A billion shekels, approved by phone. Twelve helicopters, and the money comes out of a five-year plan for Arab communities.
And that's the part that's going to get the headlines, but it's not the interesting part.
No. Daniel's question is the interesting part. He sent us the Israeli case as an entry point and then asked the thing underneath it. Not why twelve Apaches. How does any military know what it needs, in what quantity, and how urgently. Who notices a gap. Who decides that buying a thing is the answer. What analysis stands behind the numbers, and how much of the final decision is analysis at all.
And he wants us to use Israel as the door, then walk into the American process, and probably NATO alongside it.
Because that's where the vocabulary actually comes from. Let's start with what happened.
October fourteenth, 2026. The government approves a budgetary adjustment of roughly one billion and thirty million shekels for what the Defense Ministry calls urgent classified procurement. Twelve Apache attack helicopters.
Nine hundred and sixty-four million of that comes out of the 2027 to 2029 five-year plan for Arab communities. A hundred and sixty-nine million in 2027, three hundred and forty-two in 2028, four hundred and fifty-three in 2029. The remaining sixty-eight million comes from this year's budget.
And here's the bit that makes it a perfect teaching case. The deal had already collapsed. After three American extensions. The final deadline expired on October seventh.
The third anniversary of the attack.
And the requirement itself was never explained publicly. The Defense Ministry's phrase was urgent classified procurement. That's it. That's the whole public rationale.
A billion dollars, one-third of which had no funding source.
Two hundred and eighty-eight million. Because of the Numerator mechanism, which requires the government to identify a source for the full amount even when the payments are spread over years. So a deal that died over accounting gets revived by moving money out of a domestic social program, eleven days before an election.
The requirement was real enough to survive a two-year Air Force staff study. The funding was real enough to survive three deadline extensions and then die anyway. Those two facts sit next to each other very uncomfortably, and that's what we're going to pull apart.
So how does a military notice it lacks something? That's the first question, and honestly it's the least glamorous part of the whole process and probably the most important.
In the Israeli case it starts with October seventh, 2023.
The Air Force runs a staff study for over two years. Two purposes. Add capability, and replace a fleet that has been flown for more than three decades. The oldest airframes in that fleet are over thirty-five years old.
Thirty-five years is older than some of the people flying them.
That's a maintenance problem before it's a warfighting problem. Every hour on an airframe that old costs more than the hour before it. So you have a replacement driver and you have a capability driver, and they usually don't arrive at the same time, which is why the study takes two years.
And it wasn't just the Air Force on its own.
Right. The procurement gets folded into a force-build report chaired by Jacob Nagel, submitted to the prime minister and the defense minister about two years ago. The Air Force, the Defense Ministry and the National Security Council all sign off on the need. So by the time this reaches the cabinet, three separate institutions have agreed that the gap exists.
Which is not the same as agreeing on the size of the gap, or the shape of the solution.
No. And that's where the American process is worth looking at, because the Americans wrote more rules for this than anyone, and the vocabulary Daniel used in his prompt comes straight out of it.
JCIDS.
The Joint Capabilities Integration and Development System. It's the primary process the US Defense Department uses to determine weapon-system requirements, and it's managed by the Joint Requirements Oversight Council, the JROC, chaired by the Vice Chairman of the Joint Chiefs.
Note who's not in the chair. Not a service chief.
That's deliberate. The whole thing exists because the previous system was broken. Rumsfeld's memo in 2002 said the requirements system, and I'm quoting, invariably continues to require things that ought not to be required, and does not require things that need to be required.
That describes most organizations I've been in.
The fix was conceptual. Stop describing needs as systems. Describe them as capabilities. So instead of we need a more advanced fighter, you write we need the capability to defeat enemy air defenses. The acquisition people then figure out what that capability actually is.
Is that a real distinction or a bureaucratic one? Because it sounds like the kind of thing that gets said in a conference room and changes nothing.
It's real, and here's why. If you write down a system, you've decided the answer before you've done the analysis. If you write down a capability, you've left the door open for a drone, a missile, a ship, a cyber tool, or a doctrinal change. The Finance Ministry's argument in the Israeli case is exactly this. They didn't say the Apache is a bad helicopter. They said the requirement was written in a way that foreclosed the cheaper answers.
And the core analytical engine on the American side is the Capabilities-Based Assessment.
