So far, most of the effort tends to focus on austerity measures and non-technical solutions to the problem. For example, the UN recently projected about $3T per year to fix climate change. (1)

At this scale, it requires:

1. International effort

2. Rich countries carrying the load

3. Severely impacting the global economy (the US spending $1T would be enormously damaging)

4. It’s basically an “anti-technology” story

This doesn't sound like something that will work unless it's really the apocalypse.

However, it's probably possible to fix climate change at 1/10th the cost, within the range for the US or China to unilaterally solve if needed.

Keys to Fixing Climate Change:

1. Reduce the cost of clean power. The only viable options are geothermal, wind, solar, fission (traditional nuclear), and fusion (advanced nuclear). I won’t talk about this much here in this blog post, but it is critical.

2. Make CO2-free options better & cheaper than alternatives. Tesla and Impossible burgers are good examples of this technique.

3. Carbon Capture - If we make it cheap to remove carbon, we can take care of the problem without austerity. 

While a lot of people are right to point out it is more energy intensive to prevent carbon than remove it. This is why the focus is so much on taxes and guilt to try and make people consume less. I believe the taxes and guilt strategy is actually counterproductive. They are politically difficult and so distasteful that people naturally fight against the idea that there is any problem at all. Even if they believe there is a problem, if they think the solution is horrible, it will make them inclined to do nothing.

Reducing the Cost of Clean Electricity Production

I will save most of this for a different blog post, but moving to clean electricity production is almost entirely a matter of willpower. It involves essentially the following items:

• Investing in increasing the use of geothermal work for both electricity and heating/cooling. It is an excellent option in many circumstances.

• Continue the pace with wind and solar energy.

• Quickly rolling out nuclear fission, which is already incredibly safe and it can be very inexpensive. France moved to mostly nuclear power in about 15 years using 1970s technology. There is no technological reason we can’t have better, safer nuclear fission providing the majority of the baseline power in 20 years.

• Investing in nuclear fusion research. This has large upside potential and enables things like widespread carbon removal efforts.

• Investing in energy storage. Storage helps solve peak load problems.

Of these items, nuclear fission is the most obvious thing that on the margin we could be doing and aren’t.

Making Carbon-Free an Easy Decision

Tech has the opportunity to make each sector better/cheaper and emit less carbon, with no tradeoffs.

The biggest categories of CO2 emissions (2) are:

• 28% of CO2 is from Transportation (mostly cars/trucks)

• 27% from electricity.

• 10% from agriculture (mostly meat production).

• 12% from heating.

The absolute best thing for reducing carbon is to make carbon-free options better than the alternative carbon-emitting option. Tesla is perhaps the most obvious example. They made a car that is widely regarded as the best car, so people adopt it.

Impossible burger is attempting to do a similar thing by making meat alternatives that are cheaper and tastier than the original products. If they accomplish this, then we should expect meat consumption to fall.

There are newer heating products that are simply better, such as advanced heat pumps and geothermal heating/cooling. These allow much more efficient operation, providing comfort at cost savings.

Dollar-for-dollar, if you can create a better carbon-free option, that is going to be the best possible use of a dollar to reduce global carbon emissions. It is scalable, with zero tradeoffs.

Beyond that, carbon capture is the next best thing to invest incremental dollars is carbon capture.

Carbon Capture

Current estimates are that we emit 40 gigatons (40 billion tons) of CO2 per year into the atmosphere. Currently, it’s about $152 per ton, so it would cost about 6 trillion dollars per year to become “carbon neutral” globally. This means removing as much carbon as we emit each year. The global GDP is $80T per year, so to do this would be about 8% of global GDP. So even today, if we scaled up our carbon removal at these prices we could become carbon neutral. If we believed there was an immediate existential threat to do so, we could do it. But 8% of GDP is a ton, and would cause a lot of suffering to accomplish. This puts some of this crisis in perspective. Excess CO2 is not likely to doom the planet. However, even if we wanted to do this, we lack scaled-up operations and global coordination to accomplish it today. We need carbon removal to scale at lower cost.

