The cost of insuring a satellite’s launch and its first year in orbit has come down dramatically in recent years. The amount of insurance capacity has increased, with new underwriter entrants sensing opportunity. Premiums for ArianeGroup Ariane 5 and SpaceX Falcon 9 launches have dropped to around 5% or less of the insured value. The International Launch Services (ILS) Proton rocket is another matter. Underwriters continue to insist on a 12% premium. ILS says it cannot survive with this differential. Credit: SCOR

TOKYO — In what may be a first in the space industry, the cost of insuring a major veteran commercial-launch vehicle is so much higher than its two competitors as to raise questions about the rocket’s commercial viability.

Insurance premiums for launches of International Launch Services’ Russian Proton rocket, which satellite operators and insurers say is a necessary third leg for the commercial market — the SpaceX Falcon 9 and the ArianeGroup Ariane 5 being the other two — total about 12% of the insured value.

That compares with 3-4% for Ariane 5 and 4-5% for the Falcon 9.

In dollar terms, that means that ILS customers seeking a $200 million policy covering the the value of the satellite, the launch and the satellite’s first year in orbit, would pay a $24 million premium.

The same customer launching the same satellite on Falcon 9 or the Ariane 5 would pay no more than $10 million, and possibly less.

The spread has been exacerbated by Proton’s recent history, by the adoption of SpaceX’s Falcon 9 by satellite fleet operator SES, whose market influence is considerable, and by the overall collapse in premium rates in the last few years.

“We have been insuring Proton for decades,” said Stephane Rives, deputy chief space underwriter at SCOR Global P&C. “In the last five or six years, there has been a string of failures. Today’s rates are a reflection of the reality. Yes, the last failure was two years ago. Then it went one year when there was no launch of Proton. We are very happy that their recent launches have been successful.

“We need a successful Proton, like the rest of the industry because the more alternatives we have the better it is for us. But at a certain point, the market lost its patience with Proton. And now we are in a wait-and-see mode.”

Rives was speaking here Oct. 12 at the APSCC 2017 conference. Other insurance officials agreed with everything he said, but also acknowledged that Proton’s 12-launch success rate ought to move the needle at least a bit.

“SpaceX gets a break because of the cult of Elon [founder Elon Musk] and partly because of SES,” one official said. “But Falcon 9 is also launching a lot more often than Proton, which tends to mitigate” the two failures SpaceX had in 2015 and 2016.

Denis Bensoussan, head of space risks at underwriter Beazley, said the market wants a veteran vehicle like Proton to be an active player.

“As soon as satellite operators make the choice of Proton again, we will support the operators — as we always have done. At the moment there have not been many recent launches with Proton.

“Hopefully a few more are sold in the coming years and that they are able to improve the reliability and everybody will be satisfied to have another mature, reliable vehicle. In this context of new launch vehicles, we need to rely on mature vehicles to help us to support innovation. I would welcome Proton back.”

Insurance underwriters point to policies covering the first reflight of a SpaceX Falcon 9 rocket, and the first use, on the Eutelsat 172B, of an all-electric design with the thrusters mounted on robotic arms as evidence of their willingness to embrace new technology. But they worry that the current low-premium environment may not withstand the introduction of 10 new launch vehicles in the coming four years given rockets’ history of high failure rates in their initial launches. Credit: SCOR

Bensoussan was referring to the fact that all the major commercial-launch providers are introducing new vehicles in the coming years. Rocket history shows new entrants are more likely to fail than established vehicles.

By itself, that would not be a huge problem for the industry. But the low insurance rates for Ariane 5 and Falcon 9 mean total premium revenue for 2016 was insufficient to cover a single major launch failure, and 2017 looks no better.

In addition, the commercial market for large, geostationary-orbit satellites has been depressed for two years, further straining ILS’s market niche. The company is introducing a lighter version, called Proton Medium, which is scheduled to debut, in two versions, in 2018 and 2019.

Satellite operators in the past have sought to assure at least three viable launch vehicles by favoring the weakest for limited periods of time. That was true of the Sea Launch vehicle in the early 2000s, which ultimately failed to establish a foothold in the market.

But that was during a period when the major operators — Intelsat, SES and Eutelsat among them — were expanding their fleets in addition to replacing aging spacecraft.

That is less true today. Even these large fleet owners are reducing capex and geostationary-satellite orders.

All this makes insurance underwriters nervous and less inclined to look beyond their immediate concerns to the future of the industry.

Space insurance remains a profitable business.  But if a full year’s premium volume is just one launch failure away from being wiped out, and if the number of geostationary-satellite orders stays low, linsurers are hesitant to give Proton a big break on rates.

ILS President Kirk Pysher argues that management at the state-owned Khrunichev Space Center, Proton’s prime contractor, has embarked on a thorough, three-year quality-improvement program that is coming to an end.

That concern for quality, he said, is one reason why Proton was kept out of service — for Russian government and commercial launches — in 2016. The vehicle is now back.

Since its last failure, Proton has launched 12 times, all successfully.

Where is Khrunichev in its Proton quality-improvement effort?

We are nearing the end of a three-year quality improvement plan. We are seeing real results there, not only in the quality of the hardware, but also with the new technology attracting new talent that likes to play with the toys. It’s a never-ending process in producing a reliable product.

How do you assess the market and the insurance premiums demanded for your launches?

The problem is it’s a Catch-22. The insurance market wants us there, but if they are not going to be there to support us, then we are not going to be there.

Our challenge is when new, prospective customers go out to market, the premium rates they are getting back, are 12%, which automatically puts Proton out of the game. We just can’t make up that difference on launch-service price, which has already been reduced.