Three phases. Functional Area Analysis asks what are we even talking about. Functional Needs Analysis asks how good are we at it. Functional Solutions Analysis asks what should we do about it. The output feeds an Initial Capabilities Document and then a Capability Development Document.
Say the three phases again but slower, because I want to know what actually happens inside them.
Functional Area Analysis is scoping. You pick the mission area, you define the boundaries, you decide what's in and what's out. Functional Needs Analysis is the honest one. You measure current capability against a range of scenarios and you write down where you fall short. Functional Solutions Analysis is where you compare options, and crucially, materiel is only one category of option.
Which is where most of these processes fail.
Often, yes. There's a documented criticism that JCIDS over-favors materiel over non-materiel solutions. That's a nice way of saying the process can quietly assume the answer is a purchase.
If the answer is a purchase, someone gets a contract.
Someone does.
So Daniel's specific question. Twelve, not six, not twenty-four. Where does the number come from?
In the Israeli case, twelve reflects a squadron-sized buy and a favorable Boeing production slot. That's the honest answer, and neither of those is an analytical output in the way a listener might expect.
Squadron-sized means what, exactly. The unit exists, so you buy a unit's worth?
That's closer to the truth than most people would like. Squadrons have a standard establishment. If you buy half a squadron, you've bought a permanent training and maintenance problem. If you buy a squadron and a half, you're paying for infrastructure you don't have. So the organization itself constrains the number before any operational analysis has a say.
And the production slot.
Boeing has a line. If a slot opens, you can have it, and if you don't take it, the next customer takes it and you go to the back of the queue. When a gap is urgent, you take the slot you're offered, and the slot size shapes the order size.
So the number that shows up in the newspaper is partly a function of a factory calendar.
Partly. I want to be careful here, because the full operational requirement was never publicly disclosed. The Defense Ministry called it classified. So we don't know what the internal analysis actually said about twelve. We know the framing that leaked out, which is squadron plus slot.
Which is a real way to set a number. It's just not the way a listener imagines it happening.
And there's a counter-argument in the same reporting that's worth taking seriously, because it's the strongest version of the other side. Senior Finance Ministry officials challenged the Air Force's doctrine directly. The quote is, we have heard from quite a few professionals who believe the era of attack helicopters is over. And then, the military and the Ministry of Defense are opting for the most expensive route, and it is unclear to us to what extent they challenged themselves to think outside the box.
That's not a budget objection. That's a doctrine objection dressed as a budget objection.
And the Air Force's answer is specific. A former senior Air Force official said there's a huge advantage in having a battalion commander on the ground able to speak directly with a pilot hovering over the combat zone, coordinating precise high-intensity strikes, capabilities an RPA cannot match.
A remotely piloted aircraft cannot do it, in his framing, because of the conversation.
Because of the latency in the human loop. A battalion commander on the ground and a pilot overhead can build a shared picture in real time. The pilot can see what the commander can't and the commander can see what the pilot can't, and they resolve it between them in seconds.
Hovering, though. That's a helicopter doing something a fixed-wing aircraft can't do either.
Right. Endurance over a specific patch of ground, at low speed, while a ground commander uses it as a sensor and a weapon at the same time.
So the Finance Ministry says the era is over. The Air Force says here's the specific thing it does that nothing else does. Both of those could be true at once.
And that's not a failure of the process. That's the process working. The question of whether a capability is obsolete is exactly the kind of question a requirements process is supposed to surface, and in this case the argument got made in the press rather than in a review.
Because there was no public review.
Because the requirement was classified. There's no published Israeli equivalent of a JCIDS document. The Israeli process appears to run through internal Air Force staff studies, the Nagel committee, the National Security Council framework, and the ministerial procurement committee. That's a real process. It's just not a transparent one.
So the American system is the one with the rulebook. What does the rulebook say about urgency? Because urgent is the word the Defense Ministry used, and urgent is the word that gets you past a lot of gates.
JCIDS makes a formal distinction. Deliberate requirements are the normal kind. Formal, consensus-based, aligned to the major acquisition process, and they take a long time. Urgent and emergent requirements are the ones that, if unmitigated, would result in unacceptable loss of life or critical mission failure. Those get expedited review, and the target timeline is within two years.
Two years is the fast lane.
And even then, no JROC-commissioned Capabilities-Based Assessment has finished in under eleven months. The Quick Turn version runs thirty to sixty days, but that's a narrower product.
So when an Israeli ministry says urgent, what is it claiming, structurally?
