (Seaside of Truman Show fame is located there). Ok... 30A is a two lane road which most of the year is perfectly fine. However, in high season... it is a parking lot. Cape Cod in new England is similar with Rt 6 and 28.
Managing this seasonality and infrastructure takes some innovation. When I was working with Seaside, they were looking at blocking 30A and offering constantly running AV shuttles as an alternative.
The big takeaway for me is that vacation homes and residential homes are simply in different places. Michigan’s vacation homes are all in the most sparce least expensive parts of the state for example.
For the data on seasonal housing at the level of census tracts, I wonder if those data are available over time and, if those were plotted on a map, would they show the expansion of metropolitan areas. There are any number of counties at the periphery of metro areas, such as in CT, NJ and NH, that have transitioned or are transitioning from vacation land to commuter land. This could be an interesting visual confirmation of that dynamic at a census tract level.
I have some skepticism of the Illinois and Big City numbers that might be a bit revealing. One problem is how "primary" are the houses that pair with "vacation" homes? I suspect that many people who actually spend more time in Wisconsin than Chicago claim their Chicago home as primary to get lower property tax. Rules like that probably create a lot of distortions where people have a choice on which to claim as primary.
On one side sometimes claiming as primary will force you to pay income tax. In others, it gets you a homestead property tax reduction, or avoids a vacation surcharge. And if you have choice about a property tax break you'd always apply it to the higher valued property.
To avoid or at least be somewhat aware of those I'd want to know where the partially occupied primary homes were.
I believe I know why California has fewer costal vacation homes than Florida and the East coast. The state’s costal commission (the same one that Elon Musk got into a legal tiff with) is extremely NIMBY and makes all sorts of construction hard, even by CA standards. Pair that with a complete ban on the construction of houses on the beach that was put in place back in the 70s, and there’s a cap on the most desirable vacation homes, along with a challenging environment for building houses near the ocean.
I suspect that another big reason is that in California, a higher fraction of the population is already near the coast. If you live in San Francisco or Los Angeles or San Diego, there’s no need for a beach house, because you’re already pretty close - but you might want a mountain house. Whereas if you live in Atlanta or Charlotte or Philadelphia or DC or Buffalo, a beach house might make more sense.
San Franciscans definitely want houses at Tahoe, in wine country, and/or in the beautiful forested areas, either coast-adjacent redwoods or Sierra mountain pines. Each has its own supply restrictions: in Tahoe the watershed protection authority imposes a de facto cap similar to the coast; wine country is zoned to protect viticulture IIRC; and the forests are so fire prone as to make homes there very difficult to insure.
Just compare the California Coastline with Florida’s and you will understand the difference.
the "vacation homes" in Manhattan are 8- and 9-figure condos that serve as pied a terres (or straight up investment vehicles) for the ultra wealthy who don't have them as primary residences to avoid NYC income income tax
i'm curious if the definition of "vacation home" includes those held by LLCs (either to obscure ownership or because they are primarily rentals)
The number would be higher if it did, this looks like a pretty straight occupancy analysis. But there's also a long history of rich New Yorkers basically only spending part of the year in the city. The Roosevelts spent summer in Canada, Fall in Hyde Park, winter in the city, etc. etc.
I'm sitting in a cabin on the shore of a relatively small lake in northern Minnesota, so naturally I found this post very interesting. Two points: first, the "Arrowhead" region of Minnesota, where I am, is also where the Boundary Waters are located, so most of the land is off limits to any development at all. Second, you're absolutely correct about building restrictions. The state and county have VERY severe regulations designed to protect the lakes, and they rigorously enforce them. The big surprise I had when I first moved here was that the St. Louis county government is in many ways the Home Owners Association from hell. Woe betide anyone who builds near a lake without first getting the (expensive) green light from the folks down in Duluth.
