I.

There are a lot of really big houses (6+ bedrooms) on the market for around a million dollars, or sometimes less. Like this three-story, seven-bedroom house just outside Albany, NY, which recently sold for $1,181,300. Or this ten-bedroom house just south of Boston, MA, which recently sold for $990,000. Or this eight-bedroom house in Cincinnati, OH, which recently sold for $455,000. Some of these places are old bed & breakfasts; some are intended as rental properties; some of them are just big. Some of them are FRICKIN MISSILE SILOS.

Big investments generate quite a lot of money — you can draw off about 4% of an investment every year without depleting the principal, because you get back that much or more in interest. Even if you did nothing but stick the money in an S&P 500 index fund, the historical average is about 10% per year. That’s not guaranteed, but it’s pretty damn good.

If we assume 4% annually, a $3 million endowment would generate $120,000 a year, or $10,000 a month indefinitely. A $2.5 million endowment would generate $100,000 a year, or $8333.33 a month. Even a measly $2 million endowment would generate $80,000 a year, or $6,666.66 a month. 

Any of these amounts would be enough to purchase a big 6-to-10 bedroom house in many areas, with some endowment left to generate interest each month. If you sink $1 million of a $3 million endowment into a house, you still have the remaining interest from $2 million every month.

Once you’ve bought a house, you could use that interest to support a houseful of people. Exact numbers vary by location, but the interest should be enough to keep the house in good repair, pay property taxes, pay for utilities and internet access, feed everyone, buy a junky car, and even give them all a small stipend. 

The big gains are in rent — getting a decent room can easily cost you $1000 a month these days, so eight people seeking out individual lodgings would be in for $8000 a month collectively, or $96,000 a year! But if they all live in an 8-bedroom house with a mortgage of $3000, that’s only $36,000 a year, and you save $60,000 annually. (And if you purchase the house outright, then of course there’s no rent at all.) 

We could throw together a bunch of examples of different houses you could buy, in different places all around the country. Or we could just give you a link to the Charter Houseulator, and you can take it for a spin yourself! Find a nice 5+ bedroom house somewhere using Zillow (big houses in Nova Scotia are pleasantly affordable!) and plug in your best estimates for all the variables.

You’ll notice a few things. It’s clear that a mortgage is the wrong choice here. You won’t come out ahead until 30 years down the road when the mortgage is finally paid off. If you have the money, buy the house outright.

Healthcare is the big stumbling block — in a lot of scenarios, you just won’t have enough to pay for everyone’s insurance. Residents might qualify for some kind of reduced rates depending on income, but this seems to vary a lot by state.

Even in the best-case scenarios, it’s hard to end up with enough to give your residents much of a stipend. This still isn’t such a bad deal — they get their rent, their food, and maybe their health insurance all covered. They even get access to a junky car. What more could you want? 

The situation improves a lot if you start with an endowment of more than $3 million, of course, or if you assume you can get more than 4% interest per year. But even within these constraints, you can get pretty decent living conditions for 5-8 people if you choose a house in the right place and give them a shitty enough car. Go ahead and mess around with the values in the Houseulator and find out!

II.

Charter houses could be used to fill all sorts of weird niches.

Young People

Maybe you think college is a waste of time (and really who doesn’t these days). Or maybe you just think we should make it easy for young people to take big risks, and work on moonshot projects that will take years to pan out.

In that case, a charter house could be an accelerator for young people. Lots of high school or college graduates would love an opportunity to not think about paying rent and focus on their passion projects for the next several years. 

In general, young people don’t mind a slightly marginal existence, so this setup fits them pretty well. The average 30-year-old would have a hard time accepting a tiny stipend, even if rent was covered and there were no strings attached. The average 30-year-old also probably has better options, where they can make a lot more money, even if they have to work for it. But the average 22-year-old would jump at the opportunity to [checks notes] get paid to not pay rent, and most 22-year-olds don’t have access to a better deal than this. This is even more true for the average 18-year-old, especially one that doesn’t want to bother with college. 

You might be concerned that young people would like their charter house so much they would stay forever, but this is where the very small stipend becomes an advantage. From the ages of 18 to 24 or so, survival alone is pretty enticing. But as they grow up, most of your residents will begin to dream of more than a $500 a month stipend and free rent. Soon they will hunger for more space, or nicer equipment, or a car that doesn’t have holes in the floor. They’ll find a job or some other way of making money and graduate, moving out on their own. If some of them do decide to become long-haulers, that’s ok too, since it gives your house more institutional memory.

