WASHINGTON — Amazon and Blue Origin founder Jeff Bezos likes to refer to his multibillion-dollar Amazon stock portfolio as “lottery winnings” that he intends to invest in space technology. It’s Bob Smith’s job to put those winnings to good use.
Blue Origin, Smith says, ultimately must be a self-sustaining business. That’s why the company is a presence at largely commercial events like the recent Satellite 2018, where it signed an agreement with satellite fleet operator Sky Perfect JSat for a future satellite, and transformed a 2017 agreement with Thai startup mu Space into a firm contract for a 2020 launch on the New Glenn rocket.
It had previously concluded agreements with established geostationary-orbit fleet operator Eutelsat and startup broadband constellation network OneWeb.
New Glenn is a huge rocket, and Blue Origin in December took residence in its huge Florida factory.
With launches less than three years away, every year now is critical for Blue Origin. Key events in 2018 include an expected decision by United Launch Alliance on whether to use Blue Origin or Aerojet Rocketdyne engines for ULA’s future Vulcan rocket.
In an interview, Smith addressed what a positive ULA decision would mean for Blue Origin, why reusing New Glenn’s first stage makes business sense regardless of the number of launches per year, the virtues of vertical integration and the patience made possible by Bezos’s largesse.
United Launch Alliance is about to decide between Blue Origin’s BE-4 and Aerojet Rocketdyne’s AR1 engine for ULA’s next-generation rocket, called Vulcan. You appear to have the inside track. How would a ULA contract change things for you?
I’m very gratified that they’ve shown a lot of confidence in us to go select us, at least for the development discussion, and get to hopefully a production contract here very soon, because it helps us fundamentally. Being in front of a customer makes you a better customer. I think that’s always good.
They know how to do a number of things really, really well. They know how to deal with national security customers, with mission assurance customers. That’s fabulous. There’s a lot that we could learn from ULA just by being their engine provider.
Obviously the financial benefits of working with ULA and selling them engines is a good thing. The other component of it is it actually gives us additional volume and under high quality standards. Both of those give us commercial leverage with our suppliers where we can get a much better position in terms of what our commodity values are.
And larger lot buys help us. They [ULA] are actually running a much larger engine production, they would drive us to much higher engine production levels because our vehicle is completely reusable and the engines are rated for a hundred missions, so that’s a lot of missions for us. Times seven [New Glenn’s first stage is comprised of seven BE-4 engines]. If we have a couple first stages [from ULA], that’s 14 engines times 100. So that’s a lot of missions for us. They would drive us to other commercial volumes, which we would welcome.
ULA plans to recover some of their engine hardware from Vulcan flights.
They’re working on that.
ULA’s corporate parents, Lockheed Martin and Boeing, are being asked for a large financial investment in Vulcan. Parents sometimes say no.
I would say the parents’ decision process is strategic. What I mean by that is: How much do they really want to be in the launch industry? I don’t know.
But the analogy may not be entirely accurate because ULA is also going to its suppliers, or at least that’s what I’ve heard. I think they’re looking for that as well. I don’t’ know what it takes for them to close their business case. But at least from our interactions with ULA, they want to have a lot of engines and they are very bullish about their future.
Would a favorable ULA decision drive production facility sizing?
We went into a discussion with the state of Alabama for our production facility. It was conditioned on signing a production contract with ULA. If we do sign that, it will trigger the development of that facility. We may still utilize that facility for other reasons even if we don’t get it, because I think it may be a good place to go do that. There’s a lot of aerospace and certainly a lot of rocket capability there.
But that’s the real sizing question for us: What’s the production volume that we need for BE-4 engines? Largely that was going to be our place where we would get the full rates we need to support both our internal needs as well as what ULA needs.
So if ULA signs the production contract and later you find out the parents aren’t going to fund the investment…
We’ll be okay.
How is the New Glenn production plant in Florida coming along?
The facility is stunning. It’s 625,000 square feet [58,064 square meters]. You hear these numbers and it’s hard to get your head around them. So it’s always very compelling to go see this. Think about just the size of our vehicle: The shuttle stack was about 180 feet, our vehicle is well over 300 feet. It’s a big vehicle.
The rocket factory is going very well. We’re now putting in our tooling and equipment. We’ve made our first flight part for New Glenn, which is exciting to have, and our launch pad continues to get developed, which is also a heroic effort. It is amazing how much concrete we’re actually putting into that entire place.
When will it be operational?
We took occupancy last December. We’ve got people in it and now we’re ordering tooling that will be put in. Throughout the year we’re going to be giving updates about when are we getting more tooling and other things in the facility.
And final completion?
It depends on how you define completion. The hard one for us was to take occupancy in 2017, which we did, in December. That was our big thing. That way we could actually make sure we have the timing right for all the tooling to come in.
You are master of that calendar now?
