Skip to content


Post date:

We’ve been seeing a lot of microhomes lately, at least in architecture-land where they’re a regular on the competition circuit and in the design studio. Microhomes aren’t the same as the small houses that are usually set as an introduction to architectural design in first year before the projects SMLXL each year. Because a microhome’s size is typically around 25 sq.m not counting external areas whether covered or not, it’s not in the spirit of it to use a crazy amount of resources. However, a microhome designed for some idyllic countryside situation seems to me more like a weekend house or short-let accommodation and thus part of the problem it’s meant to solve. Yet, at the same time, a microhome isn’t a tiny house which, at least at the beginning, was a reduced-area house as a response to homelessness.

But what if, say, two people wanted to live in a 25 sq.m dwelling out of choice? What would make them want to do that? You can’t say “a nice location,” because that’s a function of money. The necessary spaces, furniture, fittings and appliances fit in somehow to accommodate the mechanics of daily life but what can architecture bring to the problem? Architecture generally has little to say when site, area and resources are squeezed. Over in Japan, Atelier Bow Wow have been having having a good shot at keeping architecture relevant with series of small houses. Here’s four, ranging from 22 sq.m to 37 sq.m. Small houses such as these are often seen by Western eyes as audacious or, worse, pretentious.

The ferocious inventiveness of (architect-led) Japanese house design since the 1960s is a result of Japan’s essentially feudal system of land ownership increasing the value of land so rapidly so that whatever’s built on it becomes financially irrelevant within a decade or two at the earliest, five or six max. The average is about 30 years. Transfer of land (or the rights to build on it) usually means the end of the road for whatever’s built on it – as happened to Toyo Ito’s U-House, or Shinohara’s House in Yokohama Toyo Ito’s U-House. There’s a huge difference between the real value that architecture adds, and its perceived value. We can’t do anything to change this system in Japan but we really shouldn’t be admiring houses with the lifespan of a sofa.

This isn’t a criticism of small houses, whether inventive or not. There are many forces pressuring houses to become smaller and that, when we design a small house for a competition or in the studio, we ought to be clear about what problems we want it to address solve.

  1. Will it have some worthy and specific social or humanitarian agenda?
  2. Will it be affordable?
  3. Will it be made from only locally-sourced renewable and recycled materials?
  4. Will it have a lot of technology thrown at it so we can say it is off-grid?
  5. Will it be open-source, self-build design or is fantasy mass-production or 3D printing the answer?
  6. Will it have a low environmental impact?
  7. Will it be possible to live in?
  8. Will it be architecture?

No.1 seems mandatory for competition entries but – and please don’t get me wrong here – designing micro-houses for niche and/or disadvantaged groups is undeniably virtuous if they actually get to the people they are designed for but, if they don’t (and they probably won’t), all we’ve done is reinforce the mindset that micro-houses are for other people and not for the mainstream. I’m going to approach the design of a micro-house from the other end and try to create something that’s architecture or at least has a degree of spatial interest. When this is over we can have a look at what’s been gained and what’s been ignored. If people want to say my micro house is pretentious and has ideas above its station then I’ll know I have succeeded.

I have three references. My first is Hiroyuki Asai’s 1971 Mochizuki House. It’s inclined wall making the living area more expansive and the sleeping areas more intimate has bewitched me for decades.

The second is my attempt at recalling what I liked about Mochizuki House when it was still just a memory. It’s not a “reimagining” but me running with someone else’s idea.

My third reference is Sou Fujimoto’s 2006 Final Wooden House but not because I approve of its concealed steel rods stacking, staying, and sometimes suspending some decent hunks of wood, but because this little space contains many corners in which to be. Entry is into the kitchen space with the sunken bathroom to the right. Straight ahead and up a couple of 30cm steps is some sort of sitting area, while to the right and up a couple more is one sleeping space, and then right again and up a couple more gets you to above the bathroom where all you can do is lie down. There are five different places to be and each has various views of the others. I’m not saying it’s liveable or even lived-in as I can’t see any light fittings or power outlets, but this vertical overlapping of spaces at different heights could be applied to a more conventional micro-house to give one but especially two people a choice of four or five different places to be.