With a 10x improvement it would be about $15 per ton to remove carbon. This makes the cost to remove 40 gigatons about $600B. This is expensive but possible for China and the US to do on their own without any cooperation. It represents about 12% of the US government's annual budget. There is no particular reason to believe we can’t get to at least $10 per ton, or less than 0.5% of global GDP, or perhaps even to $3 per ton. At $3 per ton, capturing all carbon would be about $100B per year, an amount the US regularly spends on a variety of programs. So how do we get carbon removal scaled up and at least 10x lower cost? Just keep scaling.

Costs for large manufacturing-like processes almost always go down with scale. For physical goods, these are typically considered economies of scale. These economies are driven by learned improvements in manufacturing processes, technological improvements, as well as shared costs for many functions (like accountants, managers, etc). "Swanson's law" is a term for this trend in this effect in solar, which describes the steady drop in cost for solar energy over time. There has been a 20% cost reduction for every 2x in solar cumulative production (not annual production), in a trend going back decades.

Following this logic, you can map out where we are today to where we need to be. For this model, I assumed that we have removed 1 million tons of CO2 to date. This seems conservative, but I couldn’t find accurate data on this. From there, we can extrapolate to see how prices change, and the cost to get there.

Swanson’s law, and the progression of solar, is a good proxy for the progress for large manufacturing processes like carbon capture. It shows a clear roadmap towards $10 per ton carbon capture, and beyond. 

The way I see this playing out is in 3 phases:

• Phase 1: Heavy R&D - At this phase, it’s all about trying things. I expect us to take about $5B-$10B per year in carbon capture for about 10 years to get carbon capture to about $30/ton.

• Phase 2: Early scale - In this phase, you can really start to spend about $100B per year to get the cost to the $10/ton range. This would extract about 10 gigatons per year or about 25% of output.

• Phase 3: Net Negative - In this phase, we can expect large-scale operations to be running. If we can get carbon costs to $3/ton, then spending about $200B per year would extract about 2 years of carbon.

What is remarkable about this path, is that it is so much cheaper than the UN’s estimated $3T per year. In fact, it may be a $3T lifetime cost to make climate change “not a big deal”, with carbon removal. It will likely cost about $10T total to restore the earth to its pre-industrial carbon levels. It is not something we can just cut a check for today and solve, it takes a lot of time, effort, and engineering. But we can and will solve it.

Our learnings from carbon renewal will also help us understand how to do more climate engineering projects, an important part of journeying to space. This may make this whole project something that is not only a cost, but rather an investment we may use many times across the galaxy.

Climate change is a risk, but it is a risk that we can solve with minimal tradeoffs if we are smart about it. There is reason to be optimistic.

Footnotes:
(1) https://www.ecosystemmarketplace.com/articles/thanks-to-past-inertia-it-will-now-cost-between-1-6-and-3-8-trillion-per-year-to-fix-the-climate-mess

(2) https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions

Here are some articles I found interesting and related (I don’t agree with everything here but they are good reads):

(a) https://blogs.ei.columbia.edu/2018/11/27/carbon-dioxide-removal-climate-change/#:~:text=Direct%20air%20capture%20started%20out,removing%20CO2%20from%20the%20air.

(b) https://stripe.com/climate

(c) How we will fight climate change

I’d like us to become a post-scarcity society. Most people probably do. Well, most people probably don’t think about it. But the handful of us that do think about “post-scarcity” want it to happen. And we probably all agree that basic income is somehow a requirement for this to happen. It’s sort of hard to be “post-scarcity” while being scarce of money. Should we go for basic income now? I think yes, let me explain.

Basic Income Has a Lot of Theoretical Benefits

Basic income is providing every citizen with regular, flat cash payments unconditionally.  If you prove you are a citizen, you get a regular paycheck.

Replacing the current system of programs with a basic income has a bunch of theoretical benefits, like:

• It’s really efficient because you don’t need bureaucrats involved. The government is actually pretty good at taking and giving money!

• It is more equitable than retirement plans, which transfer from young to old.

• It enables more people to work on what they want or get an education.

• All those bureaucrats administering the current system can go do more productive things.