You talk to underwriters and they say: This is the reality of the situation based on Proton’s flight history. They say they are actually looking at the risk profile. But then when you ask them about SpaceX, none of them can justify the rate that SpaceX has today.

So it’s no longer a risk profile they are using to insure SpaceX. Instead of using a risk-management approach for customers, it’s a gamble. You go to Las Vegas and play roulette. This big number board that has all red numbers on it. And you say: It’s got to be red again because of all those red numbers, and you put your money down.

The reality is that the odds are the same, regardless of what’s on that number board. It’s the same for launches.

Insurers sometimes talk of the cult of Elon as being a factor. How about having a cult of your own?

I’m still trying to figure that out, and to find out what sort of Kool-Aid we can serve.

Insurers say their customers, the satellite operators, are seduced by SpaceX prices and willing to overlook some risks. In addition, SpaceX is launching a lot — 15 times this year so far — and insurers like that.  SES’s support helps.

Right, but there are a lot of other operators like Intelsat or Eutelsat and they are launching, too.

But given Intelsat’s financial position, can they sell their board on an ILS launch that’s $15 million more than the competition just out of concern for the industry’s long-term health?

No, they can’t. We have contracts with them, but they can’t assign spacecraft to them because of this insurance issue. It’s not helping them.

Can you cut some kind of deal with underwriters in the absence of a bulk order from an operator that gets you a lower premium rate?

One idea that insurers have brought up is that ILS goes out and buys, say, a 10-launch deal. That’s all well and good, but the problem is that it’s not the launch cost that drives the insurance rate that is causing us problem. It’s the insurance on a $150 million or $175 million satellite.

So it has to be an L+1 [launch plus one year in orbit] policy. How do you define and L+1 policy if you don’t know that the spacecraft is? The launch cost is not the issue. We can do that readily. We can insure a launch-only policy that covers just the launch cost, but that would be a $30 million or $40 million policy. That doesn’t do us any good.

Insuring a satellite after its first year in orbit has never been less expensive — about 0.5% of the insured value. Depending on the satellite’s history and that of similar spacecraft, policies may include exclusions. Credit: SCOR

You have to bring in that L+1. One of the ways we’ve been look at to address that is through the Russian side. We’re looking into creating a captive insurance company and looking within Russia to see how we can actually fund that and do self-insurance. We’re trying to figure out how to execute on that.

If the Western insurance market doesn’t want to play in the Proton game, or wants to play only for a minimal amount, then we have to find a way to get around them and eliminate them from the equation. That means some sort of self-insurance.

And the longer this goes on, the stronger a case we have with Roscosmos and in Russia to self-insure.

Arianespace did that for awhile with an in-house insurance operation to compensate for what they saw as an insufficient insurance market.

The way we see our market going forward is a lot of Proton Mediums, very few Proton-Ms — the Medium is that lower cost point. But if we tack on a $150 million or $175 million satellite, we can’t lower the price far enough for the Medium to make up that $20 million difference from the insurance differential. We just cant do it.

So we need another alternative. Right now, it’s somehow finding a self-insurance approach to address it.

Have you made your case to Khrunichev? Do they get it?

Oh yes, Khrunichev gets it and I’ve actually had this discussion inside of Roscosmos as well and they understand it, too. But you can’t just flip a switch and make the problem go away.

I think everyone was banking on the fact that as we get more successes, the rate will gradually drop. That has not been the case in general. Some underwriters have dropped their rate, but it hasn’t dropped as fast as we would have expected it to drop. The quotes that come out haven’t much changed since our last failure.

Your rates have come down. But Ariane 5 and Falcon 9 rates have come down much further. As recently as 2012 and 2013 you weren’t that far from them. The problem is compounded by the lack of GEO satellite orders.

And then we have the non-GEO market, and it’s not sure how much they will be insuring, and their payloads don’t cost that much.

It’s going to take a big failure, God forbid, to right the course.

The recent failures SpaceX and Proton failure have been uninsured or not much insured or insured by a different set of underwriters. The space underwriters are still making a profit.

They are even still profitable on Proton, overall. The premiums still exceed the payouts. The industry today seems based on what they paid out and what they participated in to evaluate risk. Whereas the overall risk should not be assessed based on their participation.

With the exceptions in 2007 and 2013, space insurance has been profitable for underwriters in the past decade. But with rates declining, annual premium volume has dropped to where a single large launch failure would wipe out a full year’s premium for the entire sector. Credit: SCOR

A Federal Proton mission and a commercial Proton are identical when it comes to a risk profile. The SpaceX vehicle that blew up on the ground [in September 2016] is identical to the rocket that flies. That [Falcon 9] problem could have manifested itself at ignition of the rocket. It was a design issue.

So just because the space insurers didn’t have to pay out [the on-pad explosion was covered under a separate policy covered by cargo and marine underwriters] doesn’t mean it shouldn’t be incorporated in the risk profile for the sake of their investors. But it’s not.

But insurers can make up for a SpaceX failure by the volume of business coming from SpaceX launches.

The underwriters are still assessing risk then the set rates for Proton but not doing so when they set a rate for Falcon 9. They are having their cake and eating it too. Investors are behind them expecting them to do the job. Maybe they’re blind to it if they keep making money and they don’t look at the true risk profile.

And I don’t think some of the insurers have grasped the fact that Proton may be the only heritage vehicle operating in a few years after around 2021, if everyone does what they say they are going to do with new vehicles. They should want to manage their risks today for the future.

The solution that gets their attention is probably just to take them out of the equation.