Yes. It’s not a matter of building construction now, it’s really about equipment installation and hiring and doing the usual manufacturing demonstrators that would allow you to have confidence that you’ve got the process control, the process verified, and that it can be done at scale.
Many of the things we’ve already done up in Seattle. So there’s a lot that I’m not concerned about. Others need some time and effort and we’re going to have to go work through. For example, we have these very large 7-meter fairings that are 70 ft [21.3 meters] long. That means getting that done repeatedly, cheaply, is a good piece of engineering manufacturing work.
You’re building the fairing.
Could you talk about make-or-buy tradeoffs?
The make/buy process for us is largely vertically integrated. As long as we can withstand the nonrecurring [costs] and it doesn’t take us too long, and the time to market is not prohibitive for us to develop the actual system we’re talking about, we’d rather have that.
What we have then is a tight coupling between our engineering team and our manufacturing and test team, and we can actually make exactly what it wants.
If you’re not doing this on a vertical integration stack, you pay for some portion of that nonrecurring engineering, or you’d be paying something that’s largely already developed for some other program, or at best you’re going to do a build-to-print with someone.
Build-to-print generally means you have to believe that their systems, tools and capabilities are somehow better-leveraged, or that they have smarter people than we have. And at this point we haven’t found many examples of this. I can’t even think of a very large example where that’s the case. Our labor rates are competitive, obviously our capital is something that we don’t have a problem with, so vertically integrated makes a lot more sense for us.
Aren’t you giving up future flexibility if the market isn’t what it’s predicted to be?
You’re saying the fixed assets are there and therefore you’d have an absorption problem associated with those fixed assets where you wouldn’t if you’d outsourced. That’s a valid point.
You’re comfortable with that tradeoff?
Yes, we are. The reason that we are comfortable — and the thing that I’ve talked a lot about within Blue — is that we’re very very fortunate in the sense that we have great people, we have demonstrated technologies, and we’ve got patience. We were discussing market fluctuations earlier. We’re going to be very patient. We’re going to be able to go through a lot of those issues and be very committed to it.
Reusability leads to a reduced production volume, sacrificing scale economies, which is OK if you have a certain annual volume. How many launches per year makes reusability a clear winner?
The thing that I keep trying to get the industry to recalibrate to is that there are two paths here, and one business model has a very different model than the other.
If you’re an expendable launch vehicle or even a moderately reusable launch vehicle, your method to winning in the marketplace is to buy good, reliable, capable systems and getting the unit product cost down for each one of those. That means you’re generally dealing with some amount of fixed asset base for tooling and property and equipment. But most of it is tied up into your touch labor, your materials and what your production standard is for that vehicle. What you want to do is drive that down, and you want a lot of volume from the standpoint of expendable launchers.
For a reusable launcher, it’s an asset realization model, which is a very different business model. It is the difference between discrete manufacturing and continuous manufacturing. In discrete manufacturing you’re dealing with that unit product cost. On a reusable, it’s actually all asset reutilization. So I want to sweat that asset as much as I can.
Even if you’ve got fixed investment that’s not being used to the maximum extent?
It’s much more like why Southwest Airlines invests so much time and energy on turn times. They will fly a less-than-full plane, but as long as that passenger is flying, they are generating revenue.
It’s the same business model that we have. With a New Glenn vehicle, I have high reusability on that first stage and I am going to make different decisions about what we fly on top, which has a lot of advantages for us. So we can say: We’ll do this one, which is very nice from an overall utilization standpoint, and in those terms it looks very good. This one’s not as good, but that’s OK, I’m still beating the depreciation that I had from investing in this stage.
The market for large GEO-orbit satellites has been down for two years and may never return to previous levels. Has that changed your view of the future?
I would emphasize again our patience in this area. We are not looking at the kinds of time horizons that I think are very typical. That’s one of the powers of what we’re doing at Blue. I think the investments we’re making are large and sustainable. And we’re not varying off on a different paths.
We built New Shepard [the company’s crewed, suborbital vehicle and capsule] so we can build New Glenn. We built a hydrogen-oxygen engine on New Shepard and learned how to master a lot of reusable capability from an operations standpoint so we can actually apply those to New Glenn. We built BE-3 so we that know how to go build BE-4.
There are corporate sayings — Quality Is Job One for example — that are just nice. But if you look literally at what we do, it really is this step-by-step approach [the company’s corporate motto, Gradatim Ferociter, is Latin for Step by Step, Ferociously]. We’re going to build capability step by step and we’re going to continue on down that path.
Your business plan is Jeff Bezos, or is it more complicated than that?
My business plan has become the way that we get the kind of vision we’re committed to — putting people into space to live and work — and to create a self-sustaining, self-funding, sustainable business. That’s why we’re making sales. That’s why we’re going to sell to a lot of people, and why we’re going to continue along this path. It has to be a business.