Collapse of the reusable launch vehicle industry in the 1990s

[Blackstar]

Over on the Kistler thread I made a comment about how I remembered in the mid-1990s going to a press conference by an FAA official who announced that their latest projections indicated that the market for large constellations of low Earth orbit comsats was going to be much smaller than everybody was predicting.

At the time there were a lot of entrepreneurs talking about building huge constellations of comsats to ring the globe. These would mostly be accessible either directly, with handheld phones, or through local ground stations. There were a lot of start-ups talking about doing this. A quick list:

Teledesic
AMSC
Constellation
Ellipso
Globalstar
Inmarsat P
Iridium
Odyssey
Signal

and so on (see the attached table)

In essence, lots of people were expecting that the comsat industry was going to evolve from a few hundred GEO comsats to thousands of smaller LEO comsats. And there was a lot of excitement and enthusiasm about this. It was sort of a feeding frenzy similar to the space tourism feeding frenzy of last decade (with an important difference: many of the start-ups were actually off-shoots of really big companies with a lot of money and a long track record in space communications).

Because there were going to be thousands of satellites, people expected that there would be a requirement for a lot of rocket launches. And because that demand was going to be so incredibly high, people expected that only reusable rockets would make sense. As a result, by this time there were also a bunch of start up companies that wanted to develop reusable rockets. There were at least half a dozen of them.

I have since found the reports that Futron produced for FAA on this subject. The earliest one dates from 1995:

http://www.faa.gov/about/office_org/headquarters_offices/ast/media/leo.pdf

Here is their money quote:

"Many of the systems listed in Table 1 were under consideration by industry at the time of OCST's 1994 assessment. Nonetheless, given the uncertain status of many of the efforts and the apparent impact of various market and regulatory factors at that time, OCST determined that one to two Big LEO systems and one Little LEO system might ultimately be deployed in the 1994-2005 time frame (these potential outcomes were described in terms of two market scenarios)."

That statement really annoyed the heck out of these companies. After all, there were about 19 companies gearing up to go after this perceived market, and now the government was saying that they figured that at most there was room for 2-3 of them at most. Suddenly the job of getting investment capital was a lot harder because the government was essentially predicting that the market was not all that big.

Within a few years these companies all started to shrivel up. You can look at the list and see which ones survived (although survivors may have also substantially changed their business focus).

Other FAA reports are here:

http://www.faa.gov/about/office_org/headquarters_offices/ast/reports_studies/forecasts/

[Blackstar]

There was some fallout from this FAA announcement. One was that it shook up the LEO comsat industry. It may have caused capital to dry up as investors suddenly became wary, thinking that this was a riskier investment than they were being told. It's hard to understand what was happening at that time unless you were involved, but there was a widespread perception that developments in computing and electronics/communications, combined with the end of the Cold War, meant that low Earth orbit was this amazing new opportunity for private enterprise. That vanished quickly.

There was also a lag effect. A bunch of reusable companies were just getting started up, around the time that the LEO comsat craze was peaking. So these companies were starting at exactly the time that the LEO companies were all starting to collapse.

Another effect, although minor, was that this forecast put Futron--which did the study--on the map. Futron made a prediction based upon careful analysis and it turned out that they were exactly right and the enthusiasts were wrong. That later enabled Futron to get into other forecasting jobs, like space tourism and commercial launch and stuff.

[Danderman]

The key difference between LEO comsats as a market, and space tourism, is that humans are not likely going away any time soon.

Of course, their money may go away.

[Lee Jay]

I thought the reason LEO comsat constellations died was the build out of the terrestrial cell network.

[Blackstar]

So what? What's my point in all this?

See this attached slide from Henry Hertzfeld warning about promises of market demand. The LEO telecom example is really the best one, because many of the start ups were real business with lots of experience and capital and they got overly enthusiastic. That was a real bubble, but it was not caused by amateurs. Hertzfeld notes that last decade we saw the same thing, with a whole bunch of companies predicting a great demand for space tourism, and trying to develop their rockets to supply it. Lots of companies in 2003, even lots of companies in 2004-2005. Now it is a decade later and there are a lot fewer companies, and they still are not flying.

(I could add another mini-bubble around 2011 or so when suddenly there was a prediction of this great demand for scientific use of reusable suborbitals. That bubble still exists today, but they don't have the vehicles to service it, and one can expect this enthusiasm to wane as people get tired of waiting for something that never happens.)

There is a shallower bubble right now with the "sovereign markets" idea. Bigelow has based his model upon providing services to foreign governments that may want to fly their own astronauts in space. So far that has not worked out. Golden Spike has essentially the same model. That has not worked out either.

So we can learn from the past.

[Blackstar]

I thought the reason LEO comsat constellations died was the build out of the terrestrial cell network.

That was most of it. But there was the inherent problem with the costs of space. These constellations were going to be expensive to develop. The build costs alone were huge, not to mention the transportation costs. And ongoing maintenance was not going to be cheap.

[Lee Jay]

I thought the reason LEO comsat constellations died was the build out of the terrestrial cell network.