This level of security helps people figure out their comparative advantage, and lets them found more small businesses and startups, because they don’t need to make the same kind of money right out of school. Obviously that’s good for innovation.

Research and Scholarship

Here’s a question: what’s the minimum form of scholarly institution? Existing universities are huge, but every university is made up of schools and departments, and in many cases these function almost as independent entities. How small can you go and still call it an institution? 

A charter house could be an interesting experiment in marginal scholarship. Charter houses could serve as a replacement for academic departments, possibly with a mentoring component (e.g. half of the residents are students, with mandatory turnover after a couple years). You buy a house and give it an endowment, and recruit a bunch of biologists or linguists or computer scientists, and see what kind of scholarship they produce. We don’t know if it will be good, but we’re sure it will be different.

The kind of biologists who would show up to live in an abandoned church in Oak Creek, Wisconsin or an old Victorian mansion in Normal, Indiana would be a very different kind of biologist than the kind who would take an academic job at your local university. But we think this is an advantage.

Extrainstitutional Support

There’s an ongoing conversation about how we as a society can support people who have important but hard-to-compensate roles (see for example this twitter thread). There are lots of roles, especially in open source software but also in other areas, where the work is critical but no one is willing to pony up to pay for it.

These roles don’t fit within normal funding structures — they’re too small for a business to hire the person on, too small to form a nonprofit around them, and too big to be supported through individual donations. And beyond this, there are even more projects that someone should do, and which might attract support retrospectively, but no business or nonprofit would be willing to support prospectively.

Charter houses could solve this problem neatly. A charter house or two could easily be set up with positions offered to people who are filling these roles, providing them with a minimum of support — at the very least, free rent and free high-speed internet. These people are professionals, so this may not be enough for them — but there’s no reason they can’t get support from the charter house and make additional money in other ways. They can supplement that support by consulting, getting a real job, being a bounty hunter, etc. 

In fact, since people in this position might also have a part-time consulting gig or something, a charter house targeted at them might be able to survive on a much smaller endowment, only paying for their rent, and not covering their food and healthcare, for example. 

There are a lot of projects that would have no prospective support because they’re super high risk. But if we have an ecosystem for encouraging lots of high risk projects, we will eventually get a lot of crazy successful moonshots. Our society already does this a bit for open source software — we should do it for other important avenues of progress as well. Like apenwarr says: “The best part of free software is it sometimes produces stuff you never would have been willing to pay to develop (Linux), and sometimes at quality levels too high to be rational for the market to provide (sqlite).”

Unprincipled Mixture

You could also allow a totally unprincipled combination of all of these approaches, and we think that would work pretty damn well. It’s fine if you have three engineers working on a startup on the ground floor, an essayist sharing a bunk bed with a painter in a room above the garage, and two biologists in the attic. 

A mix of approaches is good and healthy. If you fill a house with biologists, they will all be competing with each other. They may even end up at each other’s throats — they are too similar. But mix in a little diversity, a few chemists and physicists, some experts in East Asian literature, and a Turkish math wiz who speaks almost no English, and things will work very well indeed.

It’s tempting to make each charter house alike in scope and subject — one house for the college dropouts, one house for the physicists, one house for the painters, one house for the startup accelerator, one house for the mystics, etc. But siloing people in this way is going to be counterproductive. Young people will benefit from sitting across the dinner table from old people; old people from young people. Biologists will benefit from playing video games in the living room with art historians. Philosophers will benefit from going grocery shopping with blacksmiths. Electrical engineers will benefit from fixing windows with clowns. Bartenders will benefit from cooking dinner with astronomers.

As Paul Graham says in his essay Hackers and Painters, “I’ve found that the best sources of ideas are not the other fields that have the word ‘computer’ in their names, but the other fields inhabited by makers. Painting has been a much richer source of ideas than the theory of computation.” 

So it’s ok, even ideal, to have a charter house where most of the residents are college dropouts, and there’s one 60-year-old living in the basement maintaining ‘runk’.

III.

Charter houses capture a number of features of other successful programs.

They’re kind of like the Alaska Fellows Program. In this program, you stick a bunch of recent college grads in a house somewhere in Alaska, where they live together for about a year. Housing and utilities are covered, and everyone gets a monthly stipend of $1000 on top of that. We hear it works great. If young people sign up for this, you can bet they would also sign up for a program with more freedom and where they didn’t have to live through the polar night.

