Henry Ford famously said that customers could have the Model T in any color they wanted, as long as it was black. This restriction was largely driven by the limitations of automotive paint technology at the time. The only cheap, durable finish available for cars was a baked black enamel. Higher end cars such as Cadillacs were available in different colors, but painting them was a long and tedious process. The colored paints available at the time could only be applied in thin layers with small amounts of pigment, and painting required multiple cycles of applying a coat and waiting for it to dry. Altogether it took 2-4 weeks to paint a car, and huge volumes of work-in-progress (WIP) accumulated. A plant producing 1000 cars a day would need 20 acres of indoor storage for cars in the process of being painted. Additionally, the colored paints then available weren’t durable, and would start to chip and flake off after a short time.
The product that changed this was Duco, a quick drying automotive lacquer developed by DuPont in the early 1920s. Duco was discovered by accident while researchers were trying to stop static electricity from causing streaks on nitrocellulose movie film. In the course of their experiments, they made a batch of nitrocellulose solution with added sodium acetate, but a power failure in the plant resulted in it sitting idle for several days. When the barrel was opened, the solution had become much thinner, and the researchers realized that the thin but high-nitrocellulose-content solution might be useful as a lacquer. Duco cut the labor required to finish a car body by 15%, and eliminated a third of the painting steps along with three 24-hour periods of drying. A car could now be finished in a single 8 hour shift. As a result, colors became widely available on cars - the Model A, which followed the Model T, was available in 78 different colors.
Plaster
It’s interesting to compare the development of Duco with a similar development in building construction: the finishing of interior walls. In the early 20th century, the interior walls of buildings were typically finished with plaster, a building technology that had been in use in one form or another for thousands of years. Plaster consisted of a mineral such as lime or gypsum, burned in a kiln to form a dry powder. When mixed with water, the plaster would re-hydrate, forming a solid surface that could be painted, have wallpaper applied, etc.
Plaster has attractive qualities for an interior finish - it produces smooth surfaces, it’s fire-resistant, it deadens sound, it's strong and durable (though susceptible to cracking), it's relatively easy to repair, and it can be molded into different shapes. But it has some drawbacks. For one, plaster is heavy, and requires a secondary framing as backing. This was traditionally done using lath, a series of wood slats placed around ½” apart. Plaster would be pushed through the gaps, which would act as keyways, holding the plaster in place (later things like wallboard and steel mesh would be used as backing material).
Plaster also took several weeks to finish, as it was applied in multiple coats which each needed to dry. The moisture released from the drying plaster also often caused problems for parts of the building sensitive to humidity, such as freshly-placed wood.
The invention of drywall
The system that replaced plaster is colloquially known as drywall (also called sheetrock, plasterboard, or wallboard). Drywall consists of a sheet of gypsum sandwiched between two layers of paper. The sheets get attached directly to the studs using nails or screws, and the edges are taped and smoothed out with a layer of joint compound (usually a mix of gypsum dust and water). The result is a smooth surface similar to plaster (it’s fire resistant and comparatively easy to paint or apply wallpaper to) that is simpler and faster to install.
Drywall was first developed in the late 1800s as a product called Sackett Board. Patented in 1894, Sackett Board consisted of several layers of gypsum plaster separated by wool felt. Sackett Board was sold as a replacement for the lath backing, which allowed the elimination of the first, heavy coating of plaster (called the “scratch”).
The Sackett Board company was later bought by US Gypsum, and by the early 1900s a variety of companies offered gypsum board products. WWI accelerated the use of gypsum board as a replacement both for the lath backing (as “plasterboard”) and for the plaster itself (as “wallboard” or “sheetrock”), as it allowed war housing to be built more quickly. The government in fact took over the entire gypsum industry to secure enough wallboard, as twice the industry capacity was needed. By the 1920s, gypsum board products were widely advertised as full plaster replacements.
Plastering, however, remained common. Gypsum board used as a lath replacement was outselling wallboard 3-to-1 by 1939, and the number of plasterers in the country would not peak until 1950 (at 64,000). This was at least partly due to building code requirements (which often specifically required plaster walls) and advocacy by plasterer’s unions (which opposed the use of an alternative wall finish that didn’t require a plasterer to install).
