This Appendix contains data and backup calculations for the Capital chapter. These are not meant to be definitive or exhaustive, but rather to illustrate orders of magnitude.
[NOTE: This appendix is incomplete and requires a lot of additional work. At present it is mostly copied from an earlier version of the Capital chapter.]
Recall from the Needs chapter that humans require on average about 550 liters (0.55 cubic meters) of pure oxygen per day. With roughly 7.5 Billion people on the planet, that means we need over 4 Billion cubic meters/day. The Earth's troposphere contains about 600 Million cubic kilometers of oxygen, or 6*10^17 cubic meters . Ignoring all other effects for a moment, the troposphere contains enough oxygen for about 1.5*10^8 days of human breathing, which more than 400,000 years .
Of course there are also lots of technological processes, most notably the burning of fossil fuels, that replace oxygen with CO2 in the air. Conversely we have the large scale process of photosynthesis that removes CO2 from the air and releases oxygen. While the balance is an issue with regard to climate change it does not pose a short term threat to breathing — CO2 at present is only 0.04% or 400ppm (this is up significantly since the industrial revolution and cause of climate change) . Conversely Oxygen is about 20% of the atmosphere or 500 times as much .
But what about clean air? We definitely have an air pollution problem in countries such as India and China that impacts breathing. But we went through a similar phase in Europe and in the U.S. and managed to clean that up. It is a solved problem technologically. For instance, cars can be outfitted with catalytic converters and a single large plant has produced 50 million of these .
There is plenty of water in the world and we have made significant advances in desalination and in filtration. There are about 10 million cubic kilometers of fresh water on the planet (not including another 24 million locked up in ice caps and glaciers) . So that's 10^15 cubic meters. Based on the recommended 2.5 liters (0.0025 cubic meters) per day, human consumption is about 1.9*10^7 cubic meters globally per day. However, we should also include freshwater used for agriculture, livestock and general domestic use. All in, freshwater withdrawals annually are around 4*10^12 cubic meters . So, relative to supply we have over 2,600 years of remaining freshwater to meet our current needs. While 2,600 years may not seem like an extremely long timeline, don't forget that technological advancements like improving desalination processes will allow us to tap into the saline water, which makes up almost 97 percent of our water supply globally.
Again, the point is not that everyone has access to clean drinking water today. People quite clearly do not. But this is not related to a fundamental water shortage. Nor is it even related to our present ability to make and produce water filtration. For instance, filtering water for one person costs about $50 per year using modern filters . In the U.S. the average household meanwhile consumes over 30 gallons of bottled water at a cost of roughly $1.50 per gallon [total spending about $12 Billion] . The World Bank has come up with an estimate of only about $28 Billion annually to provide everyone with basic Water, Sanitation and Hygiene and about $90 Billion to make these services available continuously .
The U.S. population has more than doubled in the last six decades, as has agricultural output. U.S. agriculture now uses about 25 percent less farmland and 78 percent less labor than in 1948, so agricultural productivity is largely responsible for the increased production .
Even globally the amount of land required for farming has started to decline and we have made recent breakthroughs in vertical and automated farming. For instance, the world's larges vertical farm is currently under construction in Jersey City. The Japanese indoor farming company Spread is working on a fully automated facility that will be able to produce 30,000 heads of lettuce per day . Indoor farming uses significantly less space and more importantly less water than traditional farming.
By 2010 the U.S. housing stock was just over 235 billion square feet of residential real estate, which corresponds to about 800 square feet, or 75 square meters of floor space per capita . Obviously this is not equally distributed, but it shows that we have nearly 8x as much space on average than I had identified as a basic need.
An alternative data source is the American Housing Survey. Using this table  for 2013 I get 230 Billion Square Feet. By then U.S. population was 316 Million people which works out to 230 * 10^9 / 316 * 10^6 = 727 square feet or 67 square meter per person.
Another way to look at the physical capacity of the economy is to consider new construction. From the same Census data source it appears we are building about about (2,735 / 4)*10^3 equal to 683*10^3 units per year, with average square footage of 1,737 square feet. That means we have the physical capital to add 0.683*10^6 * 1.737*10^3 square feet = 1.186*10^9 square feet (about 1 billion square feet) per year, which is more than 100 million square meters per year and enough to meet the basic need of 10 million people .
The production of textiles, which are a key part of making clothing, has become highly automated. Apparel production, i.e. making clothes from textiles, however, is still quite manual. Based on data from a study by the Federation of American scientists  U.S. textile mills output in 2013 was $31.7 Billion with 116,805 employees for about $270K/employee. By contrast, U.S. Apparel production in the same year was $13.4 Billion with 143,575 employees for about $93K/employee. The key reason for the low degree of automation in apparel is that much of the production takes place overseas with cheap labor.
