When a rocket blows up —1 in 20 fail—who pays for it?

On a cloudy day in May, a Russian-made rocket roared off a launch pad in Kazakhstan. A few minutes later, it was falling from the sky, its payload—a $300 million Mexican communications satellite—disintegrating over Siberia.

A little over a month later, one of SpaceX's rockets exploded about two minutes after liftoff from Cape Canaveral. It took with it $110 million in NASA's supplies for the International Space Station.

Companies and governments spend huge sums to get things into space, but an average of about 1 in 20 launches will fail. That's why many of today's launches—especially those putting commercial satellites into orbit—are covered by space insurance policies to prevent catastrophic financial losses.

But insuring a payload on the tip of a rocket is entirely different from insuring a home, boat or car. There are only about 50 insured launches each year paying about $750 million in premiums to a handful of companies. If just a few big accidents pile up, there is a real risk of the industry ending up in the red—and it looks like 2015 is shaping up to be a tough year.

"The nature of this business is very volatile," said Chris Kunstadter, senior vice president and global underwriting manager for space at XL Catlin. "You don't have many losses, but when you do, they're large."

The failure of the Mexican satellite mission in May will cost insurers about $390 million—the second largest claim ever in the space insurance industry and far more than the premiums that have been collected so far this year.

Not only are the potential losses huge, but there are too few launches each year to do the same sort of actuarial math as in other types of insurance. A few bad launches in an unlucky year can cause the failure rate to bounce between 3 percent and 10 percent, and accidents tend to be total losses.

But the industry has other unique upsides that have attracted investors. A company knows right away whether its coverage is a loss or a gain. Space insurance is also uncorrelated with other types of insurance, so even if a hurricane wipes out hundreds of seaside homes all at once, the space insurance market remains serenely independent. The huge premium payments are also attractive to companies.

Interest in funding space insurance has been strong recently, and the extra coverage capacity has driven down rates to around 6 or 7 percent, compared with almost 20 percent a decade ago.

"Rates are approaching historical lows, despite the fact that we have had claims of approximately $500 million and only $250 million in premiums," said Mike Vinter, executive vice president of Aon International Space Brokers. "The underwriters are certainly in a loss position for the year."

"There is a lot of money out there seeking return," said Kunstadter. "The pricing is marginal right now—there's room for profit on some programs and not others, so it's not universally profitable."

After the first half of 2015, the Russian Proton rocket and SpaceX's Falcon rocket will be grounded while the failures are investigated. Underwriters are hoping to get launches up and running quickly to make up for the early losses, said Vinter.

Generally, one of the approximately 40 space insurance companies will put a maximum of $50 million on a launch, said Kunstadter. Berkshire Hathaway is believed to be on the line for $50 million of the loss in the Mexican satellite failure, according to Insurance Insider.

Rates low despite bad year

If margins are so small and 2015 might be a loss year, will rates readjust? Not any time soon.

Policies are hammered out one or two years in advance, so the price of today's rocket explosions wouldn't turn up until far down the road. And even then, the most likely outcome is higher rates for Proton rockets—the Russian-made rocket that destroyed the Mexican satellite and has had an unimpressive track record lately.

Other rockets, like Europe's Ariane 5 rocket and even SpaceX's relatively new Falcon 9, will probably continue to enjoy low rates, said Vinter. A major Ariane failure could sour the market, driving down capacity and then rates.

"Years ago if there was a launch failure it might impact rates across the board," said Vinter. "I don't see it in the near future. It would have to be a big one to scare underwriters out of the business."

As long as there is excess capital seeking high returns feeding into the market, the price probably won't move as much as the recent failures should suggest, said Kunstadter. And even with technology improvements, rockets will continue to explode—so space programs will need insurance to defuse the risk.

"This technology is 50 to 60 years old, and yet we're still having failures at 5 or 10 percent," said Kunstadter. "We're dealing with a high-energy system, and there are always improvements being made that can be the source of failures."