Launch Cost Thresholds for Economic Activity

Introduction

The historically high cost of access to space is a major constraint on space-based economic activity. To date, there have been relatively few types of profitable commercial space activity, with communication relays and imaging being the most prominent. While many other types of economic activity in space have been proposed, none have come to fruition, and the main obstacle is the high cost of space launches.

For each type of economic activity, there is some threshold for the launch cost per payload mass beyond which the economic activity will not be sustainable. If launch cost increases above this threshold, then the economic activity will not be sustainable in the long term. In business terms, the activity must at least break even. This may not be a requirement for the activity to happen at all, as some businesses will run at less than breakeven for extended periods of time if they can obtain funding and think the eventual returns to be sufficient, but it is necessary for it to become commonplace.

Since launch cost is usually reckoned in launch cost per payload mass, the necessary breakeven or better condition may be expressed as shown in Equation (1). \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} {R_{{ \rm{gross}}}} - {C_{{ \rm{non - flight}}}} \ge {C_{{ \rm{threshold}}}} \tag{1} \end{align*} \end{document}

Equation (1) shows that 3 factors contribute to meeting the launch cost threshold: the revenue per payload mass, the nonflight costs, and the cost-effectiveness of the launcher itself. The nonflight costs are primarily a nonengineering consideration and so for the most part will not be treated in this study. This leaves 2 terms to consider: the revenue that can be earned by the activity and the cost threshold itself.

Determination of this threshold is important for both evaluation of potential economic activities and the development of new launch vehicles. For example, a highly expensive launch vehicle development program may be justified if it can make a correspondingly profitable economic activity viable. Conversely, if an economic activity requires a significant reduction in launch cost from the present state of the art, any discussion of its feasibility must include either the cost of the necessary launch vehicle development or some indication that a suitable launch vehicle is under development.

Estimation of the Minimum Launch Cost

Computing the launch cost thresholds that must be met for certain proposed economic activities to be viable is a worthwhile exercise in its own right. To make decisions about future efforts it is necessary to know what the current cost for space access is or will be at some specified time in the future. Recent events have drastically altered the landscape of the launch vehicle market, and all participants are still coming to grips with these changes.

With the successful recovery and reuse of Falcon 9 first stages, ¹ reusable launch vehicle capability can be considered to exist. Many of the operational details, especially the critical repair, overhaul, and maintenance (ROM) costs, are still unknown and can only be determined with continued maturation of the system. Nonetheless, application of such costs from previous analogous systems can provide estimates of what these costs will be for Falcon 9.

Fuel is virtually negligible as an expense for current space travel. SpaceX claims that fuel expenses are ∼0.4% of a Falcon 9 launch's costs. ² This results in a fuel cost of $10.61/kg of payload delivered to low Earth orbit. Koelle ³ gives $1.2M for launch operation costs for a TSTO (two stage to orbit) launch vehicle, which for a vehicle of Falcon 9 capacity would amount to $52.63/kg. This sets a floor at $63.24/kg for fuel and operations assuming full reuse of the launch vehicle.

Hardware cost typically increases with mass and will be assumed proportional for this analysis. This is a standard assumption in the aerospace industry; for instance, various RAND cost estimation relationships for aircraft ⁴ are commonly used to estimate aircraft costs based on gross takeoff weight. For Falcon 9, the second stage is not reusable, and its mass is ∼13% of the total mass expended. This results in a vehicle that is 87% reusable less ROM costs. The remaining reusable portion of the launch vehicle requires that some ROM cost be paid for each flight. This is unknown yet, but X-15 ROM costs were ∼3%, ^{5 *} and Space Shuttle ROM costs were ∼21%–24%. ^{6 †}

Zero percent ROM on the first stage (i.e., the limiting case), combined with the current expendable second stage, would yield a cost efficiency of $410/kg of low Earth orbit (LEO) payload capacity. Ten percent ROM yields a cost of $640/kg-LEO. While this is nothing short of revolutionary, it still represents a limiting factor for economic activity in space.

