The International Space Station continues, amidst politics and its own Arthur Anderson accounting mess, to inch forward. The current mission of the Space Shuttle Atlantis will add a key element to it–a truss that will provide the backbone for assembly tools and the additional solar panels and radiators needed to provide adequate power for the scientific research envisioned for it.

The Space Station program started formally with an announcement in President Reagan’s State of the Union address back in 1984 (though millions had been spent on conceptual studies for years prior to that). Its stated purpose was to perform scientific research, but its true purpose was to give NASA something to do as the Shuttle development wound down, just as the development of the Shuttle kept NASA centers occupied and maintained the jobs base in critical congressional districts after the end of Apollo a decade earlier. Later, its purpose transmogrified into, in addition, promoting “international cooperation” (and in the 1990s, providing “midnight basketball” for the Russian scientists who might otherwise be working for Saddam Hussein or North Korea).

And because requirements were never defined very well, and changed almost every year, and it wasn’t necessary to actually build hardware and launch it into space in order to meet any of the above true program goals, the program spent a lot of money for many years without a lot of results. The original hope was for a station in 1992, the five hundredth anniversary of Columbus’ discovery of the New World. Instead, the first piece of hardware didn’t go into orbit until almost the end of the century.

But now we have a space station, of sorts. Much is made of how large it is, and if you simply put a yard stick on it, or count miles of plumbing or wiring, it is certainly the largest structure we’ve ever put in space, but this is misleading. For all of its supposedly gargantuan attributes, it remains unable to support more people (three) than our first space station–Skylab–launched over a quarter of a century ago. Indeed, the crew size is so small, in a facility so complex, that almost all available crew time will be spent in constructing and maintaining the space station itself, rather than in doing any actual science. So on a dollars per manned-space-science basis, we’ve gone backwards since the 1970s.

I’ve previously discussed the economics of the Space Shuttle, and why it’s so expensive to operate, for reasons that have nothing intrinsically to do with the fact that it’s a space vehicle. When you need a multi-billion dollar infrastructure to operate a vehicle, and only fly that vehicle a few times a year, every flight is going to be wallet-achingly expensive. The ISS has the same problem–it has no economies of scale. It all comes down to the difference between fixed and variable costs.

The first fifteen years of the program spent a billion or two year after year, doing nothing except designing, and redesigning, and planning, and building mockups and prototypes, and eventually tooling to actually build space station components. If they had cut the program off a couple years before they had actually launched the first component, they could have spent upwards of twenty billion dollars without building anything. The cost of the actual hardware itself, in comparison to this expenditure, is quite small. That means that, at least in theory, the cost of building the station much larger wouldn’t be that much more.

Here’s a real-life example (though the numbers will be approximate). Back in the late 1980s, when I worked on the program, the estimated cost to completion was about thirty billion. The people at NASA were told that due to budget constraints, they had to cut five billion out of it. A program manager at NASA (who will remain nameless) told me, “that’s the cost of the hardware!”

In other words, in terms of actually building a space station, they could spend twenty five billion dollars and get nothing, or spend thirty billion dollars and get what they planned. And this implies that they could spend thirty-five billion and get twice what they planned, for an increase in costs of only 17%. Another way to look at it is that for a doubling of the budget, they could have a station five times the size. (This is a little oversimplified, but you get the idea.)

Why do we have such a penny-saved, pound-foolish space program?

Because space isn’t important. It doesn’t matter whether the space station is as large as it is, or half that size, or ten times that size, because the people funding it don’t care how big it is. The only thing they care about is how many jobs (not how much wealth) it creates, or how much international cooperation it promotes.

No one will ever lose an election because we don’t have a good civil space station. That’s why, if we really do care about having good space stations, and other things, it’s important to get space activities out of the Big Government Program mode that they’ve been in since the Cold War, and into the private sector. Because private enterprise will care about actual outputs and marginal costs, and value for the dollar, in a way that governments never do.