Wallboard would once again widely substitute for plaster during WWII in defense housing construction, due to the shortage of materials and the need to build quickly. And after the war, the pent-up demand for housing placed a large premium on being able to build quickly, resulting in builders frequently turning to it - Levitt, for instance, used gypsum wallboard in their large scale housing developments. Drywall would continue to grow in popularity, and by the 1980s, more than 90% of residential construction was done using drywall instead of plaster.
Drywall and prefabrication
Duco reduced the time required to paint a car in color from weeks down to hours, solving the problem of an extremely time-consuming finishing process (though automotive paint would continue to improve). But drywall can at best be regarded as a partial solution for the time-consuming finishing process for a building’s interior walls. While an improvement over the previous plastering process, installing drywall is still a time-consuming, multi-day process. Applying the joint compound (which covers edges and imperfections in the panel) is typically done over several days, as multiple layers of compound are applied and dry. And the higher quality finish desired, the more coats are required, and the more time it takes.
With conventional, site-built construction, this delay’s impact is relatively minor. Since construction isn’t being done in a factory, there’s no extra storage space required, and while the extra time means incurring the costs of extra inventory (every additional day construction takes means another day before the builder gets paid), it’s a relatively small fraction of the 7 months it typically takes to build a home in the US.
But in a factory-built construction environment, the typical negative impacts of a very slow process (longer cycle times, large accumulation of WIP that needs to be stored) appear. In addition to these issues, drywall has a tendency to crack during transportation. Because of these issues, the drywall process is considered the biggest problem station by modular builders, by a wide margin.
Drywall alternatives
Of course, drywall isn’t the only possible interior wall material available - a variety of products in this space compete with it. A few notable ones:
Vinyl-over-gypsum (VOG) - This consists of a layer of vinyl over a sheet of drywall. These sheets are attached to the studs using adhesive (instead of the screws that drywall uses), and don’t require any finishing compound - instead, the edges are covered with battens.
Fiber Reinforced Plastic (FRP) - This is a thin layer of plastic that gets applied over an existing surface, such as a block wall or plywood/OSB (in some cases the panels come attached to plywood already). FRP is largely used for facility construction, such as healthcare, processing facilities, and commercial kitchens.
PVC Panels - These are panels made of PVC that consist of two layers of plastic connected with a truss, eliminating the backing material that FRP requires. These seem to be marketed mostly as an alternative to FRP.
MDF Panels - Panels made of medium density fiberboard, often with a graphic (such as wood grain) placed on the visible surface. You also see this made to give the look of beadboard. Related products include hardboard panels, used for the fake wood panels of the 1970s, and Fibo, an engineered wood panel largely used for bathrooms.
Dekton - Dekton is a manufactured stone material, more commonly used for countertops but also available for wall panels.
Corian - An acrylic invented by DuPont in the 1970s, Corian is also largely used for countertops, but its manufacturers also make wall panels. I believe at least some prefab companies use this instead of drywall.
Metal panels - not especially common for interiors, but were notably used on the Lustron Home.
These systems all avoid the time-consuming process of applying multiple coats of finishing compound and waiting for it to dry. Nevertheless, drywall remains dominant in residential construction, even in prefabricated environments. As best I can tell, this remains true outside the US as well - both Sweden and Japan seem to use drywall for prefabricated construction, for instance.
The most straightforward reason behind this is likely cost. Despite its drawbacks, installing drywall is incredibly inexpensive. My 2022 Construction Estimator gives a cost of installing + finishing drywall at about $1.50 per square foot. RSMeans gives similar. Most of these alternative materials are much more expensive, even before taking into account the labor or extra backing materials they might require. MDF panels seem to be ~$1.50 for the material alone. PVC panels seem to be in the realm of $3.50 per square foot for just material. Fibo is closer to $10 per square foot, Corian around $50 per square foot, and Dekton up to $100 (though this is for the countertops, walls might be cheaper).[0]
A building uses a large volume of material: a 1000 square foot home will require perhaps 3000-4000 square feet of drywall for the walls + ceiling. Since drywall is already one of the more expensive trades (slightly less than ~5% the cost of construction for a single family home), more expensive material would have a substantial cost impact. And because prefab factories operate at comparatively low volume (on the order of 10 modules a day, rather than 1000), there’s less to be gained from the WIP reduction you’d achieve from accelerating the drywall process, compared the potential savings in car manufacturing. In some cases, drywall doesn’t get finished in the factory at all. If there’s going to be joints that have to be finished on-site anyway when you stitch your modules together, you may as well just wait to finish the drywall on-site.