• It reduces the marginal tax rate for the poor. (For example losing unemployment benefits and food stamps is typically the highest marginal tax rate in the US).

• It replaces unemployment, which is the worst thing to subsidize (COVID’s pandemic unemployment benefits are an example of catastrophic failure here).

• It replaces minimum wages, increasing possible jobs.

• It reduces the potential for corruption because there are fewer middlemen and fewer opportunities for special interests…because there are none.

• It provides a more stable consumer purchasing base.

• It may reduce crime as a result of lower levels of desperation, particularly among the youth.

This is a pretty good list! It probably should appeal to both traditional Republicans (more freedom, fewer government programs) and Democrats (help the poor).

I think the two main counterpoints are that people will spend it poorly and that work provides meaning. These are real concerns! But frankly, the current system doesn’t work now and there is an increasing number of unemployable people. So we need to get over it.

Redistributing Better, not More

I think of basic income really as changing the method of redistribution, not the amount of redistribution. Here is a chart:

Communism is probably in the top right of this graph. Basically, take everyone’s money, and a committee decides who should get the money. “Pure Capitalism” would do no redistribution. Basically, nobody really does either pure communism or capitalism today. “Centralized redistribution + capitalism” is the most popular category today, with varying degrees of redistribution.  This is basically what you would expect. Political elites prefer to dole out benefits, as it is a very effective way to increase power and influence.

But centralized resource allocation generally underperforms market-based allocation. This is why communism fails every time. In this case, it’s pretty clear that centralized redistribution is failing: we spend $20,610 per person below the poverty line. So in theory, we could just hand poor people that money and “have no poverty”. Maybe we should just…do that?

Note on charity: Charity is great. But I can't see a real "post-scarcity" sodiety with a huge unemployable segment of the population requiring charity to live.

How much should basic income be?

It is better to have a permanent income than to be fascinating.

-Oscar Wilde

It’s pretty much pointless to talk about Basic Income without discussing the amount.  $1,000 per year and $50,000 per year are very different propositions, with likely very different effects.

There is no “right” level.  $7,000/year might be an appropriate “minimum Basic Income” since people were considered quite wealthy in the “roaring 20s” and they made an average of $7,000/year adjusted for inflation.  But, let’s take at least a rough stab at what might make sense. I think a starting basic income should:

1. Be enough for basic necessities: food, water, sanitation, clothing, health care, & shelter in a reasonably inexpensive location in America.

2. Provide some capacity to improve one’s lot in life.

3. Be small enough to provide incentives to work and the economy to flourish.

Option 1: Just give $1 more than the official poverty line.

Our first option is basically to give $1 more than the official poverty line. Poof, no more “poverty” (definitionally).

First, let’s take a quick detour to see how the US defines Poverty. It’s a bit strange, but it kind of works.

1. In 1964, the US Department of Agriculture developed the “economy food plan” which was the least expensive nutritionally adequate food plan.

2. At that time, the average family spent 1/3 of their after-tax income on food.

3. The US set the “Absolute poverty line” at 3 times the “economy food plan”

4. The US adjusted this to the CPI over time.

So yeah, this doesn't make a ton of sense. It ignores most necessities and just assumes they add up to 2x the spending on food.  This assumption makes decreasing sense over time.  There are some specific problems that have emerged since the creation of this index in 1964.

1.  The average family now spends 34% less of their after-tax income on food, despite significant quality improvements.

(Source: USDA)

2. As average food prices increase, the “poverty level” increases. So organic and artisanal food movements would by definition drive up the measured “poverty line”, even though in practice we probably all think that if some random people try some new fancy $1,000 cheese that shouldn’t really affect the poverty line. But it does!

3. The official poverty measure excludes non-cash benefits such as food stamps, housing assistance, refundable tax credits, or other government benefits.  In other words, most poverty benefits can never officially “solve” poverty. This is pretty odd.

4.  The poverty measurement assumes traditional nuclear family arrangements with an average of 2.5 related parties living in a home.  The 2014 US Poverty line is $11,670/year for an individual, and ~$4,060 for each additional person.  It is strange to assume that some humans can live off of $11,670 and others can live off of $4,060.  

