Hacker News Reader

Comment by jeffbee | Hacker News Reader

Prefab is the wrong optimization. It targets a minor improvement in a small part of the overall budget. Most of the time and money of construction in America is tied up in land prices, permitting, and regulation. Prefab solves zero of these issues.

There is a building going up in Oakland right now where Factory_OS put out a press release that they had built it in ten days. But they spent 7 years getting approvals, 18 months in site prep, and they've been detailing and finishing the building for more than a year since the press release, so it doesn't seem like the method of construction was a very important factor.

View on HN

The advantage to me seems like it should be in quality and consistency, not cost and speed. Building houses is fast enough, and as you say the cost of the land often dominates, but the quality of houses is appallingly bad compared to consumer items mass produced in factories.

This. I want homes as well designed, engineered and built as iPhones, or at least automobiles.

FWIW, I have a hunch that in 10 years TSLA will start building homes and AAPL will likely start building condo buildings.

The quality issues that you encounter in houses aren't the framing either, though. It's the roof sealing, ventilation, trim, plumbing, ... all things that prefab framing also doesn't address.

True, I don't see a point to prefab unless it involves a lot more than the framing. But it seems totally possible to do prefab installation of most of the ventilation, plumbing, electrical, etc.

Does prefab go much beyond framing? The NAHB estimates that framing is only 17% of the cost of a house(1), so even cutting that cost in half doesn't save much on the whole project.

(1)https://www.nahbclassic.org/generic.aspx?sectionID=734&gener....

> so even cutting that cost in half doesn't save much on the whole project.

Houses are usually a leveraged investment. If you put down 5%, and prefab saves you 5%, you have massively reduced your risk of default during a downturn.

And if you expect it to take 20 years to pay off your mortgage, and framing saves 5%, that’s one year saved, which is massive IMHO.

Obviously a lot of assumptions to the above e.g. everyone else doesn’t get the 5% savings, and everyone doesn’t just bid up new house prices by 5% (because they are bounded by their mortgage which is related to their earnings, not costs).

These are fair points, and if a buyer had 100% confidence in prefab, your argument certainly makes financial sense. But for most consumers prefab is a novel approach, and therefore likely to be perceived as risky. So that savings might need to be greater to get them to try something new. Given the failure of prefab to make a big dent in the market, that seems to be exactly what's happening.

I am currently in the process of buying prefab house and after visiting couple of local factories it seems like most of them is doing just framing and competing on price only.

We are going with company that actually assemble full house inside and bring it to site fully equipped: electrical, plumbing etc.

What was the breakdown for prefab direct cost versus transportation cost (if you don't mind me being nosey)?

Transportation should be around 5-8%.

Huh? There's a shortage of carpenters so if you can go prefab it is a win. In fact the article shows that if you can build a house faster with prefab at the same cost as a traditional house that's a win.

There isn't a shortage of carpenters. There is a shortage willing to work for the wages offered. I've done carpentry work in the past, offer enough money and I'll leave my tech job. I'm not the only one who could do carpentry for the right price.

Any skilled job that can be done poorly with little training becomes a bit of a lemon market.

I hate painting. I'm also a terrible painter. However, I've given up on ever hiring painters. Every professional painter I've used has either prepped the surface improperly, or put the coats on too thick, to the point where I rub paint off if I brush up against the wall.

I'm willing to pay more for better work, but I can't easily tell who will do a better job.

> or put the coats on too thick, to the point where I rub paint off if I brush up against the wall

How many coats are they putting on the wall, 5? I've never heard of paint rubbing off because there's too much of it.

Same. This is much more likely to be a problem of poor surface prep and cheap paint than of the paint being too thick.

It is not number of coats it is thickness of coats. If you put on a thin coat of good paint it sticks well. Thick coats of even good paint won't stick.

Same here. I'm a skilled carpenter and woodworker, but for a much higher salary and benefits I can sit on my butt all day long typing into a computer. They would have to offer me a lot more money than my current six figure salary to get me to do carpentry work all day.

Around here, the carpenters I know all drive very nice trucks and are booked solid making $$$$. These guys are running their own 2-4 person shops though.

In my area, yacht racing is what carpenters do on the weekends. No joke. Brand new sails, custom everything, and, of course, when the wood trim needs fixing or varnishing, it looks gorgeous.

Me too. I worked for 5 years as a carpenter and at the end of the day it just wasn't worth the paycheck and the back pain.

We are starting to talk about externalized costs with regards to the environment, but this concept applies everywhere. We have a problem in this society paying the true cost of damn near everything and that includes physical labor.

Ok, so prefab can still be a win if it's cheaper?

Problem is it isn't much cheaper. There is still work to do on site. Prefab needs labor in the factory. Prefab generally has higher shipping costs. These mostly balance out.