The materials that are cheaper, on the other hand (such as VOG), tend to feel cheap. People don’t like the look of battens (perhaps because they’re associated with mobile homes) or 70s wood paneling, and it's thus difficult to replace drywall with cheaper alternatives. Manufactured home builders in fact will go out of their way to advertise their use of drywall.
So finding a better drywall is something of a challenge. Nevertheless, I still find it somewhat surprising that this situation persists. Though the need for a faster method of installation isn’t as pressing as it was with automotive paint, it seems like an interior wall system (whether it was a drywall compound that could be applied in a single pass, or a new material) with a one-day finish time would have substantial demand (and in fact this is what Canvas is developing, though they are approaching it by way of a robot, rather than improved materials).
[0] - Using other materials is also likely complicated by requirements to have fire-rated walls in certain conditions, depending on the type of construction.
Copied from my comment on a discussion elsewhere:
This is a fun one for me, as I've worked in an auto manufacturing paint shop and I've done drywall.
They have something else common that the author didn't mention - imperfections.
I worked on custom homes with my dad growing up, and some customers demanded "triple perfect" work. Any wall can look smooth with lights pointed straight at it, or with diffuse lighting. Most walls look really bumpy when you shine a light down the wall. During construction, you are working with natural light. The customer sees the home after overhead lights are installed and after blinds and shutters are put in. This completely changes which imperfections are visible. Small scratches become pronounced; changes in texture show up. Customers can see ALL of your work. Who knows if the electrician has some half-tight wire nuts? But we all know when the drywall is bad.
Another thing drywall and automotive paint have in common: Repairs have a 50/50 chance of making things better. Automotive paint repair in this case being during the OEM process, not a collision repair. Collision repairs have a 0% chance of looking exactly like OEM paint.
Small imperfections may be fixed or covered up, but you have to skip back several steps in the process.
For drywall, if you are fixing scratches, you just have to disrupt part of the texture and paint. If you are fixing a lumpy butt joint (seam), then you have to disrupt a large area of texture and paint. The repaired texture may not match - air flow in the texture gun, humidity, exact proportions of the texture mix, temperature - will all be different. Paint has a different pore structure and dries differently on texture vs paint. This often affects the specular look of the paint.
For automotive paint, much of the same applies. Humidity, paint mix, metallic or mica mix, temperature all play a role. As well, for a repair, you are not painting primer, you are painting paint. It's very hard to match edges, so often whole body panels or even the whole vehicle is repainted. For the paint process I'm familiar with, you only get 2 tries to fix any large problems - the vehicle has to go through the oven again for each repair, and too many passes puts you out of compliance for the paint and adhesives.
====
Anyway, how to replace drywall? I don't see it happening on a large scale. I think drywall can definitely be improved, but even for most of the replacements listed drywall will be installed under the product listed. For instance, MDF does not meet fire code. Anything that uses adhesives instead of screws is MUCH harder to repair.
Some things to ponder for replacements:
- Fire code
- Speed of install
- Look and feel
- Ease of repair ( you may not like drywall repair costs, but they're nothing compared to other products, even though I said it's very hard to make repairs look good, its much harder to make repairs in other products look good )
- Ease of demolition
- Toxicity/outgassing
I can imagine some improvements to drywall, but it will still have to be cut and custom fit on site. Something is still going to have to join the seams, even if it is something that sets in 15 minutes. It isn't just that people don't like seams. It's that people don't like seams that are arbitrarily placed as artifacts of construction.