Post Your Comment
Ahead of Elon Musk's India visit, a deep dive into the global space market
The founder of SpaceX, the world's biggest space launch company, is expected to meet India's budding spacetech entrepreneurs in Delhi. He's also likely to announce plans for his EV company Tesla in the country
This handout photo taken and released by Indian Space Research Organisation (ISRO) on January 1, 2024, shows the lifting-off the PSLV-C58 rocket carrying the X-Ray Polarimeter Satellite (XPoSat) from the Satish Dhawan Space Centre in Sriharikota. Image: Indian Space Research Organisation (ISRO) / AFP
Related stories
To better answer these questions, I analysed the industry’s current state of demand (satellite constellation manufacturers/operators) and supply (launch vehicle manufacturers/service providers). Whether you're a techno-optimist, a curious reader or a potential founder/investor, understanding the spacetech market offers valuable insights into our future, both on Earth and in space.
Part I: The Demand
The current state of demand
Data from the UCS satellite database, as of May 1, 2023, contains details on 7,560 satellites currently orbiting Earth.
There are three major categories of orbits where most satellites are in operation today: Low-earth orbit (LEO), medium-earth orbit (MEO) and geostationary orbit (GEO). At a high level, the characteristics of satellites in each of these orbital classes are different as summarised below:
Overall, LEO satellites tend to be the most common type of satellites in orbit today accounting for over 90 percent of all satellites. Due to their proximity to Earth, LEO satellites offer the lowest latency and highest bandwidth for communications while requiring reduced transmission power.
This means LEO satellites can be made considerably smaller and more compact compared to higher-orbit satellites. The median mass of LEO satellites in operation today is 260 kg compared to 800 kg and 4190.5 kg for MEO and GEO satellites, respectively. As a result, they are less expensive to manufacture and launch compared to higher-Earth orbit satellites, lowering the barrier to entry versus MEO and GEO satellites.
However, satellites in LEO travel at an orbital speed of ~7.8 km/s and take less than ~2 hours to orbit the Earth. Due to this, they have a small field of view and can observe/communicate with a fixed section of the Earth for only a brief moment in time. So, a network (or "constellation") of LEO satellites is required to provide continuous coverage.
In addition, satellites in lower regions of LEO also suffer from fast orbital decay and have shorter operational lifetimes. The median lifetime of LEO satellites is four years versus 10 and 15 years for MEO and GEO satellites, respectively. Therefore, LEO satellites require either periodic re-boosting or replacement to continue their operation, and a plan to de-orbit to prevent the proliferation of space junk.
Key applications of satellites in orbit
Most satellites in orbit are launched for communication purposes, primarily for commercial end-users. The second most popular application is in Earth observation. The unique features of each orbital class inform what applications the satellites are best suited for:
Satellites in LEO offer the lowest latency connectivity and highest resolution imagery, making them suitable for communications and Earth observation.
Satellites in MEO offer a balance between coverage area and latency, making them suitable for navigation systems like the Global Positioning System (GPS). Also read: Moving beyond launch: Realizing the benefits of the new space economy
Satellites in GEO, due to their orbit being in sync with the Earth’s rotation, are especially suitable for applications that serve a broad-but-fixed surface area, like television broadcasting, long-distance telecommunication and weather observation.
In the last five years, there has been an explosion of LEO satellites, driven by constellations for communications. This trend is expected to continue to grow, with players like SpaceX’s Starlink and Amazon’s Project Kuiper planning LEO mega constellations.
The satellite market and key players
The market for LEO satellite constellations for communications is extremely crowded and dominated by Starlink. With over 5,289 operational LEO satellites as of January, Starlink now accounts for over 50 percent of all active satellites. In addition, SpaceX is fully vertically integrated launching Starlink on their Falcon 9 rockets out-competing everyone else. More on this in the supply side of the equation.
Furthermore, SpaceX recently launched Starlink satellites with direct-to-cell capabilities essentially turning them into cellphone towers in space, a significant evolution from their existing satellite-to-dish services.