So I don’t really like the methodology here.

Option 2: Do a new bottoms-up calculation.

What if we just made a new calculation! Here is what I think makes some rough sense:

So while we probably would disagree a ton on these metrics, somehow I independently came up with a number very close to the US poverty line. So while I think the method of determining the poverty line is horrible, the number it landed on seems…to basically work? I choose to think that the US government agrees with me, generally speaking. I'll call that a win.

What I propose, I’m sure some will find controversial. This is ok! For example:

1. I only budget for a reasonable, low-cost method of accomplishing a goal.  If there is a technologically free way to do something, I allocate $0 for it.  Transportation is an example, with only $50 a month budgeted.  Not much travel is related to survival, most travel is related to work or entertainment. Work-related travel should be covered by work pay.

2. We assume living in a high-cost area is a form of “entertainment”, and there is no preference given to someone who prefers to live in Santa Monica, with an average home value of $1,135,000 vs Wyoming, with an average home value of $191,000.

3. These numbers are absolute measures of poverty, not relative ones.

Is this too little?

$1k a month is not exactly “post-scarcity”. But I’d like to (1) solve poverty and (2) afford this. To practically implement Basic Income, we need to make this as affordable and practical as possible.  While we are continuing to automate a wide range of job functions, we still need a large and motivated labor force.

You aren’t flying private jets to Aspen on this, but the average US per capita social security benefit is $12,666, and this Basic Income figure is actually higher than the national poverty rate.  So there are many things that triangulate on a $1,000 per month figure.

Will this stop people from working?

It’s a legitimate concern people might stop working, but the opposite may occur. With no minimum wage or minimum benefits, companies would have more flexibility to hire workers, and labor participation would likely go up.  The marginal tax rate on the poor would decrease, and this would encourage working. Many Americans could keep expenses low and retire young on a low income, few actually do.  So there isn’t a ton of evidence that this would drive early retirement.

But, I do think we would probably see a decrease in “low paying jobs that people hate” and an increase in “low paying jobs people like”. Which seems like a good thing? I mean that is maybe the progressive dream?

I feel $12,000/year is about right to start.

Should we do this now?

No one, as a child, ever aspired to scrub toilets or flip burgers or restock merchandise. But you had to earn money to live your life, and these were the jobs being offered to tens of millions of people.

-Marshall Brain, Manna

When should basic income start?

1) when we have to

2) when it's affordable

When will we have to?

We will need to provide basic income (or something worse than it) at some point.

If you have a robot or AI that does a job better than a human for less money, most businesses will probably use them. Some won't. Some will advertise how this is handmade. But most people will buy the better and cheaper thing.

That will force people out of jobs. Traditionally, this has been great! People went from hard farming to a variety of better-paying jobs. But this was true only because a tractor might be better at farming but was worse at accounting or filming a movie or whatever. That is not true anymore. AI can now do office work, creative work, and administrative work better, with more capability every day.

Will new categories of work emerge? Maybe…but it's hard to think what those will be. I mean a farmer in 1900 certainly couldn't exactly imagine the job of a social media marketer, but they could imagine the idea of a marketer talking to people.

If the historical average productivity growth continues, in 20 years productivity will improve by 48%.  This means almost half the number of people are needed to produce the same economic output every 20 years.

How many of those cycles can occur before we run out of new classes of jobs?

My intuition is that we will start running out of jobs in 20 years. Some might argue that it’s 40, or 60. And they may be right, but I think it’s really hard to say if 90% of jobs today were gone, there will be THAT many new classes of jobs created that humans are better at. Remember they are competing against robots with superhuman smarts and strength for these jobs.

My guess: we need basic income sometime between 20 and 60 years, which to be as a society basically means we need it “now-ish”.

When will it be affordable?

The other reason to do basic income now would be that it is affordable. And it certainly is, if we dismantle the current (most expensive) system. This includes:

• All current means-tested social safety net programs (e.g. food stamps) would be dismantled.

• All age-related social programs, such as Social Security and Medicare would be dismantled.

• Any government-paid benefit, such as pension fund obligations or VA benefits, that are less than or equal to Basic Income would be replaced with Basic Income.

