When we left off, over a few short years mobile home manufacturing had grown into a massive industry, and was supplying an increasingly large fraction of the US’s housing.
I share your current working theory in part. The chattel collapse of 2000 remains partly responsible for tepid MH unit sales. Following 2000, the sub-prime run-up to 2008 drew many traditional MH buyers into unsustainable (it turned out) single family homes. Then the crash, credit tightening and resulting regulatory response to 2008 impacted MH chattel lending and home sales. But, what explains the last 5 - 7 years as housing demand really heated up?
From here, I go to zoning and planning approvals first. The number of MH communities being built roughly equals the number being closed down. Second, as someone who talks with a lot of would-be buyers, the MH community industry has a water cooler problem - i.e. what people say when you announce, "we are looking at a mobile home."
Dave Ramsey calls MH a depreciable asset and asks "why would you do that?" MH community closures put people out on the street - e.g. in Indiana with just 60 days notice - and subject people to what the Washington Post wrote were "massive rent increases." It all contributes to reluctant consumers and that narrows demand.
My view: The industry is in need of structural change. Long-term leases and access to residential mortgage loans (i.e. home only mortgages on leasehold land) would be one why to say and mean that these are permanent homes and worthy of residential mortgage loans. Think about MHCs as "true lease" communities and present that paradigm to town planners and the buying public. Our work is in helping homeowners form co-ops to buy their communities. That's the surest way to ensure that homeowners have long-term security and affordability. www.ROCUSA.org
I’m a little dubious about that 50,000 MH community number. I suspect the figure comes from a Homeland Security database of communities, which includes a lot of parks that no longer exist and many are under 25 pads.
The financing is a huge component of stagnant production numbers, and the personal property vs real property makes capital for the financing more expensive. The public policy around MH, making them difficult to qualify for rent support, exclusionary zoning, lower tax revenue (because of the personal v real property issue) conspire to make them a less attractive housing option for public support. Finally, outside of retirement communities, they’re invisible to most people and therefore no one recognizes the affordable component of these communities.
There’s also the issue of capital for park owners and Fannie/Freddie debt availability, but that’s a much longer comment.
It's possible to accept that those particular upswings and downswings were caused by loosening of lending standards, and a predictable collapse from poor standards.
That said, it doesn't answer the question of why manufactured homes, given their significant cost advantage aren't more prevalent, and haven't become more and more prevalent as the gap between the stick built homes and manufactured homes has become wider?
In addition, a part of the explanation for the the impacts of lending bubble/crash is incomplete without looking at how repossession law differs between stick built and the majority of manufactured homes. If a manufactured home remains classified as personal property, repossession is less onerous than if classified as real property.
Public impression also seems likely to play a role. That perception isn't helped by the chassis requirement. Not only does it make manufactured homes even more obviously distinct, it doesn't make sense in a way that would cause any prospective buyer to question the entire premise and wonder what other non-sense this class of home would get them involved in. It's interesting that it can almost entirely cancel the value of the other HUD requirements, in that the perception of the post 1974 home is more closely linked to the performance of the pre-1974 home. Those improvements are fairly invisible in comparison to the chassis in the basement.
I think what need to be added to the discussion is how easy (at times) it has been to finance homes, and at historically low rates.
There has been a general rise in the size/pricing of stick built homes. It has been shown that lending is inversely related to pricing. This is true for just about everything. A comparable example is the increased university tuition costs as student lending took off. The market adjusts to the increase in money.
So while all the other issues are relevant, I think the key point is that these homes, due to their quality (perception or otherwise) are at the lower end of the market. If you have money flooding into the market for stick built homes, and the terms are so generous that the loans payments are no worse than previous payments on manufactured homes, what is the consumer going to buy? Particularly in a market hyped up with home-flipping shows, and real-estate never declines memes.
Lending generally increases spending. If it's for a supply limited asset (best example being land, or development rights), it's going to increase prices. If it's not supply limited, it's more likely to increase consumption (though you'd also expect some of that to divert to other areas of consumption). Higher consumption might not be quantitative, but qualitative, or could be a mix of both.
Your example of tuition costs could be debated in all those realms. Quantitatively, more students went to college. Supply oriented, there was a limit in trained professors (though some could be lured from other professions via higher salaries). Qualitative is the weakest as there's little evidence of the value of those educations having increased.
For homes, you could break quantitative into number of homes and square footage. Both have increased, though square footage much more. Number of households mostly via fewer multi-generational households.
