Of course, for decades all major nuclear powers have had intercontinental ballistic missiles (ICBMs) that travel through space at speeds in the Mach 20 range before reentering Earth’s atmosphere and striking their targets at hypersonic speeds. China’s RGV differs from ICBMs in that it is not limited to a ballistic trajectory, is capable of changing targets multiple times after launch, and can maneuver to avoid interception. This is certainly an advantage, but given that no one can really effectively intercept decades-old ICBMs, the question becomes: Just how important is this additional maneuverability?
But beyond general inefficiency and profligacy with taxpayer dollars, much of U.S. weapons development suffers from an acute case of what could be characterized as paralysis by analysis—analysis paralysis. Rather than building many prototypes and testing them to get the real-world data necessary to rapidly develop a working system, defense contractors spend billions on developing complex models that will theoretically predict how hypersonics perform. The problem is that no matter how complex the models, the data collected by building a prototype and actually flying it gives you better data. This is not to say that modeling is not critical to modern weapon systems development, but that if you can feed your models lots and lots of data from real-world tests, they will improve much faster than trying to extrapolate and interpolate data from limited test flights. It turns out that conducting extensive real-world testing, including destructive testing, is actually less expensive than endless modeling.
Such testing—and the mindset supporting such testing—is perhaps the number one reason SpaceX was so successful in meeting its requirements for NASA’s commercial space program, while Boeing, with its much larger budget, continues to flounder. And SpaceX continues to be willing to suffer losses of its test spacecraft on the theory that doing so accelerates development, and ultimately costs less than trying to model its way to success.
Further, when you test real-world systems, you are also able to see how the actual hardware and components perform under stress and make corrections when those components fail. Conducting many real-world tests reduces the cost of fixing things after a system is declared operational.
Once again, this is not to say that modeling is not useful; however, models populated with lots and lots of real-world data will outperform models that don’t have that advantage.
That the United States has fallen behind both China and Russia in missile technology, including hypersonics, despite spending many times more on national defense, is just another piece of evidence that spending more is not the answer. Rather, we need to spend our defense dollars far more efficiently to develop weapon systems that meet current needs and can be cost-effectively upgraded to meet future needs. This will require major disruptive defense reform far beyond what the Department of Government Efficiency will be able to deliver in a year or two.