A decent measure of affordability is probably the cost of basic income in % GDP. The level of Basic Income compared to our total economic output gives some sense to the scale of its effect.

Let’s look at the cost of providing Basic Income today for adults.

The US Federal and State governments spend 13.5% of GDP annually on programs that would be immediately and directly replaced by Basic Income.

Bottom Line: Because our current social spending programs cost 11.4%, and Basic Income would be expected to cost 7.7%, we expect to save 3.7% of GDP by switching to Basic Income.

Over time, this should become more affordable as GDP per Capita has grown over time.

Source: Visualizing Economics

I also expect things to also get cheaper, in general. Software has already driven the cost of entertainment, education, and information to nearly zero.  For example, medicine is largely an information science and should trend towards a $0 marginal cost over time. 25% of a residential construction project is field labor. Approximately 80% of a food’s cost is labor.

There have been a few studies that also conclude basic income is affordable. Pascal J. made estimates for Canada, and determined a basic income would be affordable without any tax increases by replacing welfare, unemployment, and related services.

The $1,000 a month seems affordable today.

Summary

In summary, basic income seems to have a ton of benefits. Giving a benefit of about $1,000 per month is affordable today, and we will probably need it in 20 years anyway. So we might as well get on it, and start working on it now. Post-scarcity, here we come!

Related Links:

1: http://www.freerepublic.com/focus/f-news/1130947/posts http://povertylaw.org/communication/webinars/food-insecurity

2: http://www.cbpp.org/cms/?fa=view&id=3629

3: http://aspe.hhs.gov/poverty/14poverty.cfm

4: “Quintiles of income before taxes”, Bureau Labor Statistics, http://www.bls.gov/cex/#tables

5: “Quintiles of income before taxes”, Bureau Labor Statistics, http://www.bls.gov/cex/#tables

6: Based on my average spending a month.  The 20% is technically $37 according to “Quintiles of income before taxes”, Bureau Labor Statistics, http://www.bls.gov/cex/#tables.

7: For example, the mean Health Care spending (including insurance, drugs, & services) was $119 according to “Quintiles of income before taxes”, Bureau Labor Statistics, http://www.bls.gov/cex/#tables.  Actual healthcare spending across the US was $708, according to https://en.wikipedia.org/wiki/List_of_countries_by_total_health_expenditure_(PPP)_per_capita and $468 according to http://www.statisticbrain.com/health-insurance-cost-statistics/.  For our purposes, we will make the assumption that $119 is accurate in what consumers perceive to pay, but that the other numbers are more accurate for actual expenditures.  Assuming that the majority of much spending is adopted by the wealthier 80% of the population, we can assume a normal “livable” expenditure target to be $350, which is 50-75% of the “total spend” numbers and 294% of the perceived consumer spend

8: http://www.cato.org/sites/cato.org/files/pubs/pdf/PA694.pdf

9: http://www.ssa.gov/pressoffice/factsheets/colafacts2014.pdf

10: https://www.cbo.gov/sites/default/files/cbofiles/attachments/11-15-2012-MarginalTaxRates.pdf

11: http://data.worldbank.org/indicator/NY.GDP.PCAP.CD

12: http://washington.cbslocal.com/2012/11/15/census-u-s-poverty-rate-spikes-nearly-50-million-americans-affected/

13: http://www.cnbc.com/id/101015065

14: States noted to have spent 29% of the federal budget in  http://budget.house.gov/uploadedfiles/rectortestimony04172012.pdf

15: http://www.multpl.com/us-real-gdp-growth-rate

16: http://www.bls.gov/news.release/empsit.t01.htm

17: Wikipedia.  "Instituto pela Revitalização da Cidadania". ReCivitas. Retrieved 2013-07-24.

18: Wikipedia.  "Namibian Basic Income Grant Coalition". Bignam.org. Retrieved 2013-07-24.

19: http://www.urban.org/UploadedPDF/412903-Kids-Share-2013.pdf

And we are making progress, fast progress. In 1980, the cost per kg was almost $100,000 per kg. When SpaceX first launched the Falcon 1, the cost was about $10,000 per kg. The 2021 Falcon 9/Falcon Heavy costs per kg is about $1,000. SpaceX’s Starship has a target cost of $10 per kg.

