> I don't think most people are really aware of the obscene costs of military conflicts.
Would those costs still be obscene if you were in a conflict where you’d want to use a significant number of them? Right now they’re expensive because they’re essentially just sitting around.
Speaking of Javelin missiles, mentioned upthread. In 2022, when the war in Ukraine erupted, the small stock of Javelins which the NATO countries were able to provide was spent in like first several months. After that, $300 drones carrying a $1000 armor-piercing round started to dominate the battlefield, leading to terrible losses in Russian armor, especially the newest and most expensive tanks. Similarly, having lost a number of advanced and expensive aircraft, and watching advanced and expensive cruise missiles mostly shot down during airstrikes, Russian forces turned to expendable drones imported from Iran (!) and expendable rockets imported from North Korea (!!).
In other terms, Protoss-type technology works well when you have a large advantage and need to deal a decisive blow; an example would be B-2s bombing the Iran nuclear facilities. But when you're in a protracted conflict against a capable adversary, Zerg-type technology, cheap, flimsy, and truly massively produced, seems to be indispensable.
It’s interesting to see both kinds of drones in Ukraine as well. Ukrainian drones are built for €300 or so and they’re staggeringly effective. “Western” drones as made by Helsing and other companies cost several thousand. While they may have more features, it’s not clear that they’re doing 10x more damage than the Ukrainian ones.
Ukraine plans to buy 4.5 drones in 2025. They’re definitely going with volume over software features. Further they’re allowing frontline drone regiments to earn “points” based on kills and using the points to buy their own drones instead of allocating them top down. The regiments appear to be favouring the cheap drones over expensive ones like the Helsing HF-1.
What’s interesting is that European governments are probably going to end up buying tens of thousands of the expensive drones because the laundry list of features, rather than investing in true mass production like the Ukrainians have. Going the Protoss way, rather than Zerg.
€300 drones are anti-personal ones. They unable to penetrate tank armor except when tank is abandoned and sits open. Drones in current generation are 10x more expensive even when produced in Ukraine.
This is largely incorrect. The USA has supplied Javelins steadily throughout the war, including 2022. Eg, the 1000 announced at the start of June that year [1]. Moreover they were far from Ukraine's only ATGMs in 2022. You're ignoring NLAWs and Stugna-P which certainly weren't 'spent in the first several months'. BGM-71 TOWs were supplied from the middle of the year.
The war in 2022 was primarily an artillery war. Drones were in use, but the dominance you describe came later.
No economy of scale. The cost to build one car is ~$100 million. The cost to build the second one is ~$20K. The only reason you can buy a car for $40K is because they build millions of them to spread the initial investment. The military buys missiles in units of 100s and there are no other buyers, so the cost per missile is massive.
Those contracts are put out for competitive bids. Profit margins for defense contractors aren't very high. The prices are driven by a combination of strict requirements, lack of economies of scale, and legal compliance with government mandated processes.
Consolidation of defense prime contractors was inevitable due to budgetary realities and the escalating complexity of major programs. It's unlikely that keeping a bunch of small, weak companies around would have produced better results for the military or taxpayers.
If the defense contractors figured they could get away with those costs, at higher volume? Hell, yes.
If the U.S. still had it's own (gov't-owned, gov't-operated) production facilities - as, historically, every A List nation has had - to provide honest competition? Hell, no.
Unlikely. The complexity of major defense programs has increased by orders of magnitude since WWII. Running a small-arms ammunition factory is one thing, but the notion of the government acting as it's own prime contactor for something like the Tomahawk program is just absurd and totally impractical.
> ... notion of the government acting as it's own prime contactor for something like the Tomahawk program is just absurd and totally impractical.
The small-arms ammo was just their MVP for 1777. By the late 1950's, the government was building stuff like this in it's own (gov't-owned, gov't-operated) shipyard:
I'm thinking that a Tomahawk has rather fewer parts, from fewer subcontractors, than a >60,000-ton aircraft carrier. And doesn't take multiple years of continuous work to build, either.
Nah. The complete Tomahawk weapons system is far more complex than a WWII era aircraft carrier. Beyond the missile itself there's an "iceberg" under the surface. The software alone is huge and requires major ongoing work from several defense contractors covering multiple embedded systems, mission planning, telemetry, launch platform integration (multiple different classes of surface ships and submarines, plus now ground launchers again), testing, etc. Plus customized builds for each of the export customers. You probably have no idea what actually goes into making this all work with an extremely high level of reliability.
Axiom: While, in the past, gov't organizations were quite capable of performing the largest, most complex, and most critical technological tasks that society faced, things are somehow Different Now - and only non-gov't organizations (very preferably for-profit corporations) are now capable of such things.
But what is actually Different Now is this: Our ruling classes de facto decided to reduce the gov't's core competency in a part of national security - because outsourcing those capabilities to for-profit org's was far more lucrative for them, and the nation seemed secure enough that they didn't much care about the downsides.
Humans are very responsive to their social environments, and its structure and unwritten rules. Setting the "Non-corporate" bit on the org that a human works for does not magically reduce what they are capable of. Linus Torvalds actually is the creator and BDFL of Linux. Even though he is an individual human - not a corporation, nor a secret front for one. The mathematicians who completed the classification of finite simple groups ( https://en.wikipedia.org/wiki/Classification_of_finite_simpl... ) over decades and centuries completed that massive task while generally working individually or in small groups, for wide array of colleges and universities.
Well that's your opinion. But so far we haven't seen any evidence that governments are able to build complex software as well as private industry. So I think we'll stick with the current approach.
Would those costs still be obscene if you were in a conflict where you’d want to use a significant number of them? Right now they’re expensive because they’re essentially just sitting around.