Nuclear Weapons

Update on the Reliable Replacement Warhead

05.08.06 | 5 min read | Text by Ivan Oelrich

At first glance, who could complain about replacing current nuclear warheads with ones that are more reliable? After all, since we have them they should work. But the Reliable Replacement Warhead (RRW) program may do no good, may do much harm, and will cost a lot if carried forward.

The Department of Energy’s (DOE) Reliable Replacement Warhead (RRW) Program is, as one might guess from the name, intended to develop more reliable nuclear warheads to replace existing warheads. One of the most important Congressional supporters of the Reliable Replacement Warhead Program, Congressman David Hobson of Ohio, was recently quoted saying, “This [the RRW] is a way to redo the weapon capability that we have and maybe make them more reliable, make them better mission capable.”

There are a couple of problems with this statement. First, it implies that there is some problem with the reliability of the current stockpile. I was recently at a meeting held under the “Chatam Houes Rule” (that is, I can use what was said at the meeting but I cannot attribute it) to discuss the RRW. Among several of the participants, who had extensive knowledge of nuclear weapons’ design and the stockpile stewardship program, a debate arose about the reliability of the current arsenal. Some claimed that the current arsenal is 98% reliable while others challenged that number, arguing that the reliability is better than that. No one suggested the reliability was less. The fact is, the current arsenal is extremely reliable and there are no foreseeable problems that will change that assessment. (One should note that 98% “reliable” does not mean that 2% of the weapons will not go off, but that they might explode with a yield somewhat less than specified; so a 400 kT bomb that explodes with a yield of 300 kT is considered “unreliable.”)

It is possible that an RRW could be more reliable than current weapons. If the current arsenal is 98% reliable, then an RRW could, in theory, be 99% reliable. But 98% is a pretty high bar to vault and it is not at all clear that an RRW could be made more reliable than existing weapons. Moreover, there is no conceivable meaningful difference between the two cases. If some war plan depends on the difference between 98% and 99%, then we need a new war plan, not a new warhead. Also, the warheads sit atop missiles that are reckoned to be about 90% or so reliable, which swamps the unreliability of the warheads themselves. Indeed, given the finite number of weapons and tests that might be available even in theory, it will be difficult, from a statistical point view, to even measure the difference between 98% and 99%. In this case, the RRW is clearly a solution to a problem that we either do not have or don’t need to fix. It is highly unlikely that the RRW will be more reliable than the highly reliable current arsenal and certainly not meaningfully more reliable. There are two great dangers here: The first is that discussion about the “need” for a new “reliable” warhead will make people think that current warheads are not reliable when they are, making us do something uncalled for, or even reckless, to solve a non-existant problem. The second, ironically, is that after an RRW has been introduced a new group of decision-makers will realize the thing has never been tested and might begin calling for renewed testing.

It appears, however, that the main justification for the program is not the first “R” (increased reliability), but the second “R” in RRW (to replace existing warheads). Some who acknowledge that the current stockpile is adequately reliable nevertheless claim that maintaining the stockpile, and that level of reliability, will become increasingly expensive. An RRW might reduce those costs, but not necessarily. There are, as yet, no detailed cost estimates for the RRW Program so it is merely an assertion to say that the program will save any money.

Besides, the Departments of Defense and Energy are not going to stop the stockpile stewardship program for existing warheads while the RRW is being developed. There are several hints that the RRW will be deployed, not in place of existing warheads, but in parallel with existing programs. Even if the new warhead replaces old warheads in a one-to-one exchange, the old warhead program will remain in place until it has been completely replaced by the new warhead. There is some indication that the military users will want to keep the two types of warheads deployed in parallel for some time, a decade or more, to reassure themselves that the RRW holds up to its billing, before completing any replacement. The Replacement Warhead Program might eventually be cheaper than the current Stockpile Stewardship Program, but the two programs together can’t possibly be cheaper than Stockpile Stewardship alone. And costing less is not enough if “saving” that money requires a big up-front investment. As Richard Garwin has pointed out, we have to take into account the discounted value of future savings. It makes great sense to spend $1.00 today to save $1.10 tomorrow but it makes no sense to spend $1.00 today to save $1.10 thirty years from now.

The National Labs are energized by the RRW Program because it allows them to exercise their design skills. One of the stated objectives of the Administration is to have a responsive nuclear weapons “complex” and building a new warhead will test, and help maintain, that complex. Indeed, one way to look at the RRW program is to see the warhead as just a means to an end; the end is keeping a ready nuclear design and manufacturing capability. Last autumn I wrote a short article on the RRW in the FAS journal, the Public Interest Report. Logically, it had to be written in the form of, “If the RRW turns out to be X, then Y…” because it was not at all clear then what the RRW program really was. The program is still not entirely clear, but clearer. It seems that the RRW is grist for the mill of the nuclear weapons complex. The overarching objective is indefinite maintenance of a warm nuclear weapons manufacturing base. The Department of Energy, no matter how far into the future it looks, cannot envision an America without thousands of nuclear weapons.

The Arms Control Association has a good piece on the RRW.