As we increase our presence in space and make travel regular, we will reduce the cost. Musk’s $10 per kg target involves regular travel. The average human is about 80kg, so that is basically $800 per trip. That is about the cost of international air travel in many cases. But it would be nice if we could get that price another order of magnitude lower, something like $80 per trip. Then, its basically affordable for everyone for a trip. At $8 per trip, you can commute daily to and from space for things.

How might we get 10-100x cheaper than even Starship? Probably not with rockets. To do that, we need to build large structures.

There are many designs for structures, with some good ones being called:

• Space Elevators

• Skyhooks

• Launch Loops

• Orbital Rings

In each of these cases, the upfront investment is very high, so you need to be putting a lot of things to and from space to make it worth it.

The main one I will focus on is the Orbital Ring. It’s a personal favorite of mine, and it has the advantage of being almost certainly possible using known physics and material science. In theory, it should hit the $1 per kg or $0.10 per kg range possible to allow daily commuting to/from space.

I won’t go into lots of detail, but here is how they work. 

Building an Orbital Loop

The principle behind an orbital loop is the following:

1. Put objects into orbit. We do this all the time.

2. Connect the objects in orbit.

3. Do this all around the planet, and you create a ring around the planet. The objects are in orbit, which means they are moving really fast. At this phase, you haven’t solved anything because someone leaving earth would need to get to orbital speed to land on the ring. So we need to keep going.

4. Put magnets on this fast-moving ring.

5. Use the magnets to suspend a tunnel around the fast-moving ring, similar to the way maglev trains are over the track.

6. Now you have a stationary tunnel-like platform around the earth, with a fast-moving ring in the middle holding it up with centripetal force.

4.1.2. Getting to the loop

Putting the ring about 100km up makes sense. It is above the atmosphere, which is what really matters. It has basically the same gravity as the Earth’s surface, just higher up. But it’s only 100km up, which would take about 10 minutes to travel with current high-speed trains.

100km is a bit long for a cable though, you can’t just use steel for this. You need something a bit more exotic.

Here are some materials we could use:

Kevlar is the most obvious material that can work. It has ton of extra strength, and we already produce it in large quantities. Once you have the Kevlar support structure up, you can attach plumbing, power, and other things to the line as well, that part is easy. We have oil pipelines and such much longer than that.

So now you have a platform in space, and you have cables going down to earth. What can you do on it?

Well, you can do just about anything on it like space hotels. However, the main purpose of this orbital loop is transportation, which in space really means acceleration and deceleration.

Launching from the Orbital Ring

To get places in space, you need speed. A lot of speed. Almost all the energy of a rocket is used to get up to speed, not altitude. The orbital ring gives you 2 huge advantages:

1. No atmosphere. Travelling on it is outside the atmosphere. Cutting through the atmosphere consumes a ton of energy for rockets.

2. No need to carry fuel. Normal rockets require you to carry fuel to get it into space, but then you need more fuel to carry the fuel, etc. This makes most rockest mostly fuel. Orbital rings solve this, you just launch from a maglev track. You don’t necessarily need any fuel at all.

If you flip the ship upside down, and are not moving, it will feel like 1G “up”. So if you accelerate to 2Gs on the track, while upside down you will feel 2G “up” and 1G “down”, leaving you with 1G “up”, or normal gravity.

You can easily surpass Earth’s escape velocity of 11.2 km/s with this method, and get to Mars or nearby space stations.

You can make more rings at different angles to allow you to point yourself at various locations. You can connect these other rings together. You can build larger rings, which allow you to go even faster.

When you get to another location, you can slow down in other ways: rockets, lasers, maglev tracks, or another orbital ring. The most important thing right now is making getting off Earth easy. Its much easier to leave Mars, and trivial to leave a space station.

Orbital rings, or something like it, is a really good step to making humans move seamlessly between space and Earth.

How Expensive Is It?