Following Starlink is Airbus OneWeb Satellites, a collaboration between Airbus and Eutelsat OneWeb, which to date has 634 operational LEO satellites. Another massive initiative is Amazon’s Project Kuiper which plans to launch 3,236 LEO satellites for which 83 commercial launches have already been secured.
Demand implications: White spaces and the future
Given the competitiveness in LEO-based connectivity, I list what I believe to be the white spaces and the future of demand in the coming years.
Demand white spaces
The future of space looks crowded, and satellite operators will have to invest in avoiding collisions and mitigating the build-up of space debris and junk.
As discussed previously, LEO-based connectivity is such a crowded space and comes with the inherent challenges of launching and maintaining large constellations. This creates a white space in GEO-based connectivity using single satellites to deliver communication and internet services to markets that LEO satellites do not serve.
A notable example of a company in this category is Astranis which operates its communication satellites in GEO instead of LEO. This allows their satellites to maintain a fixed position relative to the Earth, providing continuous service to a specific geographical area. Furthermore, fewer satellites are required due to their GEO positioning, which eliminates the need for a large constellation, constant tracking and complex satellite operations.
GEO-based connectivity where unserved by LEO satellites
From a technological standpoint, Astranis’s microGEO satellites are smaller and more cost-effective than traditional GEO satellites. These satellites are about the size of a dishwasher, whereas traditional GEO satellites are much larger, around the size of a double-decker bus.
From a business model standpoint, Astranis is targeting specific markets with dedicated satellites. For example, it has inked deals to launch and operate dedicated satellites for regions like Alaska and the Philippines, aiming to provide enough bandwidth to connect a significant number of people in those areas as opposed to a global coverage approach adopted by Starlink.
The Philippines is uniquely suited for such connectivity being an Archipelago of over 7,000 islands that are extremely hard to cover with traditional connectivity such as fibre. Astranis hopes to deliver this connectivity to the entire nation through just two microGEO satellites.
Going beyond communications, satellite companies can develop technology for the next biggest application in Earth observation for diverse industries like defence and intelligence, maritime tracking, insurance, disaster response, mining, agriculture and infrastructure monitoring.
Earth observation applications for LEO satellites
Players can differentiate themselves by adopting different imaging modalities used for Earth observation like optical, hyperspectral or Synthetic Aperture Radar (SAR) imaging.
One company in this category is Capella Space which operates LEO satellites for 24-hour all-weather SAR imagery at spatial resolutions as low as 0.5 metres. This is unique because SAR can penetrate clouds, fog, smog, darkness and smoke, providing reliable and high-quality imagery in all weather conditions.
The company monetises this data through analytics that highlights items of interest and changes over time (example, shipping vessels and their movements) through web-based dashboards, API integration to client workflows, and satellite-tasking software.
Differentiation can also be achieved through Earth observation delivered from different orbital altitudes. For example, operating at very low Earth orbit (VLEO) between 250 km and 450 km instead of LEO allows capturing much higher-resolution imagery due to the closer proximity to Earth.
VLEO satellites can be made smaller and at lower costs due to the absence of radiation-hardened components and bulky optics that are required at higher orbits. However, operating at VLEO presents engineering and control system challenges which, when overcome, are significant technology moats.
Most notably, there is significantly higher atmospheric drag experienced at VLEO compared to LEO and beyond requiring multiple thrusts per orbit to counteract. In addition, given VLEO orbital speeds, image stability and accuracy are also significant challenges.
A company attempting to do this is Albedo Space whose technology focuses on capturing both optical and thermal images from VLEO. Their satellites are designed to capture optical images with a native resolution of 0.1 m similar to that obtained by reconnaissance aircraft.
This is complemented with thermal imagery at a resolution of 2 m. Albedo plans to have its first commercial satellite in 2025, followed by six more, and ultimately a full constellation of 24 satellites. This would allow for five revisits per day, providing high-frequency coverage.
Next-gen LEO applications: Space manufacturing
The ability to produce and launch low-cost satellites unlocks next-gen LEO-based applications like space manufacturing in micro-/zero-gravity environments. The microgravity environment offers unique conditions that can be leveraged to create products with properties that are difficult or impossible to achieve on Earth.