I think this probably says less about elite consumption and impact on housing then it seems at first. The challenge with this is the definition of vacation home is a bit squishy. The census.gov link says their count includes fishing cabins and migrant housing. A "Vacation home" sounds like a rich person luxury, but I am guessing a lot of these are not places you would want to live long term. This is why there is a high percentage in low density states (Montana, Wisconsin, Maine, Alaska, etc). Land is cheap and a middle income family can have a rustic cabin on a lake a couple hours out of town for very little money. Lots of them shared by families, inherited and shared between adult siblings, etc. Your family can spend a couple weeks/weekends at the lake each summer for way less than other types of vacations. But this is a step above camping, and you couldn't really live at these locations.
A more interesting analysis would filter out these low value properties. Maybe look at the number of "vacation homes" that are at or above the median home value for the state.
Does a house used as an AirBNB count as a "vacation home"? I mean, to me it seems like basically the same thing, whether you own a house and rent it out 80% of the year, versus it's just a property management company that owns it and rents it out 100% of the year, it's basically the same. But I don't know how you'd count it in your data.
Is the definition here of vacation home meaning "owned by one person" or can a vacation home be a time share/VRBO/AirBnB? There is "roughly one home for every 2.4 people in the country". Sounds real but I would think home ownership is not at that rate, or at least 1 in every 2.4 people I know don't own their home. It would be interesting to see the data on what % of people own more than one vacation home, and how many does one person own if they have more than one.
Enjoyable and informative article as always!
Curious correlation with nordic ancestry around the great lakes, a people famed for their love of vacation cabins by lakes
That would depend upon someone's definition of vacation. Mine might be different from yours, or my wife's. My idea of a vacation house would be a small three or four bedroom house fifty miles from a major city, and somewhere ten miles away from a small town. Someplace that I can still access the net and interact, but someplace where I can see the stars and there's not a soul around.
My wife's idea of a vacation home would be fifteen miles outside a major city, where she could go shopping and enjoy new things.
I wrote this for VtDigger, our online newspaper here in Vermont, although they chose not to publish it. Here in Vermont housing prices are just completely out of control and I think vacation, or at least, second homes are part of the problem. And I'm not sure this gets reflected in the census data, at least as for as I could tell.
About thirty years ago, shortly after I became a property owner, I lost an argument with the zoning board in my town, East Montpelier.
The issue pertained to setbacks for building additions and the regulation stated that these needed to be 75 feet for back and side property lines in my district. Since neither of the two adjoining property owners objected to my building plans, I thought a variance would be easy, but I was wrong and denied a permit.
I had to modify my building plans in ways I felt made no sense. Several years later the town changed its setback requirements to 50 feet which would have more than accommodated my old plans, but I had already gone ahead and built something conforming to the old zoning requirements. The experience embittered me and convinced me that zoning mostly consisted of pointless restrictions which make it harder for people to get a start in life.
Fast forward to today. My kids, who are a little younger than I was when I got in the skirmish with the zoning board, have both struggled to find housing. I have been blaming Vermont’s housing shortage on the zoning restrictions, from Act 250 to local zoning, which make it very difficult to build new housing, especially affordable housing.
But I am beginning to think that the housing problem is bigger and more difficult than simply lightening up on zoning. Don’t get me wrong, regulations which lack common sense are a big problem in Vermont and they make it harder to build houses here. But even if we could magically do away with all building regulations, I don’t think it would solve the housing shortage.
The problem is more about the financialization of housing than it is a shortage of houses. Here’s how I came to this conclusion.
1- I noticed that the housing crisis and affordability crisis is spread across the entire country, including places like Texas with more lenient zoning.
2- The housing problem isn’t even confined to this country but is also the case in Australia, Canada and New Zealand.
3- I talked to a woman who told me that where she lives in central Mass., 70% of the houses in the town are second homes.
4- I added up the number of houses on our fairly typical, rural Vermont road, 21, and the number of second homes which are not rented, 2, and the number of homes that cater to short term rentals, also 2. So that's roughly 10% of the houses vacant except when the second home owner is there, and another 10% of the homes used for short term rentals.
5- I asked the Vermont tax department if they could provide me with the number of second homes in Vermont and how that number has changed over time. Short answer, no, they don’t have that data.
6- I had a brief conversation with our town clerk while paying my property taxes. The empty houses I knew about were categorized as “seasonal” properties on the grand list, same category as hunting camps. In other words, even at the town level it is hard to ferret out how many of Vermont’s four-season, habitable houses are sitting empty, and, more importantly, how this number has changed over the years.