That was most of it. But there was the inherent problem with the costs of space. These constellations were going to be expensive to develop. The build costs alone were huge, not to mention the transportation costs. And ongoing maintenance was not going to be cheap.

Those seem to be to be some of the reasons the terrestrial cell network got built out - it out-competed a LEO constellation that was designed to do the same thing.

And there are two main reasons (power and latency).

First of all, LEO is a rotten place for a comsat because of poor coverage.  The only reasons to have them there are power demands of the ground interface (your cell phone) and latency (GEO is rotten because the speed of light is too slow for low latency when you're out that far).  Well, LEO requires less power and has far lower latency than GEO, but for the same reasons, the terrestrial network has even lower power requirements and lower latency.

But the terrestrial network has even lousier coverage than LEO.  In GEO, you maybe need a handful of satellites.  In LEO, you need a few hundred.  On the ground, you need tens or hundreds of thousands.  So, it wasn't clear who was going to win the tradeoff there, but the quicker entry, lower barriers to entry and lower latency and power requirements ultimately favored the terrestrial network.

[Blackstar]

But the terrestrial network has even lousier coverage than LEO.  In GEO, you maybe need a handful of satellites.  In LEO, you need a few hundred.  On the ground, you need tens or hundreds of thousands.  So, it wasn't clear who was going to win the tradeoff there, but the quicker entry, lower barriers to entry and lower latency and power requirements ultimately favored the terrestrial network.

I agree, particularly with "So, it wasn't clear who was going to win the tradeoff there." That's why all these start-ups sprang up: it just was not clear at the time.

However, I think there was one other factor which gave terrestrial networks a big advantage over LEO, which was that a terrestrial network could plug directly into the existing (cable) infrastructure and start generating revenue immediately. Put up a few dozen cellphone towers and you could start selling phones and charging to use the towers. Money comes in immediately. That was not true for satellites, where you had to build up most of your constellation before you could use it. Very high initial capital costs.

Probably buried away in Verizon's files somewhere is their proprietary information on all that, showing how quickly after installing their first towers they were making money off of them.

[kevin-rf]

One thing I remember about the 90's period was the massive build outs of the terrestrial fiber networks. So much so that much of it has remained dark. It was never actually hooked in. They where burying lines all over the US. It was also a bit of a bubble.

So while the terrestrial out competed the LEO and GEO system, mistakes where made on all sides. Not every market materialized.

Think of it this way, you now have to look far and wide to find (can still find a few in WV) remaining microwave relay antennas of the old US backbone. They have mostly been replaced by fiber and satellite.

As a side note on fiber, while Verizon (major US telecom carrier) still sells Fios, it is no longer laying new Fios fiber and expanding the last mile coverage areas. Fios was a grand Verizon plan a few years back to bring last mile fiber to every subscriber in it's geographic area.

[Lee Jay]

However, I think there was one other factor which gave terrestrial networks a big advantage over LEO, which was that a terrestrial network could plug directly into the existing (cable) infrastructure and start generating revenue immediately. Put up a few dozen cellphone towers and you could start selling phones and charging to use the towers. Money comes in immediately.

Yes, this is precisely what I meant by a lower barrier to entry and quicker entry.

[Danderman]

(I could add another mini-bubble around 2011 or so when suddenly there was a prediction of this great demand for scientific use of reusable suborbitals. That bubble still exists today, but they don't have the vehicles to service it, and one can expect this enthusiasm to wane as people get tired of waiting for something that never happens.)

I never heard of this one.

Are you suggesting that we are not going to have reusable suborbital vehicles any time soon?

[Danderman]

There is a shallower bubble right now with the "sovereign markets" idea. Bigelow has based his model upon providing services to foreign governments that may want to fly their own astronauts in space. So far that has not worked out. Golden Spike has essentially the same model. That has not worked out either.

This is a not a bubble, this is a quasi-religious belief on the part of people who require large sums of money to make their dreams come true. 

To be a bubble, there has to be some reality to it, some actual base number of qualified customers (with the result that expectations exceed the reality).

[Blackstar]

This is a not a bubble, this is a quasi-religious belief on the part of people who require large sums of money to make their dreams come true. 

To be a bubble, there has to be some reality to it, some actual base number of qualified customers (with the result that expectations exceed the reality).

Well, I don't want to get to arguing semantics with you, but I don't think you're right. Tulips are the classic bubble. And there are lots of people who are convinced that their religious beliefs are real.

[Blackstar]

(I could add another mini-bubble around 2011 or so when suddenly there was a prediction of this great demand for scientific use of reusable suborbitals. That bubble still exists today, but they don't have the vehicles to service it, and one can expect this enthusiasm to wane as people get tired of waiting for something that never happens.)

1-I never heard of this one.

2-Are you suggesting that we are not going to have reusable suborbital vehicles any time soon?