Making Falcon 9 completely reusable (less ROM costs) or designing a new completely reusable launch vehicle has the potential to reduce launch costs below this. How low is dependent on relatively small differences in ROM costs and second stage structural fraction that are difficult or impossible to prognosticate at the present. Table 1 contains cost effectiveness values for different values of ROM fraction and second stage structural fraction.

Of course, there are other factors in launch price than the costs required to carry out the launch. As of this writing SpaceX is planning to offer a 30% discount for a customer willing to use a reused first stage, ⁷ which results in a far higher cost than $640/kg-LEO. Several factors could play into this. First, ROM costs could be much higher than expected. For example, the recovery of first stages used for certain payloads still has a high failure rate, leading to a high ROM cost. Second, SpaceX needs to recoup the cost of developing the reusability capability, and third, they simply charge what the market will bear, as there is no competitor in the reusable rocket market.

All of these factors are subject to change over time. ROM costs will decrease over time as a result of increased experience with the system combined with the opportunity to redesign components for increased life span and reduction of booster losses. Eventually, the research and development costs will return, and it is likely that eventually a competitor will reproduce the technology and bring competition to the market. Two competitors offering theoretically similar capabilities, Blue Origin's New Glenn ⁸ and ULA's Vulcan, ⁹ are under development. However, the analysis presented above represents the minimum amount that can eventually be paid assuming that it is not possible to significantly reduce the cost of actually constructing and launching the vehicle from what it is at present and that some amount of ROM costs must be paid.

Communications

Telecommunications was the first commercial application of spaceflight. Before the advent of geosynchronous telecommunication relays, intercontinental voice and data (telegraph) communications relied on subsea cables. An example of this was AT&T's TAT-1 (Trans-Atlantic Telephone 1), completed in 1956. This cable cost $35M in 1956 dollars, equivalent to $245M in 2017 dollars using the implicit price deflator, and had 36 telephone circuits, ¹⁰ resulting in a cost of $972k per circuit. Each of these circuits could carry 1 telephone call or up to 50 concurrent telegraph signals. ¹⁰ The use costs were correspondingly high for the end user, with transatlantic telephone calls costing $4 per minute during business hours, or $28 per minute in 2017 dollars, a staggering sum today.

On April 5, 1965, less than 10 years after Sputnik, Early Bird (Intelsat 1) launched from Cape Canaveral. Whalen states that the total cost to COMSAT (later Intelsat) for Early Bird was approximately $7M in 1965 dollars ($42M 2017$). ¹⁰ Of that cost, Early Bird itself accounted for slightly less than half, the balance being the Delta D required to launch it. For this mission, the launch vehicle had a cost per payload delivered to geosynchronous transfer orbit of $55.5k/kg or approximately $3.72M total launch costs in 1965 dollars.

Early Bird was tiny compared to modern communication satellites, with a mass of 67 kg, including the apogee motor, compared to an average geosynchronous transfer orbit payload of 3,282 kg over the period examined in Boone and Miller. ¹¹ However, it had 240 telephone circuits. This resulted in a cost per circuit of $29.1k in 1965 dollars. Whalen states that the ground stations for early telecommunication satellites in most cases cost as much as the satellite and launch vehicle, putting the hardware cost for the system at $58.2k per circuit (351k 2017$ per circuit). AT&T's lowest estimates at the time for the cost of constructing a follow-on subsea cable were approximately $100k per circuit (603k 2017$ per circuit).