Ideally here too one could find data to analyze clothing output in terms of actual unit data instead of financial data. In the meantime here is an attempt to compare this to minimum needs. An international comparison suggests that people may be able to meet their minimum clothing needs with as little as $200 per year or even less  and .
The global apparel market was $1.7 trillion in 2012 . At the time the global population was roughly 7 billion. That works out to $242 per person and supports the idea that we have enough capital in the world to meet everyone's basic needs in clothing.
Importantly, going forward automation is coming to apparel in the form of automated knitting machines  which have been around for some time and the newer development of robotic pattern cutting and sewing machines .
Great data source here 
Highways 2012 car vehicle miles (in millions) 2,664,445 (note: includes light trucks and SUVs), 2012 passenger miles (in millions) 3,669,821, so average travelers/car = 1.38 for highways. Further supported on a separate page which shows that 76% of people commute alone.
Light Duty vehicles 233,760,558 in 2012 up from 220,931,982 in 2002 compared to U.S. population in 2012 of 313 million. That is 233.7 / 313 = 0.75 light duty vehicles per person.
Utilization of private cars is around 4%  but can be increased substantially through car sharing.
The role of capital in providing healthcare is difficult to assess. First, we are still figuring out what it means to live healthily in the first place. For instance, our knowledge of good nutrition is still quite primitive. Second, other than a few machines (e.g. for imaging) relatively little medicine requires expensive equipment. A lot of medication is expensive to buy but not expensive to make once the research has been completed. Labor accounts for 66% or more of the total expense of the healthcare system and capital equipment for around 10% or less . Third, we are just at the beginning of our ability to deliver personalized medicine and to manipulate the human genome.
Given how I have defined the basic need for healthcare though it is clear that we already have enough capital to provide it in the U.S. as our life expectancy is already above 75 years. Gains in life expectancy around the world have been tremendous in recent years. This great chart by Max Roser beautifully sums up these gains  it shows that about 50% of world population already is at or above the 75 year mark. Another 37% is between 65 and 75 and only 13% is below. The chart also shows how much of these gains was achieved since 1950.
The progress that we have made in computation is nothing if not extraordinary. I remember how excited I was when I got my Apple II in the early 1980s which came equipped with 48KB of RAM and an 8-bit processor at a 1 MHz clock speed. At the time the machine cost about $1,300 which is about $5,000 adjusted for inflation. Today a Raspberry Pi 2 computer board costs $35 (down by 99.3%) and comes equipped with 1 GB of RAM (up 21,000 fold) and a quad core 32-bit processor at 900 MHz clock speed (up 14,000 fold). Smartphones are a bit more expensive but a high performance model from Xiaomi can still be had for $100 unsubsidized. Global output of smartphones in 2015 was roughly 1.4 billion units . So without a doubt we have the capacity to equip everyone in the world with computation.
While not quite as dramatic as computation we have also made tremendous progress in networking. When I first received my Apple II was also the time when modems became popular for connecting to so-called Bulletin Board Systems. The early modems had a speed of 300 bits/second or about 40 characters/second. Today my phone on an LTE connection here in New York has a download speed of over 70 Mbps and and upload speed of nearly 30 Mbps (that's a 100,000 fold increase). Now obviously a big investment in infrastructure is required to provide everyone around the world with such blazing wireless speed but less than one might at first assume. For instance in unregulated spectrum a wifi access point can serve a small village by providing 200 or more simultaneous connections of 4 Mbps per connection for about $1,500. A 1 Gbps microwave link to cover about 4 km is about $7,500 on each end. A significant portion of the existing cost of networking has to do with the cost of spectrum as well as the cost of patents and closed source software.
Encouragingly, we have made dramatic progress in recent years with clean (from a CO2 perspective) energy sources. For instance, in 2017 Germany broke its previous record by generating 85% of its electricity from renewable sources for the day of April 30th, and this is expected to be the norm for the nation by 2030 . And in the U.S., 61.5% of new electrical generation added in 2016 came from renewable sources (biomass, geothermal, hydropower, solar, wind), the second year in a row that renewables have dominated new generating capacity . We have also made strong progress with batteries to distribute loads. And nuclear power can be provided in ways that are much safer than our large historic reactor designs. Beyond that there is nothing in physics that would prevent us from building fusion reactors. We just haven't figured out how to do it yet.