A few interviews with two launch service providers, one active (Rocket Lab) and one aspirational (Astra), highlighted a supply chain trend that should have been obvious but wasn’t (to me). The space supply chain isn’t something I have focused on. Considering some of the previous analyses I’ve written about satellite manufacturers, the statements made during those interviews require a short logical step and make sense.
Both interviewees noted the reasons why their companies were building rockets their way. One particular reason, a supply chain that can’t handle their manufacturing requirements (specifically their speeds), was the common element in their interviews. In an Astra introduction interview, the answer below was provided to the question, “What are you doing differently about supply sources?”:
“Astra is currently identifying next-generation suppliers. Long lead times stifle innovation by limiting the original equipment manufacturers’ ability to release new products quickly. By applying strategies from the automotive and consumer electronics playbooks, we’ll be able to significantly increase the pace of innovation.”
Rocket Lab’s CEO, Peter Beck, provided the following answer during a Via Satellite interview:
It’s certainly early stages. What I would say is that our focus for the last couple of years has been on building the infrastructure to be able to build infrastructure on orbit. Standing up the Space Systems division, and ensuring that we have a good supplier of spacecraft components. One of the challenges that we found when we stood up the spacecraft division was the fragility of the space supply chain. There are lots of amazing companies building tens or hundreds of things a year with really long lead times. At Rocket Lab, we’re highly vertically integrated. If we can’t get it in nine weeks, we’ll just go and build it ourselves.
What we’re trying to do with the Space Systems group is make sure that we have the full selection of components — everything we need to be able to build anything we need, for either ourselves or our customer.
Note, these reasons from Astra and Rocket Lab are also why SpaceX builds most of its rocket and satellite parts. Planet builds its cubesats. In other words, they are increasing capital expenditures for their focused businesses. For very logical reasons, the aspirational or successful and operational companies in the new generation of space companies have decided to go all-in for building their rockets and satellites, using only external suppliers if they can keep up and build quality parts. And, as you’ll see in at least one example, some external suppliers can keep up.
This trend raises a few questions:
- Is it desirable for the space industry?
- What are the trend’s alternatives?
- Which of those alternatives is sustainable?
Is it Desirable for the Space Industry?
Companies such as Rocket Lab and Astra respond this way because legacy suppliers are used to production schedules requiring years, building quantities, as Beck noted, in the “tens or hundreds of things a year...”. For companies intending to launch much more than that, the “business as usual” rate of legacy suppliers is far too slow. Both Rocket Lab and Astra want to pump out more rockets than was usual, which means it’s also desirable to have a supply that can match that output--in this case, by building it themselves. Is the act of building parts in-house desirable for the industry?
Frustratingly, the answer is yes...and no.
Yes, because the old way needs to change. Suppliers used to legacy satellite and rocket manufacturers get paid well to take their time building their products. Nearly ten years ago (and during the decades before then), that pacing was just how it was in the space industry and accepted without question. The industry moved glacially, with no one company moving notably faster than the other. Despite public hand-wringing and multiple inspector general and government accountability reports, the industry never moved faster. It didn’t need to. And the companies that moved slowly back then still build that way. But that needs to stop, and it will take the newer generation of companies to do that since the other incentivizer, the customer, will not.
A domino effect drives the pacing, ultimately started with government customers (military AND civil). The government awards contracts to space industry manufacturing companies while rarely enforcing punitive options for schedule delays, cost increases, etc. (actually rewarding that behavior by continuing those programs). Under the general officer-driven mandate of “make it happen,” flexible ethics because of career concerns, coupled with an unwillingness to rock the space industry “boat,” means everyone gets paid for this lackadaisical pace while the mission (eventually) gets accomplished. So, why go through all the pain of changing and moving faster--especially for companies who won’t pay them as much (as the government does)?
These processes are so ingrained and accepted within government customers that NASA recognized the issue and subsequently created a middleman to speed the processes up. The middleman, the Rapid Space Development Office (RSDO), helps government customers get “...fast and flexible procurement of spacecraft and spacecraft components for future missions.” With much due diligence and hustle, the RSDO can help a customer obtain a satellite (a small one) in as little as...24 months? The office has an updated catalog of pre-vetted satellite manufacturers, some of who advertise longer satellite manufacturing times. So much for rapid.
NASA isn’t the only one who attempted to “shorten” the process. The DoD’s Operationally Responsive Space program was the military’s attempt to work around lengthy manufacturing times. The Space Development Agency today has similar mandates (on a more extensive scope).
The RSDO’s existence highlights the problem of accepting the tortoise’s speed within the older parts of the space sector. That industry is so slow-moving (with the justification that slow=better mission assurance/quality) that two years to build a satellite is considered “rapid.” This middling result of a government middleman dedicated to the concept of rapidity makes it easy to understand why newer companies are building components in-house and why the initial answer to industry desirability is “yes.”