Comparing Early Bird to TAT-1, which was actually in service at the time of its launch, the economic case for communication satellites at that time is clear in terms of hardware. Early Bird's total hardware cost, including a ground station, was less than 6% of the cost of an equivalent existing subsea cable. Given this enormous comparative advantage, the launch cost threshold is very large. If the entire system merely had to be equivalent in price per circuit to TAT-1, the flight hardware could cost up to $486k per circuit, with the balance being required for the ground stations. Early Bird itself cost approximately $13.6k per circuit, leaving $472k per circuit for launch costs. The geosynchronous transfer orbit mass per circuit is 0.279 kg per circuit, leaving a geosynchronous transfer orbit launch cost of $1.68M/kg of payload in 1965 dollars or $10.1M in 2017 dollars. For comparison purposes, most current launch vehicles cost less than $20k/kg to geosynchronous transfer orbit. Even comparing Early Bird to a theoretical $100k per circuit subsea cable as planned but not yet built by AT&T at the time, it still has a significant upside, costing less than 60% as much. In this case the launch cost threshold is still $130k/kg (783k in 2017$), more than twice what Delta D actually cost.

Of course, this estimate includes no research and development or other costs for either system. Whalen notes that several organizations both public and private had spent significant sums on communication satellite development, much of which they had no opportunity to recover. ¹⁰ Nonetheless, it is clear from his work that decision-makers in telecommunications had thought about the potential of telecommunication satellites before Sputnik. Such satellites were built as soon as was practical, for the reason that even the earliest launch vehicles were well below the cost threshold, which was very high due to the great expense of subsea cables.

Since Early Bird, the telecommunication market has changed. The advent of fiber-optic cables dramatically improved the cost-effectiveness of subsea cables by greatly increasing their bandwidth, and these cables transfer the majority of voice and data communications today. Cables are also not affected by atmospheric conditions, which can interfere with satellite communication relays. As the cost-effectiveness of subsea cables improved, communication satellites shifted to television and radio transmissions, which are their primary purpose today. They are also useful in austere environments where it is impractical to run cables. As a result of technological advancements in other fields, the launch cost threshold for this economic activity has shifted and, consequently, shifted the paradigm in this field back to cables.

Tourism

Space tourism is probably the most commonly proposed “next big thing” in space, with a number of start-ups devoted to advancing this form of commercial activity. These proposals fall into 2 categories: suborbital excursions lasting a few minutes and proposals for orbital hotels or habitats for extended stays in space, such as the Bigelow BEAM. ¹² This work will only consider orbital stays, as orbital and suborbital systems are not directly comparable, since attaining an orbit where the perigee is above the atmosphere requires much more energy and is a far more difficult task than simply crossing the von Karman line into space for a short period.

It might be argued that the threshold for space tourism has already been met, since space tourists have flown to the International Space Station. ¹² However, these individuals were both extremely wealthy and space enthusiasts and so represent an infinitesimally small portion of the total world population and the total population of people who take vacations.

Therefore, this analysis is somewhat different than those for economic activities producing some commodity that has a singular value. Individuals in any society each have a certain amount of disposable income that they can spend on luxuries, such as vacations. This availability of funds and the individual's enthusiasm for going into space both play into the cost threshold. Research conducted in Japan in 1993 also provided direct insight into what individuals there were willing to pay for travel to space. ¹³ In a 2001 survey across the United States, Japan, Germany, and the United Kingdom, 10% to 20% of respondents indicated that they were prepared to pay as much as 1 year's salary for the opportunity to take an orbital vacation. ¹² Crouch surveys previous work on the demand for space tourism in “The Market for Space Tourism: Early Indications.” ¹⁴ The results of the studies reviewed there indicate that the worldwide market for tickets to space costing more than approximately $250,000 is in the hundreds. Given that only 3% of the United States' households have a higher income than this, his results are in line with the previously referenced research.

This analysis will assume that the average sums spent on vacations are relatively constant. The Bureau of Labor Statistics states that Americans spend roughly 3% of their household expenditures on vacation travel. ¹⁵ Studies in the travel literature corroborate this^16,17 and provide additional information on the distribution of expenditures, with Melenberg and van Soest determining that less than 2% of the Dutch population spends more than 20% of their annual expenditure on vacations.