1-Starting around 2008/9 there was a sudden increase in discussion about scientific use of reusable suborbital vehicles. NASA even started to throw some seed money at the subject around a year later. It is most exemplified by the Next-Gen Suborbital Researchers Conferences:

http://www.enhancedonlinenews.com/portal/site/eon/permalink/?ndmViewId=news_view&newsId=20110301007479&newsLang=en&permalinkExtra=NSRC/Alan-Stern/SwRI

The 2010 one was a "hit." The 2011 one was a "hit." So were the 2012 and 2013 ones. The number of attendees at these conferences keep going up. And the blogosphere was filled with prognostications by people who even predicted that science research might be a more important market for these vehicles than tourism (which was supposedly the market demand between 2003 and today, but still hasn't really gotten them off the ground).

All of this implies a growing market demand for reusable suborbital rockets for research purposes. It is my contention that this demand is over-hyped and may even be mostly an illusion (fueled by excitement, not actual money). I'd note that despite three years of wildly successful conferences, nobody is yet flying their experiments. If you were a grad student starting work in 2011 and hoping to fly, right now is about the time you would be picking a different topic for your dissertation so that you can actually finish your degree. Eventually the hype fades. It always does. I'd note that after doing conferences in 10, 11, 12 and 13, they are skipping a year and the next one is in 15. Does this indicate a bursting of the enthusiasm bubble?

http://nsrc.swri.org/

2-I was not suggesting that, but I'd note that we were supposed to have reusable suborbital vehicles "soon" ten years ago (in 2005 somebody assured me that it would happen by 2007). Are there going to be any available for regular scientific research this year? Next year? in 2016?

[savuporo]

(I could add another mini-bubble around 2011 or so when suddenly there was a prediction of this great demand for scientific use of reusable suborbitals. That bubble still exists today, but they don't have the vehicles to service it, and one can expect this enthusiasm to wane as people get tired of waiting for something that never happens.)

I never heard of this one.

Are you suggesting that we are not going to have reusable suborbital vehicles any time soon?

Yes, that bubble exists
https://flightopportunities.nasa.gov/platforms/

That page listed "STIG" as "in operation" a while ago, and has been up for years.
Apart Up Aerospace plain old sounding rocket, there is nothing else flying.

A bunch of conferences have taken place and quite a bit of hype around this.

[Danderman]

Yes, that bubble exists
https://flightopportunities.nasa.gov/platforms/

hmm.. some of those vehicles are flying today, and contracts have been let for future research opportunities.

I am not sure that this is simply confusing the reality that development of new flight systems inevitably takes longer than expected with the idea that they will never fly.

What happens if SS2 starts flying research payloads in 2015? Does this mean this wasn't a bubble?

You can't say it was a bubble if the supply side simply fails to meet demand, bubbles are the other way around. I don't see a huge amount of potential customers for suborbital research, it seems to me to be a niche market, even if SS2 and Lynx and Xaero fly a few payloads over the next few years.

[edkyle99]

So what? What's my point in all this?

See this attached slide from Henry Hertzfeld warning about promises of market demand. ...

Of all of the points listed on the slide, I believe that the Little-LEO bubble had the most substantial, and detrimental, effect on the space launch industry.  Remember, for example, Teledesic, the "Intenet in the Sky", backed by Bill Gates and Boeing and Motorola and other big hitters, with an original plan for 840 LEO satellites?  Boeing, I suspect, took the biggest hit on the launch business side from all of these new broadband in space concepts when the commercial backlog for Delta 3 and 4 dried up suddenly.  Well, that and the EELV contract scandal ...

 - Ed Kyle

[Oli]

At the advertised prices suborbital tourism probably only supports 2-3 providers (I think a read that somewhere), but with time prices will fall.

Tourism is the only industry I can see creating big demand for spaceflight in the next decades. Although I'm kind of surprised there is demand for suborbital tourism. Who pays 200k for 3 minutes view from 100km? Crazy people

[savuporo]

Yes, that bubble exists
https://flightopportunities.nasa.gov/platforms/

hmm.. some of those vehicles are flying today, and contracts have been let for future research opportunities.

Only UP Aerospace is flying suborbital payloads. Future contracts are future contracts.

The part of the point here is that time windows for getting a service in operations actually do matter. If someone actually flew a reusable orbital rocket by mid '90ies, a few other constellations apart from Iridum would probably be in orbit and doing some sort of business.

[Archibald]

Folks,

The bottom line is: why the LEO satellite constellations ? and why and when did the frenzy started ?

Motorola & Iridium. It started as early as June 1990. The reason why the sats were thousands is because they had to be in LEO, not GEO, because of the delay.

Next question - what did happened to satellite phones, by the way ?

The reason why the LEO constellations collapsed was that, at some moment in history, ground-based mobile phones bet satphones into submission.

There were two reasons for that IMHO

- GSM: it was started as early as 1982 by the Europeans, but even a decade later noone could believe it would ever work, making ground-based phones a reality.
Ground-based relays meant that mobile phones were dependant from frontiers unless of course some worldwide agreement was found. Well, such agreement was ultimately found, but after Motorola started the satphone frenzy in 1990.

- technical reasons: my limited understanding is that satphones did not worked well inside buildings or even in places like Manhattan were the sky is obviously blocked by buildings even outside. Ground-based mobile phones have no such issues...

So the RLV collapse was only the tip of a much larger iceberg. That is, sooner or later satphones were doomed by GSM.