I personally think prefab is a win if it results in a faster construction time.

this is especially relevant in places with short build seasons or bad weather.

It's about scale, as the person you are responding to clearly said.

Technically, it might be a win. It's not enough of a win, however, to bring many new converts. It simply doesn't scale to enough large savings to make much of a difference in the construction budgets OP actually was writing about.

> Most of the time and money of construction in America is tied up in land prices, permitting, and regulation. Prefab solves zero of these issues.

I'd hypothesize that access to affordable, reproducible, medium-scale home production would allow a lot of new places to try to develop. Land prices, permitting, & regulation are very different in different places. The places where these are not the major factors could have a better chance, if home construction could be more economical & effective & appealing, which pre-fabs might be able to help with.

You say it’s the wrong optimization but your argument seems to be that there are bigger problems. I tend to agree, particularly with pernicious stuff like single-unit zoning and such, but that doesn’t mean that prefab is “wrong” overall. I don’t see it as a cost optimization, I see it as a speed and consistency optimization. Do we know if long-term maintenance costs of prefab construction are lower?

Seems like it would be hard to reason about. Consistency is great but it means if you order assembled wall panels from a factory, your foundation has to be exact in dimensions and dead level, or it's not going to come together. On-site framers aren't really bothered by this. I think it trades one kind of complexity for another.

This is really key. The quality of the end product will always be determined by the crew that sets the house. Same will be true for custom stick-built.

The biggest challenge we faced, even with a skilled crew, was simply getting the pieces into some of these lots in the ADK. Between the tight winding roads and lot elevation, it was quite a battle.

Source: My family has been in the modular and manufactured housing business since the 60s.

Well, you’ve certainly proved that I don’t understand this well enough to evaluate alternatives, fascinating though I do find this subject :) I guess the dimensions seem like they’d be easier to get right than leveling, though, maybe?

I'm sure a shim kit would be included to even out discrepancies. If the tolerance is too high for that then your foundation needs to be fixed.

Having done framing in the past I assure you foundation tolerances are not close enough for a shim kit in general.

Seems like people are getting away with shoddy work then.

It kinda seems cheap for people on a forum for software developers to criticize the quality of, well, just about anything.

It costs more than it is worth to have them do better. As a framer is takes me a couple minutes more to compensate. To have the do it would take hours of labor.

I presume you're talking about scribing the bottom plate, or perhaps another board placed between the foundation and the bottom plate? Why couldn't a layer of that same sort of scribed board be placed on the foundation before placing the prefab modules?

Depends on where it was. The worst case I saw was a half wall, so we cut each stud a different length.

There are options for prefab as you saw - but when building on site we have the saws and other tools needed anyway, so it isn't a big deal to use them. With prefab they don't normally use those tools so you have to account for the cost of getting them out putting them away.

"What do you mean you didn't finish all the tickets in this sprint? Must be shoddy work."

You're looking at single pieces of a complex and already optimized-for-several-factors system and assuming it would be trivial to do it better for one factor without taking others into account...

That depends heavily on where you're building. There are plenty of places you can pick up a few acres for $10-$20k (with electric/water access) and have a minimal permitting process that's over in at most on the order of weeks.

The San Francisco Bay area is not all of "America".

> There is a building going up in Oakland right now where Factory_OS put out a press release that they had built it in ten days. But they spent 7 years getting approvals, 18 months in site prep,

I'm currently gut remodeling a small house in Oakland. Permitting and regulations are a whopping 3% of the cost.

The land (the existing property) is 50%, and 40% is building costs.

Land costs (addressed via incremental upzoning) and build costs (addressed with more efficient building technology) are the 2 biggest things to address to make housing affordable, not the cost of permits.

Remodeling an existing structure won't really expose you to the regulatory burden, which is good for you but isn't the full picture. If you wanted to buy a SFH and build a duplex or, God forbid, a triplex, then you'd be paying serious money, mostly in the carrying costs of your debts while the government dithers over your application. This is why in California one of the most important recent state law changes was 2017's SB35, which mandates that local authorities must approve certain types of projects in 90 days or less. But the types of projects which benefit from this are fairly few.

On top of the carrying costs there are impact fees. These vary by city but you can see how crazy they are. In Fremont you will pay $160k per dwelling, that's just cash out of your pocket. Totally crazy.

https://www.spur.org/news/2018-05-09/it-all-adds-growing-cos...

I don't fully agree given that housing and materials prices rise every year. 7 year delay adds how much $ to a project in inflation?

Well delays can vary. My delay was 3 months, but that was due to COVID.

Permits and regulations should be solved too. Prefab houses can be identical or use a finite set of components that could be pre-approved easily.

I know of at least one prefab option where permitting and inspections are streamlined at the factory.