A combination of these data with Census data on the income distribution of the United States' population ¹⁸ and the Bureau of Labor Statistics' survey data on household expenditures ^{19 ‡} yields the average sum that each household spends per year. This analysis will consider 2 scenarios as follows: a yearly vacation using the vacation expenditures for a single year and a “once-in-a-lifetime” experience costing 30 years' worth of this vacation expenditure. Two spending levels will be considered as follows: the average of 3% and an upper estimate of 20% based on the Melenberg and van Soest article. Table 2 presents the results of this analysis.

It is important to note that the models for vacation travel expenditures developed by Fish and Waggle and Melenberg and van Soest are regression models intended to cover a certain band of incomes and/or annual expenditures and do not accurately model the spending behaviors of the extremely wealthy. In the Fish and Waggle model, vacation expenditures rise above 100% at a total annual expenditure of $5.4M per year in 1990 dollars, and in the Melenberg and van Soest model, vacation expenditures become negative at annual expenditures of greater than $523k in 2014$. The significance of this is that the wealthy, who are the most likely first adopters of any luxury, are not modeled well in the literature on vacation expenditures. Melenberg and van Soest do have real survey data for the distribution of the percentage of expenditures spent on vacations, but this distribution is capped at 20% of expenditures.

Estimation of the mass requirements and ROM costs for manned space vehicles is somewhat uncertain at present. There is a turnover in this currently, with the first commercial manned space systems soon to come. Three new private manned space systems are in an advanced stage of development: Boeing's CST-100 Starliner, Sierra Nevada Corporation's Dream Chaser, and SpaceX's Dragon v2. It is unclear what maintenance costs will be for any of these systems, although they are all designed to be reusable, in contrast to current systems such as Soyuz.

Dragon v2 carries 1 pilot and 6 passengers while massing ∼6,000 kg ²⁰ and will serve as the baseline for this analysis. This results in a launch vehicle payload mass requirement of 1,000 kg per passenger. The analysis will not consider ROM costs for the passenger-carrying portion of the vehicle, as it is uncertain what any of these spacecraft will actually cost, so estimation of ROM as a percentage is not helpful.

The contribution of fuel costs to total launch cost for Falcon 9, at $7.96/kg of LEO payload capacity, is more than the launch cost threshold for a single person's average yearly vacation expenditures for all but the wealthiest income group in the Census data. This fuel cost represents a physical limit on the minimum cost of chemical rocket powered access to space ^§ . However, the Census data put all incomes of $250,000 per year and higher in a single group, so the very wealthiest people in American society are not sufficiently differentiated to determine where the limit would be. For the “once-in-a-lifetime” scenario, things look more promising. The $15,000–$19,999 bracket could pay the fuel cost for a single individual, and the $65,000–$69,000 could pay the fuel cost for a family of 4.

For the top 1.7% of vacation spenders, the thresholds are much lower. As stated earlier, Melenberg and van Soest do not provide a breakdown of the vacation expenditures of this group, but it is no less than 20%, and so 20% may be used as a conservative estimate. Americans who fall into this category and make more than approximately $45,000 a year with this amount of expenditure could pay the fuel cost for a single person. Wealthier persons could actually afford yearly trips to space given a fully reusable launch vehicle with low enough ROM and fixed costs. For the “once-in-a-lifetime” scenario, single individuals could actually afford to travel into space on partially expendable systems such as the current Falcon 9.

Of course, the price of actual tickets will greatly exceed the launch costs in the long term. Section 4 of the FAA report on “Economic Values for FAA Investment and Regulatory Decisions” presents data indicating that the direct and indirect costs of operating aircraft excluding crew expenses are ∼49% of an airline's expenses, with 28% of the total being fuel and oil costs. ²¹ This cost is analogous to space launch costs in a space enterprise and implies that the final ticket price will be at least twice the passenger's share of the actual launch cost.