It's claiming the condition in that JCIDS sentence. Unacceptable loss of life, or critical mission failure, if the thing isn't done. And in this case that claim was contested. Finance called the linkage spin. A legal opinion warned the emergency allocation could bind an incoming government and prioritize party interests ahead of the October twenty-seventh election.
Urgency got decided in a room where the operational question and the political question were sitting at the same table.
And the budget mechanism made it worse. The Numerator rule is what killed the deal the first time. It requires a source for the full billion even though the payments were spread across years. So a program that was affordable on a cash-flow basis was unaffordable on the government's own accounting basis. Two hundred and eighty-eight million with no source, and the deal dies.
Over a third of the cost. Not over the mission, not over the requirement. Over the accounting convention.
That's the collision the whole episode is about. The military analysis and the budget mechanism are answering different questions. One asks whether we need the capability. The other asks whether we can name the shekel today.
And a capability that is needed but cannot be named in a budget line does not get bought. That's not a failure of the military. That's the system working exactly as designed.
Which is worth sitting with. The Israeli requirement survived a two-year staff study, a Nagel committee report, and agreement across three institutions. It did not survive an accounting rule.
Now trade-offs. Because a requirement is not a purchase.
No. And this is where capability-based planning comes in, and it's worth naming where the term comes from, because it's not a marketing phrase. Paul Davis at RAND defined it in 2002 as planning, under uncertainty, to provide capabilities suitable for a wide range of modern-day challenges and circumstances, while working within an economic framework that necessitates choice.
Necessitates choice. That's the entire business in two words.
It's the honest framing of the problem. You cannot buy everything. So the question is not what do we need, because the answer to that is everything. The question is what do we buy given that we cannot buy everything.
And the distinction Davis draws is against threat-based planning.
Threat-based planning designs a force against one or a few well-understood threats. You know who the enemy is, you know roughly what they have, you build to beat that. It worked well when the threat was stable and legible.
Which describes exactly none of the last twenty years.
Capability-based planning considers a wider range of scenarios and confronts uncertainty rather than discounting it. The point isn't to predict the next war. The point is to buy things that are useful across many possible wars.
And that's specifically relevant to Israel after October seventh, because the threat picture itself was uncertain. The thing everyone was planning for was not the thing that happened.
And it wasn't the thing that happened in a way that was knowable in advance. That's the argument for building capability rather than matching a threat list.
There's a piece of this that I think gets lost, and it's the strongest critique of the Apache buy. Buying the platform is not buying the capability.
That's the DOTMLPF point. Doctrine, Organization, Training, Materiel, Leadership and Education, Personnel, Facilities. Materiel is one of seven. And the Australians, the Canadians and the British all have their own versions of the same list, which tells you something about how universal the problem is.
Everyone rediscovered the same failure independently and gave it a different acronym.
Australia calls it Fundamental Inputs to Capability. Canada calls it PRICIE. The UK calls it TEPID OIL.
Tepid oil.
Training, Equipment, Personnel, Infrastructure, Doctrine, Organization, Information, Logistics.
Whoever named that one was having a day.
The point stands though. If you buy twelve helicopters and you don't have the pilots, the maintenance crews, the spare parts pipeline, and the basing to run them, you have bought twelve objects. You have not bought a squadron.
So the honest version of the Finance Ministry's question is not is the Apache obsolete. It's whether the constraint was ever the airframe.
And there's a further twist, because some of those inputs take longer than the airframe. Training a new attack helicopter pilot is not a six-month project. If the personnel pipeline is the constraint, twelve helicopters arrive and sit.
How do you cost that, though? Because the sticker price is the thing everyone quotes and it's clearly not the number that matters.
That's the Analysis of Alternatives. It compares materiel solutions on total life-cycle cost. Research and development, investment, operating and support, and disposal, over an identical period of analysis, using present-value discounting.
Present-value discounting meaning a dollar spent in 2040 is worth less than a dollar spent today.
Using the Office of Management and Budget's Circular A-94 rate. Otherwise you're comparing programs with different spending profiles as though the timing didn't matter.
And operating and support is where these things actually get expensive.
It usually is. There's a specific line item in the American handbook for the Fully Burdened Cost of Fuel, which is the cost of the fuel plus all the personnel and assets needed to move and protect it. In a theater where you're running convoys to move fuel, the fuel itself is the cheap part.