Ecosystem Concerns (or, why No)
But there is also a good reason to answer “no” to the question of this trend’s desirability. If a larger, more robust, more responsive space supplier ecosystem is desired, then vertically integrating everything does the opposite long-term, potentially stunting industry growth. Certainly, closed supply systems may be more dependable and more responsive to a company’s needs, but such systems tend to narrow down customer choices. Worse, if a vertically-integrated company like Astra suffered an unrecoverable business failure, its supply line would disappear. Is that desirable for a market everyone wants to see grow?
There are alternatives—a very notable exception to the new generation of space companies building their supply chains: OneWeb.
The rapid supply chain and relationships OneWeb fostered for fast satellite production is a different approach to how most new-generation space companies deal with suppliers. It is a point of pride for that company--even if there are quite a few Airbus-tied suppliers involved. That pride is well-earned, considering the speeds at which OneWeb/Airbus manufactures their satellites. Spreading the ability to manufacture quickly among suppliers instead of building components in-house appears like the healthier space industry approach with potentially far-reaching, positive impacts (unless those suppliers focus only on their customer’s needs).
However, OneWeb’s bankruptcy last year exposed the weakness inherent in rapid suppliers catering primarily to a company like OneWeb: dependence. If the company disappears or pauses, that company’s suppliers risk suffering the same fate. Instead of building a healthy, fast-reaction supply chain, the ecosystem could collapse if it’s too dependent on a single customer. The fact that OneWeb continues pumping out satellites after its bankruptcy displays some supplier ecosystem flexibility. Perhaps OneWeb is growing a healthy and responsive supplier ecosystem.
All the above describe the trend and its alternatives:
- The old way, requiring extraordinary time and money but very inclusive of suppliers.
- The “new generation of space” way, building almost everything in-house, speeding up satellite manufacturing time from years to a day.
- OneWeb’s way, a hybrid of supplier inclusivity and new generation rapidity.
To be clear, OneWeb’s way of building up suppliers is the healthiest and most desirable option for those wishing for a growing space economy. The ecosystem of suppliers is sustainable and not unusual in other industries.
Why Space is Special (AKA, the “old” Way)
The reasons for the specialness of the space industry worldwide are typically associated with advancement, both in exploration and science. While those noble and education-focused words are bandied about all the time, I would suggest that the space industry is unique for another reason: it does things the complex and expensive way (generally with the justification that space is unique). Other, more mature industries, including the tech sector, show it doesn’t have to be complicated and expensive. We see this, ironically, with companies like Rocket Lab, Planet, and SpaceX.
Tesla aside, most automobile manufacturers have conducted their manufacturing “OneWeb’s Way” for decades. That industry is one of the most competitive on the planet, producing immense varieties of their goods. The computer hardware industry works similarly. The growth and success in such industries, in which suppliers can keep up with manufacturing rates in industries producing millions annually, indicates that OneWeb’s supplier ecosystem can work--so long as the suppliers aren’t dependent on OneWeb for their fortunes. When the newer space companies take component manufacturing in-house, those external suppliers may not be able to wean away from dependence.
The old way of satellite manufacturing does not seem sustainable by modern industrial standards even though it’s gone on for decades. Nor is it desirable to those who want to see a growing space industry. In many ways, legacy manufacturers assemble their satellites with processes resembling the way horse-drawn coaches were manufactured for royalty (sustained for hundreds of years). As the government does with satellites, the royalty paid well for a durable and comfortable coach.
Indeed, the coach parts were of the highest quality from the most reputable suppliers of the land. Those parts were assembled in well-lit sheds by specialists with care to meet customer expectations, but the processes and the mindset for building those coaches never resembled mass production.
The same could be said of supplier and manufacturing processes for a Viasat-3, Galileo, or JPSS--they are high-quality, but low-production. With the old way, whether it’s the RSDO, ORS, or SDA, each throws money at the same coachbuilders (there aren’t many) to use those same slow processes, expecting faster results.
The trend of companies like Astra, Rocket Lab, and SpaceX building up internal supply chains is their response to that low-production rate with high costs. The logic guiding their decisions addresses the circumstances today, but (based on history in other industries) their internal supply chains aren’t inevitable, and won’t remain internal forever. If there’s a market (for another analysis--who is that market now?), external suppliers will choose to compete and provide better products in shorter time frames. Rocket Lab and Astra will need to respond, which could help make the supplier space more vibrant and competitive:
- Splitting off their own supply chains to become space suppliers themselves (voluntarily or coerced through government anti-monopoly actions).
- Investing in and eating up suppliers (a la GM and Delphi) as requirements exceed expertise.
Like Apple with its products, SpaceX may keep up a “closed garden” approach to its manufacturing. It has publicly stated other incentives than just profit for building better rockets and satellites. And, as with Apple, such an approach may be sustainable for SpaceX.
The trend is a bridge for other space industry players, which leads to an increasingly diverse, responsive, and reasonably-priced supply chain. We’ve seen it in the other industries. OneWeb has crossed that bridge with its suppliers. Rocket Lab, Planet, and the others are in the middle. And most of the older industry suppliers are on the other side, understanding they should cross but reticent to do so.