Several products such as semiconductors, optical fibres, protein crystals, metal alloys, ceramics, pharmaceuticals, biological tissues and organs can theoretically be synthesised defect-free in microgravity. The unique conditions in microgravity, such as the absence of sedimentation, buoyancy, and natural convection, enable these manufacturing possibilities.Also read: Nasa chief's visit spotlights India's space tech potential
Microgravity also allows for container-less processing, which can provide an ultra-pure, contaminant-free environment for manufacturing or study of materials in their molten state.
A company in this category is Varda Space Systems which is pioneering the field of microgravity manufacturing by designing and operating spacecraft that manufacture materials in the microgravity environment of space.
Their approach involves using a three-piece vehicle system, which includes a spacecraft, a manufacturing module and a heatshield-protected capsule that can re-enter through the Earth's atmosphere and land safely back on Earth. The company's technology is applied primarily for the production of pharmaceuticals.
Their first mission, which included a 27-hour drug- manufacturing experiment in orbit, successfully grew crystals of ritonavir, a drug commonly used to treat HIV. The experiment demonstrated the feasibility of orbital drug processing outside of a government-run space station, marking a significant step in commercialising microgravity and building an industrial park in LEO.
The near-Earth space is already quite crowded. This, coupled with the fact that the majority of satellites being launched are to LEO with short lifespans in expendable launch vehicles, is a strong signal that the build-up of space debris and junk is of serious concern.
Satellite operators will be required to retire their satellites at end-of-life to mitigate the build-up of space debris and junk. Methods such as active propulsion systems to avoid collisions, planned de-orbit (from LEO), and demise during re-entry have been proposed in mitigation plans from major players such as SpaceX and Amazon.
Furthermore, there will be opportunities for third-party orbital debris removal companies such as Astroscale to offer end-of-life, active debris removal and life extension services.
Part II: The Supply
The current state of supply
The supply side of the ecosystem includes launch vehicle manufacturers and service providers. Much like the demand, in this section, I first look at the current state of supply. Next, I evaluate implications by identifying white spaces and map the possible future of supply.
SpaceX is the unequivocal leader in the launch industry and the numbers are staggering. As per the UCS satellite database, among the top 10 launch vehicles recorded since 2000, SpaceX’s Falcon 9 launched 3.5x more satellites than the next nine launch vehicles combined, 99 percent of which were small satellites launched to LEO.
In addition, 84 percent of these satellites were their own Starlink satellites which, as we discussed previously, now dominate the proportion of satellites being launched to LEO. This number is only going to grow with SpaceX planning to launch and operate a mega-constellation of as many as 42,000 Starlink satellites in the future.
Falcon 9 rockets fly rather full. Of the 91 Falcon 9 launches in 2023, 69 percent (63 launches) exclusively carried Starlink satellites. Furthermore, 85 of the 91 launches had documented payload mass information. The median payload mass was 16,800 kg which corresponds to 91 percent of the total LEO payload capacity of the reusable version of the Falcon 9 (18,400 kg).
This coupled with SpaceX’s plans for a Starlink mega-constellation and uncertainty around when Starship becomes operation means that securing allocation on the most frequently operated launch vehicle will be difficult for the next few years.
Who might close the gap
Following SpaceX, there’s a long tail of launch companies vying to capture market share. Top private companies in the West include United Launch Alliance (ULA), Arianespace, Rocket Lab, Blue Origin and Relativity Space.
ULA is a joint venture between Lockheed Martin and Boeing, and it's one of the most experienced launch companies in the US, with over 155 consecutive launches and a 100 percent mission success rate. The company both manufactures and launches its vehicles, including the Atlas V and the Delta IV Heavy, which will both be retired in the coming years.
In January, ULA launched the first certification flight of their latest heavy-lift launch vehicle, Vulcan Centaur.
Arianespace is the world's first commercial launch services company. It operates a full family of launchers, including Ariane 5, Soyuz and Vega.