Neither of those however seem to run against manufactured homes. The qualitative aspect has some merit, given what is available in the marketplace. In a sense, manufactured homes could be higher quality than they are, given demand for those homes. In the fringe the cost case for them breaks down if quality is considered "customized", but I have a hard time believing there wasn't a huge number of homes that the quality impressions of buyers would be willing to give up some customization for the cost benefits.
The most compelling idea that occurs is that if the market was continually growing up (quality or square footage) more than out (number of units), there would be a strong effect of the new units needing to meet the "up" requirements, and the "out" at other price points being met by filtering units down. For example, simplifying and looking only at square footage, is in 1970 the 20/40/60/80 percentiles for sq. ft. were 600/800/1000/1200, and in 2020 it was 900/1200/1800/2400, most new homes would be 1800+ sq ft, and the bottom 40% would be filled via older units. A bit oversimplified, as there was lots of remodeling/expansions, etc., but I think the thought is still useful.
So with that, most of the new units would be built to meet that part of the market which was most distant from the part of the market manufactured homes were both best suited to meet, and most prepared. While I'd say there was a part of this market that logically it seems like manufactured homes should have been well suited to.. the gap between current business practices and what would have needed to change may have simply been too great for manufactured home suppliers to gear up to, given the need to compete with stick homes, the overall uncertainty of those markets, and a number of policies that favored stick home construction over manufactured homes.
The construction industry is a famously inelastic supplier. That is one of the main reasons that pricing swings so greatly with changes in demand. I believe the main driver of supply inelasticity is labor availability. Land is also obviously inelastic. The other reason for the pricing swings would be that various forms of leverage (straight up construction loans, but also accounting practices like the old SOP81-1.
Lending driving up college tuition costs is not a particularly odd concept. At the point were Forbes is willing to discuss it, it's not even particularly controversial:
I've enjoyed this series, and in general when you cover niches that most of us normally wouldn't think about.
I'd be curious if it's possible to compare the all-in *monthly* cost of a site built home vs a similar manufactured home in the 1970s, vs. today. I'd guess the decline in financing costs have narrowed the total cost gap between the two types over the last 50 years, enough to offset shifts in manufacturing costs, especially if there was a lending bubble in mobile homes back then.
Some of it might be the Alchian-Allen effect, in this case higher land costs narrowing the cost gap. I used to live in San Luis Obispo County in CA, a rural place where there are still a ton of mobile home parks from that era. I imagine if land use laws were different, those would all be homes and condos now, just like the rest of the region, because that land is much more valuable now. Greater urbanization crimps the economics of mobile homes, I imagine.
There's a natural decline in consumption of low end goods as society gets wealthier. If incomes are growing faster than construction costs (which I think they mostly have been) you expect consumption of the low end options to decline. People today drink more craft beer and less Busch for example, compared to the 1970s. The durable nature of homes means that production will *really* decline when consumption stagnates, because there will be enough existing supply to satisfy most future demand for a long time.
Thank you Brian for this excellent series. You present solid evidence that debunks the conspiracy theory about sinister forces at NAHB and HUD colluding to wipe out the mobile home industry. Because of the capital intensive nature of the construction and the different lending standards it's pretty reasonable to expect a boom/bust cycle in this particular shelter industry. From this perspective, the production peaks in the 70's and 90's demonstrate a glut headed to a crash rather than a displacement of site built homes.
The Boston Globe had a good article today in the Sunday real estate section about how mobile home parks are enjoying a modest rebound in eastern Massachusetts. Retirees can take advantage of low carrying costs, the community of the park, and a planned home. There will always be a niche market, but I'm guessing it won't exceed 15% of housing starts on average.
HUD’s explanation for rejecting the ANSI standard was that they were reluctant to hand over code writing to a private group? This is odd because the building code for site built housing has been privatised for decades. It currently is produced by the International Codes Council ; previously regional codes were produced by three regional organisations.
I could see it being a bit more complex. Perhaps they did not want to hand over code writing to THAT particular private group. They might have looked at the code proposed, compared it to site-built housing and been unimpressed by the quality of the private group's code. (the fire fatality rates speak to that)
Anybody contemplating living in a manufactured, mobile, trailerwhatchamacallit should look at the aftermath pictures of Mexico Beach FL after being hit with a hurricane in 2018. Many died with their houses both wood frame and manufactured. They and the rubble were washed out to sea in the storm surge. Or check tornadoes, wildfires, etc. - they offer zero protection - code or no code.
Worth noting, is that most people living in these tincan shelters are renters; not owners. After the owners (usually retirees living off SSA pensions) died, families sold them to the trailerpark which then rented them out - so many are very very old.