How much would it cost to build an orbital ring? If Starship works, it becomes in the range of building now. Here is roughly how it might work:

1. First you build a small loop around the earth, something as small as a wire. It has to go around the 40 million km circumference of the earth, which is ballpark 10 million kg for the wire. The stationary ring might be 20 million kg, for a total of 30 million kg.

2. If Starship hits their target of $100/kg, this is $3B in launch costs.

3. The materials cost would be about $500M.

4. The labor would cost…a lot. We won’t have astronauts doing this work. Oil crews can lay about 1 mile per day, so this would take about 29,000 days to complete. But if you have 100 robots working in parallel, you can complete the thing in under 1 year. I’m not sure anyone has priced this out completely, but having 100 robots seems pretty good. Its not bespoke, nor large scale, so I wouldn’t be surprised if each robot cost like $50M, so like $5B in robot costs.

5. This doesn’t feel like the hardest R&D task ever done in space, so maybe $10B in other costs? In some ways, this feels easier than building a tunnel through Boston. But in other ways, its space?

This puts an estimated total of $18.5B. You probably want to just double it because its crazy, so maybe its $40B.

Once the first ring is up, it can then be used to shuttle up mass to bulk up the rings and get it ready for launch. You can probably re-use the robots and techniques you learned, so probably the transportation and materials cost are the main blockers. Now launch costs are much cheaper now, so 10x the size of the ring is probably now more like $5B. If you wanted to 1000x the size of the ring, you probably will keep seeing more scale benefits, and can probably do that for $200B. At that point, its probably strong enough to carry and launch things without issue.

So you are maybe looking at $250B to get a usable orbital ring, able to launch things into orbit. On the one hand, this isn’t something you cavalierly construct. On the other hand, the SLS rocket will cost something like $25B to develop, and that just gets us an old-fashioned rocket. And we spend way more than $250B on projects all the time (e.g. COVID relief, wars, etc).

This is also something that someone can just start doing? I mean, for about $200M you can probably build a robot load up a starship with some material, and send that thing to space, and just start building.

Can we do this tomorrow? It seems like “yes”, but we probably shouldn’t. I mean waiting on Starship is a no brainer. Then we probably want to do

An orbital ring is something we should seriously consider and plan for as a follow-up to Starship. It’s my favorite approach right now to make traveling to space as easy as a daily commute.

My favorite way to explain this is to talk about disassembling Mercury. Yes, taking the whole thing apart.

Mercury is the smallest planet, it is about 1/3 the size of Earth.

It is literally a rock, floating around the sun. No life. There are probably some Mercury enthusiasts, but it can’t be many.

So let’s take it apart. I mean, when we talk about taking it apart, we will create Mercury preservationists, campaigning to keep the pristine rock beauty intact for nobody to see.

It’s a rock that has a mass of about 3.285 × 10^23 kg, or about 300 billion gigatons. That is a lot of stuff! We could live on top of the rock, but that is pretty inefficient. If you gave me a giant pile of lumber, my first instinct wouldn’t be to make a wooden mountain and live on it. What if we make it into rotating space stations instead?

There is a type of space station called an “O’Neil Cylinder”, which is basically a large rotating tube. These stations are designed to be about 8km in diameter, and up to 32km long, so about 900sq km. This area is a bit smaller than the US state of Rhode Island and about the size of Los Angeles. So living in one would be…pretty normal for most people? Most people I know in LA don’t leave that often, it’s not a major restriction. Rotating the habitat generates centripetal force as a form of artificial gravity. This way, when you stand on the inside of it, it feels like Earth gravity…or mars or the moon, whatever you like. Lower gravity seems like it could plausibly create less stress on the body, and it’s one way to lose weight. So I suspect something like 80% of Earth’s gravity is optimal for humans, but who knows, we haven’t tried.

The best estimates for the mass of these things are 4-6 gigatons, mostly made of dirt, metal, and water.

But let’s assume that’s wrong, and it’s actually about 10 gigatons. Let’s also assume, and only about 1/3 of Mercury is usable then we can make about 10 billion stations.

This creates a living space of about 9 trillion sq kilometers. Earth is only 317 million square kilometers. This means we can make the landmass of over 28,000 earths, just out of Mercury, using very conservative assumptions.