ArianeGroup, a joint venture between Airbus and Safran is the lead contractor for the design and production of the Ariane rocket series. Avio is the lead contractor for the Vega rocket series. Arianespace has placed more than 550 satellites in orbit and serves over 96 customers from the public and private sectors.
In June, Arianespace plans the first launch of their next-generation medium-heavy lift launch vehicle, Ariane 6.
Rocket Lab is a launch vehicle manufacturer and launch service provider that operates and launches lightweight Electron orbital rockets—the second-most frequently launched US rocket delivering 172 private and public-sector satellites to orbit. In 2024, Rocket Lab plans the first launch of its medium-lift launch vehicle, Neutron.
Blue Origin is a major potential competitor in the launch services industry. The company's New Shepard suborbital vehicle has conducted several crewed and uncrewed flights. In addition, it makes rocket engines for ULA’s Vulcan Centaur. Blue Origin is also developing the New Glenn, a reusable heavy-lift launch vehicle, which is expected to significantly increase the company's presence in the satellite launch market.
Relativity Space is a launch vehicle manufacturer and launch service provider that developed the Terran 1, which was the world's first 3D-printed rocket to reach space in March 2023. After its first and only launch, the company retired the Terran 1 and is focusing on its medium-heavy lift 3D printed launch vehicle, Terran R.
What differentiates these companies from the rest of SpaceX’s competitors is that they’re developing or operating medium-/heavy payload (>10,000 kg to LEO) launch vehicles to directly compete with SpaceX’s workhorse Falcon 9. This differentiation has not gone unrecognised with the largest sources of demand flocking towards these specific launch operators.
For example, Amazon’s Project Kuiper secured up to 83 launches from ULA, Arianespace and Blue Origin, providing enough capacity to launch the majority of their constellation.
Outside of these companies, there are numerous small payload launch companies. To help facilitate comparisons between these various companies, based on publicly available data, I’ve considered the relationship between the estimated cost per kg of payload to LEO and the total payload to LEO for various launch vehicles operational today, recently retired, or planned within the next few years.
The cost per kg of payload to LEO is simply the ratio of the estimated total cost of launch and the total payload to LEO. However, it is to be noted that not only does the cost of launch remain fixed nor does the payload capacity always get filled.
Regardless, these assumptions help directionally inform trends that show how competitive the supply side of the market truly is. In addition to all the private launch companies, I’ve also considered a few leading state-owned launch manufacturers and service providers like the Indian Space Research Organization (ISRO) and China Aerospace Science and Technology Corporation (CASC).
As per the UCS satellite database, ISRO’s PSLV and CASC’s Long March series launch vehicles feature in the top 10 launch vehicles by number of satellites launched.
Why others will struggle: Economies of scale and the sweet spot
The cost per kg of payload to LEO decreases exponentially with increasing payload capacity most noticeably as we reach 10,000 kg total payload. Economies of scale come into play as the fixed costs associated with a launch (such as launch pad operations, mission control, and vehicle production) are spread over a larger payload, reducing the cost per kg.
The larger the payload capacity of the rocket, the more these fixed costs can be distributed, leading to a lower cost per kg. The nature of rocket propulsion also contributes to this trend. We know that adding more propellants to a rocket results in a less-than-proportional increase in its payload capacity to LEO (that is, diminishing returns).
In other words, more and more propellants are required to carry the same unit increase in payload capacity with increasing payload. However, as described previously, larger rockets with more fuel can through brute force simply carry more payload to LEO and better spread fixed costs associated with the launch, reducing the cost per kg.
It is to be noted that the wet mass is almost 90 percent propellant mass mostly used to generate sufficient thrust in the first stage of the rocket to overcome Earth’s gravitational pull. So even obtaining a 5 percent LEO payload fraction is extremely challenging.
The LEO payload fraction increases somewhat meaningfully with payload to LEO up to the current performance characteristics of Falcon 9 before diminishing returns are observed with increasing payload capacity as we move towards the performance characteristics of Starship.