I share your current working theory in part. The chattel collapse of 2000 remains partly responsible for tepid MH unit sales. Following 2000, the sub-prime run-up to 2008 drew many traditional MH buyers into unsustainable (it turned out) single family homes. Then the crash, credit tightening and resulting regulatory response to 2008 impacted MH chattel lending and home sales. But, what explains the last 5 - 7 years as housing demand really heated up?
From here, I go to zoning and planning approvals first. The number of MH communities being built roughly equals the number being closed down. Second, as someone who talks with a lot of would-be buyers, the MH community industry has a water cooler problem - i.e. what people say when you announce, "we are looking at a mobile home."
Dave Ramsey calls MH a depreciable asset and asks "why would you do that?" MH community closures put people out on the street - e.g. in Indiana with just 60 days notice - and subject people to what the Washington Post wrote were "massive rent increases." It all contributes to reluctant consumers and that narrows demand.
My view: The industry is in need of structural change. Long-term leases and access to residential mortgage loans (i.e. home only mortgages on leasehold land) would be one why to say and mean that these are permanent homes and worthy of residential mortgage loans. Think about MHCs as "true lease" communities and present that paradigm to town planners and the buying public. Our work is in helping homeowners form co-ops to buy their communities. That's the surest way to ensure that homeowners have long-term security and affordability. www.ROCUSA.org
I’m a little dubious about that 50,000 MH community number. I suspect the figure comes from a Homeland Security database of communities, which includes a lot of parks that no longer exist and many are under 25 pads.
The financing is a huge component of stagnant production numbers, and the personal property vs real property makes capital for the financing more expensive. The public policy around MH, making them difficult to qualify for rent support, exclusionary zoning, lower tax revenue (because of the personal v real property issue) conspire to make them a less attractive housing option for public support. Finally, outside of retirement communities, they’re invisible to most people and therefore no one recognizes the affordable component of these communities.
There’s also the issue of capital for park owners and Fannie/Freddie debt availability, but that’s a much longer comment.
It's possible to accept that those particular upswings and downswings were caused by loosening of lending standards, and a predictable collapse from poor standards.
That said, it doesn't answer the question of why manufactured homes, given their significant cost advantage aren't more prevalent, and haven't become more and more prevalent as the gap between the stick built homes and manufactured homes has become wider?
In addition, a part of the explanation for the the impacts of lending bubble/crash is incomplete without looking at how repossession law differs between stick built and the majority of manufactured homes. If a manufactured home remains classified as personal property, repossession is less onerous than if classified as real property.
https://www.nolo.com/legal-encyclopedia/manufactured-home-foreclosures-repossessions.html
Public impression also seems likely to play a role. That perception isn't helped by the chassis requirement. Not only does it make manufactured homes even more obviously distinct, it doesn't make sense in a way that would cause any prospective buyer to question the entire premise and wonder what other non-sense this class of home would get them involved in. It's interesting that it can almost entirely cancel the value of the other HUD requirements, in that the perception of the post 1974 home is more closely linked to the performance of the pre-1974 home. Those improvements are fairly invisible in comparison to the chassis in the basement.
I think what need to be added to the discussion is how easy (at times) it has been to finance homes, and at historically low rates.
There has been a general rise in the size/pricing of stick built homes. It has been shown that lending is inversely related to pricing. This is true for just about everything. A comparable example is the increased university tuition costs as student lending took off. The market adjusts to the increase in money.
So while all the other issues are relevant, I think the key point is that these homes, due to their quality (perception or otherwise) are at the lower end of the market. If you have money flooding into the market for stick built homes, and the terms are so generous that the loans payments are no worse than previous payments on manufactured homes, what is the consumer going to buy? Particularly in a market hyped up with home-flipping shows, and real-estate never declines memes.
Lending generally increases spending. If it's for a supply limited asset (best example being land, or development rights), it's going to increase prices. If it's not supply limited, it's more likely to increase consumption (though you'd also expect some of that to divert to other areas of consumption). Higher consumption might not be quantitative, but qualitative, or could be a mix of both.
Your example of tuition costs could be debated in all those realms. Quantitatively, more students went to college. Supply oriented, there was a limit in trained professors (though some could be lured from other professions via higher salaries). Qualitative is the weakest as there's little evidence of the value of those educations having increased.
For homes, you could break quantitative into number of homes and square footage. Both have increased, though square footage much more. Number of households mostly via fewer multi-generational households.
Neither of those however seem to run against manufactured homes. The qualitative aspect has some merit, given what is available in the marketplace. In a sense, manufactured homes could be higher quality than they are, given demand for those homes. In the fringe the cost case for them breaks down if quality is considered "customized", but I have a hard time believing there wasn't a huge number of homes that the quality impressions of buyers would be willing to give up some customization for the cost benefits.
