Golden Dumb Update
Fraud, Waste and Abuse
Since my original post on the Republican's anti-ballistic missile defense system revival [1], there has been quite a bit of churning but very little in terms of either accomplishments or transparency. This post attempts to clear up a few misconceptions and elaborate a little on some of the earlier points.
A systems engineering activity called systems synthesis precedes design activities, and includes answering questions such as “What is the system supposed to do?”, “How much will it cost?”, and “How well must it do it?” for examples. This brings us to the first layer of problems with Golden Dome. The executive order hints at what it's supposed to do but as we will see is ambiguous.
The cost question is the source of much disinformation and wishful thinking. The White House pegs the cost at $175 Billion over three years while most in the technical community, and some politicians as well, put it in the Trillions, with a far greater amount of time required to develop, test, produce and deploy. My own prior estimate of more than $10 Trillion is at the high end of the range because I assumed that an extremely robust system would be required. In a sense, my estimate took the mandate of the Executive Order at face value. Others arrived at lower estimates by simply assuming less protection, but we agree the cost would be more than $1 Trillion [2], [3]. In a masterpiece of bureaucratic diplomacy the Congressional Budget Office (CBO) provided a cost estimate that would appear to correspond roughly to Trump's number, if you were to read only the first two pages. In it, they offer a cost estimate in the range $161 - $542 Billion, depending on assumptions about performance. [4] Unfortunately, the scope of the pricing exercise was limited to that of a system capable of engaging one or two unsophisticated ICBMs launched by a rogue nation like North Korea, and not the general nuclear war scenario spoken of in the Executive Order. Thus, while the CBO has not yet produced a cost estimate for the full blown system, we can be confident that it will more or less align with the technical community's estimates (in the $Trillions). The CBO can hardly be blamed for providing only a partial cost estimate, because the Defense Department was charged with submitting “. . . a reference architecture, capabilities-based requirements, and an implementation plan for the next-generation missile defense shield. The architecture shall include, at a minimum, plans for: [a list of eight capabilities]” within 60 days [5] of January 27, 2025. Let's see: It was due on March 28th; as I publish this on June 26th it is 90 days overdue. So sad, the Missile Defense Agency had to cancel their plans for the coming out party. But, in Hegseth's defense, he had to arrange for the highly successful Donald Trump 250th Birthday parade on No Kings Day. [6] Not to mention renaming military installations and ships after traitors, purging books and DoD documents of any mention of a word that might have a sexual or DEI connotation, firing officers based on their gender or race, and oh, yes, planning the war of aggression against Iran.
Incidentally, it should be noted that the unreality of White House cost claims is recognized even by members of Trump's own party. In announcing the Senate Caucus for Golden Dome, Senator Sheehy (R., Mont.) said “It will not be a $25 billion or $35 billion project. It will likely cost in the trillions if and when Golden Dome is completed.” [7]
Before concluding I need to point out that all of the estimates possess a degree of incredibility because they posit costs for complex items that have never been built, and in some cases not yet even conceived. Ordinarily one develops a design concept before roughing out a cost. But the design concept is based on requirements, in effect committing to an answer to the third question, something that has not yet been done. So there are bound to be cost surprises, and if the past history of defense procurement is indicative, unpleasant ones.
The final question: “How well must it do its job?” is yet to be answered, and needs to be, and in a public venue, so that the American people will clearly understand that a system that truly prevents a catastrophe is completely unaffordable, even should it be technically feasible. In my earlier piece I posited an engagement kill probability sufficiently large so that there was only a 1% chance of a single ICBM getting through to its target, while the other cited studies assumed an average leakage rate of 10%, i.e. one ICBM in ten hitting its target. But we must be clear when we use statistics; if the expected number of missiles in a ten missile raid getting through is 1 in 10, there is a nearly 35% chance of none and a 58% chance of one or two. (There is also a 7% chance of three or more: a big risk.) Once we have decided how many hydrogen bombs we can tolerate exploding over our country, we can use statistical methods to determine what the probability of defeating a single ICBM must be to ensure that no more than that number get through at a given probability.
There are three approaches to deciding how many H-bombs we can allow to hit us:
The moral approach, which is to say “None”. The problem with that approach is that we cannot possibly afford it;
The “Assured Destruction” approach, harking back to the McNamara days, the criterion being that a country effectively ceases to function when 25% of its citizens have been killed; and
The “Nuclear Winter” approach, which limits the number of weapons used by both sides to a level well below that at which all advanced life forms on the planet are driven to extinction.
In the US, just over 40% of the population live in just 25 metropolitan areas [8], and I estimate that no more than about 60% of them would be killed with the judicious use of 150 thermonuclear weapons of 100 kilotons yield, each. As this is just below the notional number required to induce a nuclear winter, I'll use it in the remainder of this post. [9]
For the case in which the Russians launch 1500 ICBM warheads at us, almost all they are allowed under the soon-to-expire New START treaty, we will require Golden Dome to have the single engagement effectiveness values tabulated below in order to ensure that the probability of more than 150 ICBMs getting through is less than the indicated value. A 90% effectiveness is insufficient.
This calculation is but a straw man, based on an assumed threshold for destruction of our country, the assumption that said threshold is below that for causing catastrophic climate change, and completely neglects other delivery vehicles such as manned aircraft and nuclear tipped torpedoes. It is up to the Defense Department to do realistic and unbiased studies of the effects and inform the public of the results. We as citizens need to pressure them to do so, and also put pressure on the Administration to seek a real solution to the nuclear threat, which is to say negotiate a follow-on to the New START treaty with reduced weapons stockpiles. To succeed at that we need to cancel Golden Dome and the Sentinel ICBM as signals of good faith.
I promised myself and Timothy Snyder that I would constantly remind my readership of the need for everyone who believes in freedom and democracy to behave in accordance with those beliefs, every day. To remind ourselves of what that means, see
Notes
[1] https://stephenschiff.substack.com/p/transmuting-iron-to-gold
[2] https://www.aps.org/publications/reports/strategic-ballistic-missile-defense
[3] https://doi.org/10.1080/10242694.2024.2396415
[4] CBO letter dated May 5, 2025, Re: Effects of Lower Launch Costs on Previous Estimates for Space-Based, Boost-Phase Missile Defense
[5] https://www.whitehouse.gov/presidential-actions/2025/01/the-iron-dome-for-america/
[6] I know, there have been lots of photos showing the reviewing stand and the area around the Washington Monument, but this one, taken from above Constitution Avenue and showing the entire parade route, leaves no room for doubt as to crowd size.
[7] https://spacenews.com/senate-forms-golden-dome-caucus-to-champion-missile-defense-shield/
[8] https://www.statistica.com/statistics/183600/population-of-metropolitan-areas-in-the-us/
[9] As always, I will gladly provide copies of the calculations, in this case a Jupyter notebook and an Open Office spreadsheet, to any paid subscriber requesting them.
Maybe this administration will view a warning and time frame for the first bombs/missiles as being acceptable in their risk assessments.