7- According to the US Census data between 2010 and 2023 (the 2023 numbers are estimates based on extrapolations from the 2020 census, I think) the Vermont population grew by 3.5% but the total housing stock grew by almost 6%.
We’ve all seen the profusion of out-of-state license plates and heard the stories about neighbor’s houses being purchased by people who viewed the seemingly high prices as reasonable and were willing to pay cash.
The problem is that houses have become another financial asset. As the country and world were flooded with money after the Great Financial Crisis of ‘08/09, and then again during Covid, a lot of that money inevitably ended up in the hands of wealthy people looking to invest it. Obviously some of it went to the stock market where, by some measures, prices are higher than they were before the crash of 1929, but people want to diversify their investments and second houses are an obvious choice. I don’t blame them. I might buy another house if I had enough extra money.
But this means that no matter how many houses we build, if we are just creating investable “assets” for people in the rest of the world, we will never be able to build enough houses to increase supply enough to bring prices down to a level that regular people can afford. The world is too vast and the amount of money created is too vast and Vermont is way, way too small to solve this problem by increasing the supply of houses here. Even building “affordable” houses, while helpful to the people who can qualify for them, doesn’t solve the problem because it simply makes more other houses available for people to buy as investments.
Six years ago I drove across the country with my son (you can read about that in Digger archives here). We drove up to Crested Butte, Colorado which he wanted to see because he had friends from Vermont who were working there. The hillsides as we neared Crested Butte were covered with huge ski houses, rows and stacks of them, mostly brown and fitting into the landscape pretty well considering how many there were.
We spent the night in a makeshift campground set up below the ski area to accommodate the workers catering to the summer people. There wasn’t enough affordable housing for the workers in town. Shop owners in town told us that almost all of the houses we saw lining the hillsides were empty and owned by rich people, mostly from California or Texas, who stayed in them a week or two a year.
Crested Butte seems like a cautionary tale for Vermont. We can’t build our way out of our housing crisis. This is not to say that we don’t need to build more housing, but simply increasing the supply of housing won’t be all it takes to solve this problem.
Nice Work... one observation is that vacation homes are "solar" of real estate. That is, they are highly seasonal and quite dependent on the fluctuations of the economy. This combination makes them very difficult from an infrastructure planning perspective. Let me give you an example.... 30A along the Florida's panhandle has a massive concentration of vacation homes (Seaside of Truman Show fame is located there). Ok... 30A is a two lane road which most of the year is perfectly fine. However, in high season... it is a parking lot. Cape Cod in new England is similar with Rt 6 and 28.
Managing this seasonality and infrastructure takes some innovation. When I was working with Seaside, they were looking at blocking 30A and offering constantly running AV shuttles as an alternative.
The big takeaway for me is that vacation homes and residential homes are simply in different places. Michigan’s vacation homes are all in the most sparce least expensive parts of the state for example.
For the data on seasonal housing at the level of census tracts, I wonder if those data are available over time and, if those were plotted on a map, would they show the expansion of metropolitan areas. There are any number of counties at the periphery of metro areas, such as in CT, NJ and NH, that have transitioned or are transitioning from vacation land to commuter land. This could be an interesting visual confirmation of that dynamic at a census tract level.
I have some skepticism of the Illinois and Big City numbers that might be a bit revealing. One problem is how "primary" are the houses that pair with "vacation" homes? I suspect that many people who actually spend more time in Wisconsin than Chicago claim their Chicago home as primary to get lower property tax. Rules like that probably create a lot of distortions where people have a choice on which to claim as primary.
On one side sometimes claiming as primary will force you to pay income tax. In others, it gets you a homestead property tax reduction, or avoids a vacation surcharge. And if you have choice about a property tax break you'd always apply it to the higher valued property.
To avoid or at least be somewhat aware of those I'd want to know where the partially occupied primary homes were.
Always fun.
No mention of AirBnB/rentals in the article. I thought that would've had some effect that shows up in the data.
The data that you linked seems to only go up to 2000. Did you use other census data to derive this?