They’re also kind of like medieval guilds. A guild was an organization devoted to a specific kind of skill, one with practical applications, and that saw to training and organization. They pooled funds and sometimes shared tools or workshops. The first universities started out as guilds of students, who banded together to hire tutors (the first professors) and for mutual protection. Other medieval examples include various religious orders, like the Franciscans or the Poor Clares. In these particular examples you personally owned no property, but you still had a place to stay. Religious orders often owned buildings (monasteries, convents, abbeys, etc.) and conducted various forms of scholarly work together. Gregor Mendel, the father of genetics, was an Augustinian friar and abbot. 

Something like charter houses already exists during college. Particular dorms will have a particular theme, or a subset of all the people in a club or frat will live together. When people graduate from college, it’s pretty common for them to share an apartment with friends for a couple years. It’s clear that people enjoy living together like this, as long as they get their own space.

This is pretty good evidence that, given the option, young people would try living in a charter house. And it seems like this is just straight-up competitive with college in almost every way. You have to pay for your housing in college, but in Soviet Russia, house pays you a charter house pays you. In most colleges you have to share a tiny, cramped room with other people, but most charter houses would be big enough for everyone to have their own room. In college you have to study some predetermined topic and take classes, but in a charter house you can spend your time on projects that actually teach you what you need to know. In college you have to hide your drugs, but in a charter house, the chemist who lives in the walk-in closet is synthesizing LSD in the bathtub.

An example of a similar successful model is Hampshire College in Amherst, Massachusetts. At Hampshire, upperclassmen don’t live in dorms, they live in mods (“modular housing”) of 6-10 students, which are like medium-size apartment buildings. The mods are big — almost everyone gets a single, and the few doubles are huge. And you can work on whatever you think is important because of the traditional Hampshire package of no majors, no tests, and no grades (yes, really!). The only downside is that you still have to pay, but charter houses fixes this. And we know this crazy system works — Hampshire has produced alumni like Ken Burns, Elliott Smith, Lupita Nyong’o, and Eugene Mirman, the man who voices Gene on Bob’s Burgers and deliverer of the best commencement speech of all time. 

The benefit of college is of course the fact that it’s large — there are lots of people you already have something in common with, which makes it easier to build community and a strong social network. A single charter house can’t compete with that, but if you put a bunch of houses in the same town, they can support each other in various ways.

We’re not just talking about community — they can share skills and resources. The charter house full of musicians is the only house with a grand piano, but residents of the other charter houses can visit to use it. The chemists sprang for a projector or a giant TV, so everyone comes to their place for movie nights. The videographers living in the garret of the old B&B help record the experiments the electrical engineers are doing in the charter house down the street, and put it all on YouTube.

Replicating the benefits of college without the headaches isn’t just for college-age kids. Most people who went to college don’t miss the exams or the food, but a lot of them miss the sense of community and the ability to casually hang out with interesting people. Charter houses could be designed to be attractive to almost any age group.

IV. 

We’re not financial advisors, so we can’t advise on how to set up the institution behind a charter house. But we can advise a little on how we think you should organize it. 

In brief, we think a charter house should have very few rules. 

Certainly you do want some rules. You probably want to have one resident who is on all the paperwork, who can collect the interest from the endowment every month, and who is responsible for paying all the bills. You want some legal firm or something to oversee the endowment. You want rules about what happens if the endowment grossly underperforms or overperforms — what happens to a house if their $2 million endowment shrinks to $1 million, or grows to $4 million? You want rules about what happens if the house ends up being abandoned. 

(A growth rate of 4% per year does seem pretty conservative, so we support a rule that if a charter house’s endowment gets too big — if it ever reaches double the original endowment, if it breaks $5 million, something like that — it should be forced to split in half and spin off a sister house nearby.)

Other than that, we don’t think you want many rules at all.

There are many rules that do seem enticing at first glance. If your charter house is intended for biologists, you might want to make a rule that only biologists can live there. If your charter house is meant to be an accelerator for young people, you might want a rule that no one over 26 can live there. You might want a rule that no one can live there for more than 4 years, to encourage turnover and give lots of people a chance to live in the house. If the house itself has only eight rooms, you might want to make a rule that no more than 10 people can live there at a time. You might want to make sure at least a few people are living in the house at all times. Maybe you want to make a rule, “no girlfriends/boyfriends”, or at least “no families/kids”. And you would probably want some rule about how people are chosen to join the house.

These seem like good ideas, but we are against them for a simple reason: they are really hard to enforce. Who is going to go check that everyone living in the house is a biologist? If the guy playing guitar in the living room says “no I’m a biologist”, what are you going to do? If you try to enforce a maximum number of residents, how will you tell who is living there and who is just visiting? How long can someone visit for, before they count as living in the house? A week? A month? 