How many of these wealthy individuals with high vacation expenditures are there? As stated before, there are not much data extant on the vacation expenditures of the wealthy. Three percent of the households in the United States earn more than $250,000 per year, so if 1.7% of these households spend more than 20% of expenditures on vacations, this amounts to slightly more than 65,000 households. It is difficult to rate whether this is a conservative or optimistic assumption from the works cited in this article, as they do not consider incomes this high and extrapolation of their 2 models to high incomes actually results in opposing conclusions.

As discussed elsewhere in this work, it is difficult to estimate in advance what the ROM costs will be. Experience does indicate that great improvements are possible over time. As an example, the first successful production axial-flow turbojet, the Junkers Jumo 004B, had a time between overhaul of only 50 h. ²² In contrast, the Pratt and Whitney F100-PW-229, a mature contemporary turbofan design with a similar mission, has an overhaul interval of 6,000 cycles, ²³ which is roughly equivalent to 4,300 h. While these values do not translate directly into a proportional reduction in ROM costs, it is clear that continued development in both design and material science can drastically increase the life span of most components and systems, which will usually have the effect of reducing the ROM costs.

Since the launch cost thresholds are above what most Americans could actually afford, it is valuable to consider the inverse of the question; that is, what income is required to afford a space vacation at current launch costs. The results indicate that chemical rocket powered access to space is unlikely to lower the launch cost enough for the majority of the population to ever engage in it on a routine basis. At $100/kg-LEO, a price point that requires a fully reusable vehicle with less than 2% ROM costs, a single ticket to low Earth orbit would cost at least $106k, irrespective of overhead or profit, and not including any maintenance on the spacecraft other than ROM costs for the launch vehicle itself. However, at this price point, individuals with sufficient determination and willingness to sacrifice for the goal of visiting space will be able to achieve it.

While the discussion of private manned access to space is typically examined in terms of pleasure tourism, reduction of costs to these levels would make other reasons for manned space travel more attractive and probably more practical than tourism. Aside from greatly increased possibilities for human-conducted scientific research and the possibility for organizations to send their own scientists rather than rely on NASA astronauts, industrial ventures often spend quite large sums of money maintaining employees in severe environments, if the profits of the venture justify it. Industrial ventures also have the potential to drive colonization, which from an individual of modest means' perspective makes much more sense than tourism at the chemical rocket's price point. Few people might be inclined to spend a significant portion of 30 years' income on a vacation, even one to a once-in-a-lifetime destination, but many more would be willing to make an investment in their own future if there were some kind of economic activity that was sufficiently profitable to repay their investment.

It is important in this context to consider that a personal space adventure has no collateralized value (aside from marketing), and as such, the emergence of a mortgage type business to fund this expense up-front is unlikely but not absent when considering creative payment structures. This will have a braking effect on space vacations at this price point; even those highly motivated to take them may be unable to save for 30 years to do so. Other uses of human space travel, such as permanent colonization or industrial ventures, are unlikely to suffer from this problem.

Discussion of current airline expenses provides an opportunity to investigate another sort of threshold. Some as-yet unknown combination of fixed costs, ROM costs, and reusability is required to meet the long-desired “airline-like operation” criterion, where the percentages of the total cost dedicated to fuel are equal between spaceflight and terrestrial flight. If the fixed launch costs were negligible, ROM of up to 0.29% per flight with full reusability would allow launch prices of $37.89/kg of cargo if upper stage structural fraction remained at its current value.

It is not clear to what degree it is possible to reduce the fixed launch costs. If 1 considers such necessities as air traffic control and runways to be fixed costs for the launch of terrestrial aircraft, it is clear that there will always be some fixed cost. If significant reduction in the fixed cost is not possible, it is difficult to reduce launch costs for current medium-class launch vehicles to less than $100/kg-LEO, regardless of ROM cost reductions.