This essentially indicates that there’s a sweet spot between ~10,000 and 40,000 kg payload to LEO where the LEO payload fraction continues to increase somewhat meaningfully (that is, the launch vehicles are increasingly efficient) and the cost per kg of payload to LEO falls below ~$5,000. Beyond 40,000 kg payloads, more and more propellants are required to carry the same unit increase in payload resulting in diminishing returns.
Furthermore, larger payload rockets may be harder to fill requiring them to fly less frequently or partially full resulting in higher costs per key of payload.
However, due to the technical complexity, cost and ground infrastructure requirements, only a few Western players outside of SpaceX have been able to move into this sweet spot, most notably ULA with Delta IV Heavy (operational) & Vulcan (planned), Arianespace with Ariane 5 ECA (operational) & Ariane 64 (planned), Rocket Lab with Neutron (planned), Relativity Space with Terran R (planned), and Blue Origin with New Glenn (planned). The vast majority of players are still planning small-lift launch vehicles.
Supply Implications: White spaces and future
Given SpaceX’s dominance and the technological/operational challenges to build launch vehicles in the 10,000 to 40,000 kg LEO payload sweet spot, how can the myriad small-lift launch players differentiate themselves? Short of building larger launch vehicles, I list what I believe to be the white spaces and the future of supply in the coming years.
Supply white spaces
Differentiation through technology
Although likely the hardest way to differentiate, doing so through technological advances that meaningfully impact demand can build the most lasting moat. Ways to drive such impact would be to bring down the launch cost while still making the unit economics work and dramatically increasing the launch cadence allowing customers to iterate effortlessly.
A company that truly stands out in this category is Stoke Space which is developing a fully reusable rocket. This is a significant departure from traditional two-stage rockets, where typically only the first stage is reusable at best. While the first stage accounts for 60 to 70 percent of the production cost and is the natural place to start with reusability, without a fully reusable rocket, the only way to drive up launch cadence is to increase production throughput.
By developing a 100 percent reusable rocket that can be rapidly refurbished, Stoke Space aims to fully amortise the entire production cost of the rocket throughout its lifetime while maintaining a high launch cadence.
The company's unique second-stage engine features a distributed thruster system with an integrated, actively and regeneratively cooled heat shield. This means that the heat shield uses the rocket's fuel to absorb and dissipate the heat generated during re-entry.
This design allows the second stage to return to Earth somewhat like a space capsule, base first, with the regeneratively cooled heat shield protecting the vehicle from the intense heat of re-entry. Traditional cooling methods used on re-entry capsules like heat-resistant tiles require significant time and man-hours to refurbish between launches driving up launch costs and lowering launch cadence.
Stoke Space, through its novel cooling method for re-entry, aims to fly daily with minimum refurbishments. In September 2023, the company successfully demonstrated vertical take off to an altitude of 30 feet and landing at a planned landing zone of a fully reusable second stage.
Differentiation through the business model
Smaller launch vehicle companies are carving out their niches with innovative business models. One way to differentiate from other small-lift launch players is to emulate SpaceX and vertically integrate. Rocket Lab exemplifies this vertical integration, by offering a turnkey solution from manufacturing to launch, and ground support to on-orbit operations.
Rocket Lab is undoubtedly a major competitor in the launch market with the successes of the Electron and their expected launch of the Neutron. However, the company's business model goes far beyond that. First, Rocket Lab owns and operates the world’s only private orbital launch site in New Zealand providing 120 launch opportunities annually to LEO.
Second, the company designs and manufactures its own satellites, using many of its own subcomponents, including solar panels, star trackers, reaction wheels and avionics. Their Photon platform is a configurable satellite bus that can be tailored to meet the specific needs of a mission.
Third, Rocket Lab has partnered with Kongsberg Satellite Services (KSAT), the world’s largest provider of ground station services, to provide ground segment support for the Electron launch vehicle and Photon satellite bus. Finally, the company has also been contracted to operate its satellites in orbit on behalf of the US government.
Vertically integrating is no easy feat. A few small-lift launch companies are differentiating in other ways. One such way is to offer on-demand launch services like Skyrora which is positioning itself as a space-bound taxi service for customers with small payloads that prefer to launch privately.