So our recommendation is, don’t make these rules and don’t waste time and effort on trying to enforce them. It’s fine to tell a house, “I set this up for chemists” or “this house is to support open software” or “I want to support young people, so try to graduate when you can.” But don’t try to enforce these rules — trying to enforce them will just lead to internal squabbles. 

Let the residents have friends over. Let them stay as long as they need. Let them decide how they’re going to pick their housemates. And let them learn to govern themselves. This teaches them that 1) they are capable of self governance and 2) specific tips and tricks on how to actually run a small organization/government. Pretty pro-democracy. 

So we think charter houses should have as few rules as possible. On the other hand, they should definitely have traditions. Each house should have a name, house colors, maybe a crest. A motto if they can come up with one (maybe, “I am a beautiful animal! I am a destroyer of worlds!”). Perhaps an official song or chant. Traditions like a house movie (may we suggest WPDR) or a monthly poetry contest. And of course, a party every year on the day it was founded.

You can speculate and plan all you want, but you won’t know what works and what doesn’t until you give it a go. You really want someone to try it, to start some charter houses and see what they come up with, what problems they run into, and what solutions. 

You want to invite the people who will live in the house to be your co-conspirators. If you make up a bunch of rules, even good ones, and try to enforce them, your residents will resent you. But if you bring them on board, and let them tinker with it, they will surprise you.

“Let yourself be second guessed,” says Paul Graham. “When you make any tool, people use it in ways you didn’t intend, and this is especially true of a highly articulated tool like a programming language. Many a hacker will want to tweak your semantic model in a way that you never imagined. I say, let them; give the programmer access to as much internal stuff as you can without endangering runtime systems like the garbage collector.”

We feel the same way — let them get at everything except the metaphorical runtime systems. In hacking they call this the “Hands-On Imperative”, and while actual code may or may not be involved, the charter house is more than a bit of a hacking project. “Hackers can do almost anything and be a hacker,” said Burrell Smith, the designer of the Macintosh computer, at the first Hacker Conference. “You can be a hacker carpenter. It’s not necessarily high tech. I think it has to do with craftsmanship and caring about what you’re doing.”

You want these houses to be very different, and you want to use the power of evolution. Lack of diversity is so bad that in biology, they call it genetic erosion.

You want “speciation” — you want to release ideas into the world and get feedback from their success and failure. We’re going to continue with the Paul Graham quotes for a second, because charter houses are more than a little like a combination of startups and startup accelerators. “If you release a crude version 1 then iterate,” he says, “your solution can benefit from the imagination of nature, which, as Feynman pointed out, is more powerful than your own.”

V.

Most plans to change the world require a lot of coordination. You have to argue with senators and NGOs and the university PR department, on and on and on. But anyone who can spare a few million dollars can set up a charter house unilaterally.

Haha yes, “spare a few million dollars”, you laugh, but donations like this are made to nonprofits and universities all the time. We don’t want to take food out of the mouths of hungry children, but let’s just say that some of these donations are more inspiring than others. Like the $250 million gift from Charles B. Johnson to Yale in 2013. Or the $350 million gift from Michael Bloomberg to Johns Hopkins in 2013. Or the $400 million gift from John Paulson to Harvard in 2015.

We’re not even necessarily talking about bringing in new money — you could do a lot just by redirecting donations that are already being made. If Bloomberg wants to give several hundred million dollars to Johns Hopkins (estimated endowment: $8.8 billion), who are we to judge? But in a world where we can’t seem to stop talking about stagnation and academic decline, doesn’t it seem worth it to try a different model? How about you spend $10 million to set up three charter houses with endowments of $3.3 million each, and give Johns Hopkins a mere $340 million? Or set up ten charter houses with endowments of $5 million each, and see if Johns Hopkins can survive on $300 million?

What’s gonna give you more bang for your buck, giving Stanford a shiny new engineering building and filling it with smartboards and swivel chairs and all of the engineering students who would have gone to Stanford whether it had a shiny new engineering building or not, or giving some of those engineers a house where they can work on stuff they think is cool, and enough food to keep them alive while they do it?

Another thing: those engineering students will take on like a hundred thousand dollars in debt if they go to Stanford! An advantage of charter houses is that nobody has to take out a loan.

So instead of giving $20 million to an institution that you are certain will muddle on in acceptable mediocrity, split that money up among several charter houses. Some will fizzle out; a couple may even explode. But others will become self-sustaining little critters that will spark and wriggle and lay plans of their own.