If each of these calendars had 1/3 the population of Los Angeles, then you could fit 10 Quadrillion people in them. For most people, a “much more spread out Los Angeles” doesn’t sound like hell, but hey we don’t need to go for 10 Quadrillion people. We could go for like 1 Quadrillion people? I picked the subheader of the blog targeting 1 trillion people because I thought that was the largest number that people could sort of understand. At 1 trillion people, everyone basically has their own city’s worth of space. So this is far from hell, this is paradise.

It’s also rare to have visions of the future where you can be off by a factor of like 10,000 and still be right.

How we do it

Given that we CAN easily get a trillion people in space, HOW do we actually do it?

There are not really big unknowns from a physics perspective. There are 2 main things that we do need to start figuring out, and these will take a while:

1. How does the human body withstand long periods under artificial gravity?

2. How do we actually build something this big?

I’m not too worried about each of them, but I think getting a great start would be trying to build a rotating space habitat that is large enough to run tests, maybe 1-4 years long, to see how people do. SpaceX’s starship makes this kind of thing much more viable. Each of them is significantly bigger than the ISS. Tie two of them together and give them a spin, and you have a good experiment of the medical effects. ISS astronauts live in much smaller spaces for a year, and some people don’t leave their neighborhoods for a year. So this doesn’t seem that hard to test.

I’m interested to see tests on low-G (like the moon or Mars gravity). Based on what tests we have seen in zero gravity, there are negative health effects of zero-G. But it isn’t THAT bad. When we first sent astronauts up, we didn’t know if they would all die after 1 year in zero-G. So that’s a win! It’s pretty reasonable to think that zero-G is bad (our bodies were designed for gravity) but again low-G is maybe good?

Building something this large also doesn’t seem “that hard”, it’s just a matter of scale and time. If I asked a warehouse building company to build a warehouse 10x the size of the largest one, they could do it. It’s a lot of engineering and work, but it is something that can be done. Each individual O’Neil calendar seems easier than the Large Hadron Collider, for example. It doesn’t have to have the really hard things that are here on earth like “dealing with politicians” and “coordinating between countries” and “finding suppliers in good districts.”

We have a lot of work to do before actually trying to disassemble Mercury. And maybe it won’t work. But you probably haven’t thought about disassembling Mercury before, and it is my favorite example of how big we can make humanity without even leaving our cosmic backyard.

This is rare, and I know it’s rare. Most people do not think about where the world should be and think about how to get there.

This blog is a small effort to try and improve that. There is a huge gap between what we can do, and what we should focus on. “The possible” is much better than commonly imagined. We need to upgrade our optimism. My pithy tagline for this is Let’s try to get a trillion, rich superhumans in space.

I expect to talk about anything that fits this theme of expanding our optimism. I’ll speak both of short-term policy changes and big ideas that are not given the attention they deserve.

Here is a shortlist of some big ideas that we can do as humans if we focus on it:

• We can have trillions of humans. Actually much more than that, but a trillion is about the highest number people can think about.

• We can populate space. Space does not have to be expensive, and commuting to and from locations around space can be fundamentally inexpensive.

• We can become superhumans. Healthy, strong, exotic, and immortal.

• We can be rich. Starting with energy, we can have a huge abundance of things. Not unlimited, but it’s “easy” to have a trillion people each with 10x the energy budget of today.

• We can live quadrillions of years. The universe will be habitable for a very, very long time.

In the short term, I have some weird ideas that could dramatically improve the US, if it chose to use them. Note, I will avoid politics and stick to ideas far outside the political norm. The main ideas that don’t get as much attention as they should are:

1. Real Estate should be deliberately a "bad investment”. People should be able to live in fabulous homes at low cost.

2. Healthcare needs to focus on cost reduction, not the price people pay. These are two fundamentally different things.

3. Immigration. We should try to get a billion people in the US with immigration, targeting about 3%/year growth. This growth would make the US vibrant, richer, and more powerful.

4. Basic income is probably as a social safety net than our current programs.

There is a lot of nuance that needs to be discussed, and I’ll add to this list over time. But I will get into these in future posts.