Space-Based Solar Power

Dr. Peter Glaser first proposed space-based solar power as a means of generating terrestrial power in 1968. ²⁴ Since that time, a number of studies have examined the concept in detail. The most famous of these was the joint study carried out by the Department of Energy and NASA ²⁵ in the late 1970s. This plan envisioned a massive 50 square kilometer solar array assembled by human workers in geosynchronous orbit. The projected cost was consequently enormous.

A more recent proposal by John Mankins ²⁶ offers a prototype of a space-based solar power solution with innovative cost-cutting techniques incorporated into its design. In short, SPS-Alpha is a design for a modular, self-assembling, space-based solar power system. ²⁶ This drastically reduces costs by 2 distinct and complimentary routes. First, it leverages learning curve effects to its advantage, because it is composed of a few types of modules produced in mass quantities, and second, its ability to self-assemble obviates the need for the highly expensive human construction workers or sophisticated robotic assemblers required by other space-based solar power schemes.

Mankins presents economic analyses for a series of progressive alternatives for orbital power generation installations. These are design reference mission (DRM) 3 Case 1, a pilot plant, DRM 4 Case 1, a full-sized plant that assumes minimal technological improvements over the current state of the art, and DRM 5 Case 4B, a full-sized plant designed for serial production that assumes significant improvements in solar panel and radio frequency power conversion efficiency. The prices per kilowatt-hour for each of these cases are $3.26/kWh, $0.15/kWh, and $0.09/kWh, respectively. These results assume a launch cost of $500/kg-LEO.

Estimates presented in “Space-Based Solar Power” indicate that 1 GW of continuous terrestrial solar power generation costs approximately $20.9B in 2013 dollars, with 30 years of ROM costs being $4.1B. ²⁴ This equates to a cost of $0.095/kWh if the cost of the plant is to be amortized over 30 years.

$500/kg to low Earth orbit is at the edge of current capabilities. If Falcon 9 first stage ROM costs reach ∼4% or less, it will pass this threshold even while expending the second stage. Given that the X-15, with a similar flight envelope to the Falcon 9 first stage, achieved 3% ROM costs in a one-off, experimental program 50 years before the first Falcon 9 first stage was successfully recovered, reaching this threshold without significant advance over the current state of the art seems quite possible ^** . A fully reusable TSTO based on or similar in cost to Falcon 9 could reach this cost threshold even with relatively conservative second stage structural fractions and ROM costs, as indicated in Table 1.

DRM 5 Case 4B assumes, as the author puts it, “aggressive technological advances.” The more realistic DRM 4 Case 1 costs more than terrestrial solar generation even with an optimistic $500/kg-LEO launch cost. Mankins' very detailed cost analysis makes it possible to compute that for DRM 4 Case 1 to be equally cost-effective as a terrestrial installation would require a launch cost of approximately $155/kg-LEO, far below anything currently available. A highly reliable, fully reusable vehicle with ROM costs of 3% or less is required to make this project viable in the near term.

The preceding is assuming space-based light harvesting for transmission and consumption on Earth. However, as space-based operations increase, the demand for energy is most likely to come from these emerging operations in which space-based power production will have no substitute or rivalry from Earth-based power production. However, prognosticating the rate at which industrial development will occur in space is difficult at best and so cannot be relied on as a source of future demand. As was the case with the earliest New World colonies, any future space enterprises must be centered on producing commodities for use on Earth for many years to come.

Space-Based Mining

Space-based mining has recently received some attention as a potential commercial use of launch vehicles. The total amount of materials in the asteroid belt is massive, with enough iron to cover the entire surface of the Earth to a depth of 80 m ²⁷ and large quantities of other materials.

Several processes have been proposed for resource extraction from asteroids or comets. They differ in their approach to obtaining the resources, separating them from the ore, and delivering them to their final destination. The cost of each, and consequently the launch cost threshold, depends heavily on the specifics of each scheme. The launch cost threshold also depends on the value of the resource to be mined. This can be complex as well, as some of the schemes proposed, such as mining water for conversion into station-keeping fuel for geosynchronous satellites, involve a resource whose current value is not known with any degree of accuracy and is dependent on future changes in markets.