The stairs would have to be habitable or useable in some sense to stop them becoming inefficiently used area and this is what happens with this exercise of Fujimoto’s. The other idea of many different corners just might make small houses more attractive. This own space doesn’t have to be isolated cells. I took my own project as the starting point. The first image shows how it already had two very different spaces, not counting the bathroom or kitchen.

My first thought was to put an additional space above the bathroom and occupy the corner currently occupied by the shaft seen in the image at the top right.

Forgetting about the kitchen for now, I thought this new space would be accessed by steps up the roof. And making another volume to balance it on the other side was no problem either. The space beneath the roof was always going to be the sleeping area and the space beneath the upper volume was always going to be the bathroom but adding these spaces now took me way over 25 m2.

There was also the problem of how to get to this new platform space. This next image shows my first thought. I imagined a low-ceiling space with low furniture such as beanbags and a coffee table. It’s no tea ceremony room but I gave it a round window anyway. These things are sacrificial.

The space beneath that platform was never going to be part of the 25 m2 so away it went. And nor was the space above the bathroom but I didn’t yet know that.

The bedroom area was small but there was an opportunity for a nice window below the roof and a paired one the other side of the house. The bedroom is accessed by the passageway to the right of the front door (in the plan below) but I’m feeling the squeeze. The space beneath the stair is now a void to reduce the area, there’s no space for the bathroom and for some reason I’m still thinking it’s okay if the kitchen goes upstairs. (!?)

I came to my senses and the kitchen came downstairs and the plan began to make some sort of sense when I moved the stair to what became an upstairs outdoor space, to the middle of the room and partially consolidating the two stairs. For a while I let myself think this was interesting.

The area was still on track but the sleeping area was just a dark corner with an inclined roof/wall that could be lower but this would 1) decrease its access headroom and 2) increase the floor area of the living room. I learned that moving an inclined wall – especially one at 45° – separating two different levels changes the height of one space and the area of the other. Strange stuff happens when plan and section mirror.

I thought it was over when I made the outside terrace larger so I could have a single stair that wasn’t such a feature in the middle of the house. Also, the illusion of making a small house feel larger by getting good aesthetic value from the space is what this exercise is about so I lowered the roof above the living room to make the the living room roof vanish into “infinity” instead of just crashing into the end wall. Not bad. It’s no Laurentian Library but the door at the top of the stairs is underscaled. “Cheers Michaelangelo! This device of yours makes a small space seem larger than it is.”

Still, I worried that the inclined wall/roof I so loved at the beginning had little presence apart from above the bed. I thought it might be a good idea to reduce the size of the terrace and have the roof continue on the other side of the stair. It wasn’t. The two parts couldn’t both vanish to infinity if the horizontal roofs were at different heights. [left, below]. Doing the opposite was equally pointless [right, below] although it’d be the thing to do if the terrace continued up.

Micro-houses have limitations on area and, although nothing is said about excess volume, I began to be suspicious of this almost-double-height space above the kitchen and table. It wasn’t in the spirit of a micro-house so the idea of removing excessive volume was now driving the design. The one thing I hadn’t yet thought of was best.


So was it all worth it? What’s been lost and what’s been gained? Total area is 25.33 sq.m which is slightly over but can be reduced easily enough. I was curious to see if there was any area or volumetric advantage over a single-level house and produced a reference design using the same layout. Its volume turned out about 20 m3 more despite my simplifying the roof but it was a false comparison as there’s no reason to have a ceiling as high as the room is wide. It’s a different house and, if we don’t count the bathroom or outside space, has only two distinct spaces in which to be.

The high ceiling is only there because of the upstairs outdoor area and if we take that away and reduce the height, and reposition the bedroom door to give more wall to the living area then, in two simple moves, we’re left with a very ordinary layout that, though better than many, is now determined by cost. Whatever the construction is, I can now cost the difference between my reference design below, and my final design, and see if it the value I think I am adding is at least proportional to its cost.