This contrasts SpaceX’s rideshare programme which is more like a bus service that operates on fixed dates and currently has a ~1-year lead time to LEO. Skyrora’s service will likely cost 3x that of SpaceX’s rideshare programme, but they look to serve a niche segment.
Differentiation through geographic presence
Another way to differentiate is by providing strong and reliable non-US-based launch service alternatives.
These alternatives will require the support of leading non-US national space agencies. With geopolitical tensions engulfing Russia and China, India is truly this alternative with its world-renowned national space agency—ISRO. For context, ISRO has several stellar accomplishments to its name. Its family of four launch vehicles has launched over 342 foreign satellites from 34 different countries.
Furthermore, as per the UCS satellite database, ISRO’s medium-lift launch vehicle Polar Satellite Launch Vehicle (PSLV) is third on the list of top 10 launch vehicles by number of satellites launched since 2000. Beyond the launch market, ISRO's notable missions include the Mars Orbiter Mission (MOM), which made India the first nation to reach Mars on its first attempt and the fourth space agency to reach Mars orbit.
The Chandrayaan missions, with Chandrayaan-3 successfully landing on the moon, made India one of the four nations to achieve a soft lunar landing. In June 2020, ISRO established the Indian National Space Promotion and Authorization Center (IN-SPACe) to allow private companies access to ISRO’s infrastructure and expertise.
Being associated with ISRO and building in India can serve as a tremendous differentiating factor for private small-lift launch companies offering non-US-based launch services. Skyroot Aerospace and Agnikul Cosmos are well-capitalised examples of such companies. It is to be noted that both companies are building small-lift launch vehicles to differentiate through technological advances.
Agnikul’s lightweight monolithic 3D-printed rocket engine Agnilet allows for the integration of all engine components during the build process, eliminating the need for bolts, screws or welds. This results in an extremely light engine, weighing between ~5 and 6 kg, compared to similar thrust engines that can weigh up to 25 kg. In addition, the company has developed a modified truck-based launch platform called Dhanush for their Agnibaan rocket providing flexibility in launch operations.
Skyroot Aerospace successfully launched India’s first privately made rocket in 2022 in partnership with ISRO. The company has also successfully tested a solid rocket propulsion stage made using lightweight carbon composite and its 3D-printed cryogenic engine, Dhawan-II. However, the icing on the cake for these two companies is their partnerships with ISRO which serve as a strong signal for demand. Both Agnikul Cosmos and Skyroot Aerospace have entered into an MoU with ISRO for access to ISRO facilities and expertise for the development of their launch vehicles.
Short-term shortfall, long-term surplus
In the near term, the space industry may face a shortfall in the supply of medium- and heavy-lift launch services. This is due to the retirement of many medium and heavy launch vehicles (for example, Atlas V, Delta IV Heavy, Ariane 5), and the fact that most remaining capacity is already booked. The bulk of the demand driven by large constellations is expected to be for medium and heavy launches.
A report by McKinsey & Co expects the annual demand to be 15,000 tonnes if all proposed satellite constellations are to be launched through 2030 which far exceeds supply. Even in the projected base case where less than half the proposed constellations are launched, the annual demand would be 4,500 tonnes which would require Falcon 9 to be launched ~2.5x its 2023 launch frequency.
However, new launch capacity from companies like ULA, Blue Origin, Arianespace and Rocket Lab may come online as soon as 2024 adding to the supply.
In the long term, the supply of launch services is expected to increase significantly. This is largely due to the potential capabilities of SpaceX's Starship, which could theoretically offer a launch a day by 2030. However, the balance between supply and demand will be influenced by a variety of factors, including technological advances, cost dynamics and the strategies of key players in the industry.
If SpaceX's Starship and other heavy and super-heavy launch vehicles come online, there could be an oversupply in the market. However, the actual outcome will depend on the evolution of demand, which is influenced by factors like the lifespan of satellites, their subsequent removal from orbit and the general growth of the private sector in space.
(The writer is a staff scientist at Samsung Semiconductor and an investor with the India-based venture builder platform GrowthStory. He holds a PhD in medical engineering from Caltech)