The asteroid retrieval mission proposed by NASA ²⁸ proposed returning an entire asteroid to lunar orbit, where it could be investigated by a team of astronauts sometime in the future. While not strictly speaking as an attempt to extract resources from an asteroid, it does provide a detailed estimate of the cost of material retrieval using current technology. For an initial mission, it proposes retrieval of a 500-ton asteroid, with a total mission cost of $2.6B, for a total cost of $5,200/kg of retrieved material. Follow-on missions would cost approximately $1B or $2,000/kg of material. These costs assume that the mission requires an Atlas V 551, with an 18-ton capacity to low Earth orbit costing $288M, making the assumed launch cost for this mission $15,300/kg.

Examination of the cost estimates provided in the “Asteroid Retrieval Feasibility Study” shows that their estimate for a follow-on mission assumes $336M for the spacecraft itself, $288M for the launch vehicle, $117M for mission operations, and a 30% reserve. Removal of the launch vehicle and the reserve leaves a bare minimum of $906/kg of retrieved material.

This yields several conclusions of importance. The first is that simply retrieving asteroids and returning them to Earth to be processed is unlikely to be profitable with current space technology irrespective of launch cost. The value of an asteroid depends on its type, as material compositions vary widely, but none are worth $906/kg as a whole delivered on Earth. Some of their component elements are worth much more, such as the platinum group metals, but any scheme to recover them must rely on refining in place.

The second conclusion is that economic recovery of bulk materials from asteroids requires a leap forward in the cost-effectiveness of space hardware. The above $906/kg of recovered material assumes that launch costs are zero. Reusing the platform a great many times can reduce this cost, but it is limited by travel time. Even if 1 load of material could be recovered a year, a 100-year life span would reduce the cost to $9.06 a kilogram, assuming no additional ROM costs on the in-space hardware. Given that iron ore costs reached a high of $116 per ton in 2012, ²⁹ or 12.8 cents/kg, it is clear that a significant reduction in spacecraft cost, or some alternate approach such as a mass driver, is required for asteroid mining of bulk materials to become practical.

The corollary to this, however, is that assuming sufficiently light mining and refining equipment were available, the threshold for asteroid mining of some of the more valuable asteroid materials has already been met. Gold ($41.1k/kg) and the platinum group metals ($14.9k/kg) ²⁹ are both valued far in excess of the $2,000/kg estimated recovery cost and could be returned economically today, if a suitable apparatus for mining the ore and extracting the metal could be developed and deployed.

Conclusions

Exciting developments are under way in the space industry. The advent of reusable launch vehicles opens vast new possibilities for the expansion of human activity in space. However, launch cost is still a formidable obstacle to many proposed economic activities in space. As yet, communication relays and commercial imaging have been the only significant commercial uses of space.

The advent of reusable launch vehicles provides a solution to the launch cost problem. While they cannot reduce the cost of space travel to that of terrestrial travel, they are capable of as much as a 100-fold reduction in the current launch cost. This will drastically change the current environment, where only a few space economic activities are feasible, to 1 where many sorts are possible, including some that require significant manned presence.

Of the activities examined in this study, current space launch technology achieves the launch cost threshold for mining of precious metals, assuming that the attendant technical challenges remarked in this study are readily solvable. The question of a threshold for space tourism is more complex, but current technology has crossed it for sufficiently rich individuals, and with sufficient reductions in ROM and fixed launch costs space tourism will be open to the upper middle class. Moderate reductions in ROM cost enable current launch vehicles to cross the threshold for space-based solar power to become competitive with terrestrial solar power. Telecommunication has been and continues to be a viable space-based activity, even at today's launch costs.