The lightweight stair was a good idea so I appropriated that. Since it’s a matter of constructing this house as inexpensively as possible, the higher of the two roofs will have to go. Confession: I avoided showing it in a section but there was some complicated construction where the outside terrace was longer than the bathroom below. I belatedly fixed this by making the bathroom longer, and then reduced the area of the bathroom by making its walls thinner.

This helped the elevations but having to make more space for the bed meant making both the bedroom and living area wider. These two good things meant fine-tuning the dimensions so the area was bang on 25.000 sq.m. The void beneath the living area will probably be filled with rubble. Or perhaps not. I’ll stop it here.


  • I really dread these kind of multi-storey small solutions. Past 80 y.o. one basically needs to move out. Most of the times it’s not possible to live only on the ground level.

  • The problem with these things – especially if they’re meant to be an answer to homelessness or a shortage of housing – is how they are meant to work in an urban context with significant density. Most studies or photos I’ve seen have the thing on a plot that could take a London terraced house or three flats, and if they really are to be deployed one by one we’re bailing out the boat with a teaspoon. (Consider the Gas House above – drop the funny requirement that a tinyhouse is a single family building and the offstreet parking and you could get more out of that.)

    I don’t think I’ve ever seen a study (still less a completed project) for a street of tinyhouses or a tinyhouse estate, and there’s a good reason for that! If you were in a position to build a street or a subdivision’s worth, it would be idiotic to do that when you could instead build substantially more apartments, townhouses, or some other typology.

    They might, as you say, make sense as a solution for a holiday let (or an architect’s personal getaway…). More charitably perhaps they could be rural public housing, added in small numbers to an existing village site.

    Although you really really wouldn’t want to be in one with your children on a rainy day would you.

    • says:

      I totally agree. I’m okay with the isolated micro-house as a design exercise – especially a teaching one, but if we’re really going to get serious about smaller houses as a form of mainstream housing, then we really need to see what happens when they’re repeated, stacked or otherwise aggregated. We also have to shift away from the mindset that they’re a solution to homelessness or urban housing. They’re already the solution to suburban housing but we’re not there yet although, as you say, one person or two persons max. and preferably adults.

      • Suburban? The basic bargain of suburbs is that you put up with commuting to everything in exchange for more space, so I’m not sure what the point of “you still have to cram your life into a poky flat, but now you have to commute as well and give up the possibilities of the city!” is. If there is demand from single people in the ‘burbs, wouldn’t it be better to concentrate on building a meaningful town centre around mass transit? Then at least the commute is more bearable and there’s something to do locally.

        I think this is a solution in search of a problem. They would make lovely holiday lets or whatever, and that’s not a bad thing! Holidays are good! But they’re no kind of housing solution. People put them forward for things like emergency camps for refugees, but this is silly – imagine trying to fit families across two or three of the things.

    • says:

      I’m currently going back through the last three years of ArchDaily to see what I’ve missed, and I came across this little house which I think does all the right things, and for a small urban site as well. I’m very impressed. A tsubo is a traditional Japanese unit of area measurement but, in this case, I think it refers to the footprint. 6 stub = about 21 sq.m. I like how the mezzanine and bathroom has been used to stretch the house upwards. You’d still need to have some Japanese children if you were to share it.

      Sorry – I just saw your latest reply. You’re right in that it’s a solution in search of a problem and that they probably have-not found the right problem yet. In Australian cities at least, where one house stood 40 years ago there are now six or eight, still pretending to be detached houses. The fundamental contradiction might be the high proportion of surface wall compared to the amount of volume contained. In other words, it is the detached house typology that’s the problem and not the size.

      In my recent ArchDaily binge I did see a village of tiny houses but these were shelters for the homeless in LA.

      • That is rather nice. I think what’s doing the work is building upwards – it’s essentially a skinny townhouse or a diminutive sliver tower. You’d need a very strong wingman to keep the inevitable expectant little queue out of the kitchen while you were cooking in it.