The most compelling idea that occurs is that if the market was continually growing up (quality or square footage) more than out (number of units), there would be a strong effect of the new units needing to meet the "up" requirements, and the "out" at other price points being met by filtering units down. For example, simplifying and looking only at square footage, is in 1970 the 20/40/60/80 percentiles for sq. ft. were 600/800/1000/1200, and in 2020 it was 900/1200/1800/2400, most new homes would be 1800+ sq ft, and the bottom 40% would be filled via older units. A bit oversimplified, as there was lots of remodeling/expansions, etc., but I think the thought is still useful.
So with that, most of the new units would be built to meet that part of the market which was most distant from the part of the market manufactured homes were both best suited to meet, and most prepared. While I'd say there was a part of this market that logically it seems like manufactured homes should have been well suited to.. the gap between current business practices and what would have needed to change may have simply been too great for manufactured home suppliers to gear up to, given the need to compete with stick homes, the overall uncertainty of those markets, and a number of policies that favored stick home construction over manufactured homes.
The construction industry is a famously inelastic supplier. That is one of the main reasons that pricing swings so greatly with changes in demand. I believe the main driver of supply inelasticity is labor availability. Land is also obviously inelastic. The other reason for the pricing swings would be that various forms of leverage (straight up construction loans, but also accounting practices like the old SOP81-1.
Lending driving up college tuition costs is not a particularly odd concept. At the point were Forbes is willing to discuss it, it's not even particularly controversial:
https://www.forbes.com/sites/prestoncooper2/2017/02/22/how-unlimited-student-loans-drive-up-tuition/?sh=81f757b52b63
I've enjoyed this series, and in general when you cover niches that most of us normally wouldn't think about.
I'd be curious if it's possible to compare the all-in *monthly* cost of a site built home vs a similar manufactured home in the 1970s, vs. today. I'd guess the decline in financing costs have narrowed the total cost gap between the two types over the last 50 years, enough to offset shifts in manufacturing costs, especially if there was a lending bubble in mobile homes back then.
Some of it might be the Alchian-Allen effect, in this case higher land costs narrowing the cost gap. I used to live in San Luis Obispo County in CA, a rural place where there are still a ton of mobile home parks from that era. I imagine if land use laws were different, those would all be homes and condos now, just like the rest of the region, because that land is much more valuable now. Greater urbanization crimps the economics of mobile homes, I imagine.
There's a natural decline in consumption of low end goods as society gets wealthier. If incomes are growing faster than construction costs (which I think they mostly have been) you expect consumption of the low end options to decline. People today drink more craft beer and less Busch for example, compared to the 1970s. The durable nature of homes means that production will *really* decline when consumption stagnates, because there will be enough existing supply to satisfy most future demand for a long time.
Thank you Brian for this excellent series. You present solid evidence that debunks the conspiracy theory about sinister forces at NAHB and HUD colluding to wipe out the mobile home industry. Because of the capital intensive nature of the construction and the different lending standards it's pretty reasonable to expect a boom/bust cycle in this particular shelter industry. From this perspective, the production peaks in the 70's and 90's demonstrate a glut headed to a crash rather than a displacement of site built homes.
The Boston Globe had a good article today in the Sunday real estate section about how mobile home parks are enjoying a modest rebound in eastern Massachusetts. Retirees can take advantage of low carrying costs, the community of the park, and a planned home. There will always be a niche market, but I'm guessing it won't exceed 15% of housing starts on average.
HUD’s explanation for rejecting the ANSI standard was that they were reluctant to hand over code writing to a private group? This is odd because the building code for site built housing has been privatised for decades. It currently is produced by the International Codes Council ; previously regional codes were produced by three regional organisations.
I could see it being a bit more complex. Perhaps they did not want to hand over code writing to THAT particular private group. They might have looked at the code proposed, compared it to site-built housing and been unimpressed by the quality of the private group's code. (the fire fatality rates speak to that)
A modular home manufacturer goes under in UK https://www.insidehousing.co.uk/news/legal--general-to-cease-production-of-new-homes-at-modular-factory-81335
Anybody contemplating living in a manufactured, mobile, trailerwhatchamacallit should look at the aftermath pictures of Mexico Beach FL after being hit with a hurricane in 2018. Many died with their houses both wood frame and manufactured. They and the rubble were washed out to sea in the storm surge. Or check tornadoes, wildfires, etc. - they offer zero protection - code or no code.
Worth noting, is that most people living in these tincan shelters are renters; not owners. After the owners (usually retirees living off SSA pensions) died, families sold them to the trailerpark which then rented them out - so many are very very old.