Mark Whittington has a useful overview of COTS. The only problem is in this paragraph:

The Falcon 9 is designed to launch up to 24,750 kilograms into low Earth orbit for a cost of seventy eight million dollars, according to the SpaceX web site. That compares to a cost of two hundred fifty four million dollars to launch 25,800 kilograms into low Earth orbit estimated for the Delta IV Heavy, a competitor to the Falcon 9 built by the Boeing Corporation.

No. Those numbers are the price, not the cost. Confusing the two words is one of the reasons that people get confused about whether or not we've made any progress in reducing launch costs over the years (partly because we don't really know what launches actually cost, particularly in Russia, but also with the Shuttle, due to opaque bookkeeping).

Price is what is charged to a customer. Cost is the amount of resources that the launch provider has to devote to providing the service. If cost is less than price, then the provider makes money; if it's the other way around, then the provider is operating at a loss. I'm sure that SpaceX costs (at least its marginal costs) are less (and probably quite a bit less, to account for the business risk factor of developing it) than the published price, or there would have been no point in going into the business. I'm also sure that Lockheed Martin is not losing money on Atlas launches.

In both cases, of course, the average cost is highly dependent on flight rate. This is one of the reasons that EELV prices have gone up dramatically over the last few years. In fact, I used that example in my piece in The New Atlantis a couple years ago as an explanation to why vehicle design is at best a secondary issue of launch costs, while flight rate is a primary one. There's an appalling amount of ignorance, even within the professional space community, as to the reasons for high launch costs, not to mention low reliability (see comments in this post for an example), which is one of the barriers to improving the situation. And of course, the problem is made worse by the lack of recognition of their lack of knowledge. As the old saying goes, it's not what we don't know that hurts us, it's the things we know for damn sure that are wrong.

Just FYI, the sources for the launch cost is SpaceX's website for Falcon 9 and the Encyclopedia Astronatica for the EELV respectively.

Mark, it doesn't matter what the source is. Those are not costs--they are prices. SpaceX is not going to disclose their costs, nor is Lockmart. Words mean things.

For the buyer (NASA) of the COTS launch services, the price is the cost.

Rand, I am with John on this. Our curmudgeon used the words correctly. When supplier prices are input into a budget, they are costs.

Your point about the flight rate is well taken, but you need a better lead-in for the blog article.

I think it's a matter of sematics. Rand is right from the point of view of the launch service provider. John is right from the point of view of the launch service customer.

They may be the out-of-pocket cost to the payload user, but they're not the intrinsic launch costs, which is what we're trying to reduce. They are the price. As I said, words mean things. The very fact that some seem unable to make this distinction is, as I said, one of the things holding us back.

Rand's point being, essentially, that the price of a launch is an insufficiently stable and meaningful value *for measuring progress*, because it includes profit for the launch company, which can vary wildly based on the contract.

The cost *to the launch provider*, that is, the point at which the launch provider would pay bills and salary but make no profit, is a much more stable and meaningful baseline to be measuring by.

Think about, for a second, measuring price/sales vs. price/earnings. Price/sales can be a useful metric, but in the end, it is more important for investors to know what the bottom line is, with costs and expenses taken out. In this case, we're looking at it the other way around, focusing on costs, rather than earnings; but still, adding both together blurs the picture.

Yes, if you're in the market, you have to pay the launch provider's price, not their cost. But if you're measuring the progress of the market as a whole, you really need to focus on the cost.

From an economic perspective, I think it makes more sense to look at price right now - because there is an effective monopoly on launch services. Cost comes into play as more competition arises - so that will be the important metric soon, just not now.

Price is an indicator of the state of the market. Cost is an indicator of the state of the technology.

Rand is right. If you drive down cost through improved technology/design, then you have the option of lowering the price or making more profit.