So a helicopter with a worse fuel burn has a hidden multiplier on it that never shows up in the headline price.
And there are two metrics that matter for comparing programs. Average Procurement Unit Cost, which is the procurement total divided by the quantity, and Program Acquisition Unit Cost, which rolls in development as well.
So the headline number and the number that should be compared are routinely different numbers.
Routinely. And there's an affordability question on top of the cost question. Affordability analysis asks whether the program's life-cycle cost fits inside the long-range investment and force-structure plans. That's a different question from whether the program is worth its price.
It can be worth the price and still not fit. Those are different problems with different solutions.
And Cost as an Independent Variable, which is the American version of this, asks what gives the greatest capability for a fixed cost. You fix the budget and you vary the requirement, rather than fixing the requirement and letting the budget float.
Which is the opposite of how militaries normally behave, in my experience.
It's the opposite of how everyone normally behaves. The interesting question is when it actually gets used.
And that's where the Israeli case becomes a really clean test. Because the Numerator mechanism forced them to fund the full billion before ordering, even though the payments were spread over years.
Which is cash-flow budgeting colliding with lifecycle cost logic. Lifecycle costing says the program is affordable over twenty years. The Numerator rule says you need a source for the lot now. Those two statements can both be true, and only one of them decides whether the contract gets signed.
And the same pressure is governing everything else. Two KC-46 tankers, about five hundred million, no funding source. Six CH-53K transports, about a billion, same problem.
And the broader budget picture explains the squeeze. The March budget allotted defense a hundred and forty-three billion shekels, about seventeen percent of total spending. July added fifteen billion. Defense is asking for twenty-five billion more, targeting a hundred and eighty-three billion for 2026.
And claiming a shortfall of forty billion.
Claiming a shortfall of forty billion. Which is a number that has to be heard alongside the fact that a seventeen billion shekel transfer to defense was expected the following week.
So when Defense says urgent, Finance hears a number it has heard before.
And that's not cynicism. That's the actual structure of the argument. Every ministry has a list of urgent needs and a limited pot.
Let's take the urgency question seriously, because Daniel asked it directly and it deserves a real answer.
The formal definition is out there. Urgent and emergent requirements are the ones that if unmitigated would result in unacceptable loss of life or critical mission failure. That's a high bar. It's not desirable but deferrable.
It's the difference between we lose people if this isn't done and we lose readiness if it isn't done.
And readiness is a real cost. Deferred readiness doesn't show up as a casualty. It shows up as a decision you can't make in a future conflict.
Which is precisely why the urgency label is worth fighting over. It's the gate that lets a program skip the queue.
And in this case the label was contested from three directions at once. Defense said undelayable. Finance said the linkage was spin. A legal opinion said the emergency allocation could bind an incoming government and prioritize party interests before the election.
Three different definitions of urgent, each of which is internally coherent.
And there's an honest answer underneath all of it, which is that in this particular case the urgency was partly an artifact of a budget mechanism. The Numerator rule is what created the deadline. The deadline is what made it urgent. Strip out the accounting rule and you have a requirement that could plausibly have gone through normal channels.
So the operational urgency and the bureaucratic urgency were tangled together and nobody could fully separate them in public.
And a defense source said the quiet part out loud, that at best funding to accelerate procurement wouldn't be secured for another six months, and by then the entire world would be caught up in an arms race.
That's a real concern about timing.
It's a real concern. It's also a statement that could be made about any purchase at any time, which is why it doesn't itself settle the question.
Now the question that's actually hardest, and I think the one Daniel cares about most. How much of the final decision is analysis and how much is politics.
And I'd push against the framing slightly, because the analysis and the politics aren't on opposite sides of a fence. The analysis feeds the politics. It gives the political actor a defensible basis for the choice.
Which makes all of it hard to grade.
It does. And I think the straight answer has two parts. The analysis sets the menu. It establishes what is on the table and roughly what each option costs. Then the political layer picks from the menu, using criteria the analysis doesn't weight the same way.
Like what.
Industrial considerations, domestic politics, alliance relationships, budget negotiation between ministries, and in this case the timing of an election. None of those are illegitimate. They're just not in the analysis.
So the analysis is doing work, but it's not doing the deciding.
That's arguably what it's for. It makes the political choice accountable. If you buy twelve helicopters and the requirement document says you needed twenty-four, someone can look at that. If you buy twelve and there is no document, nobody can say anything.
Which loops back to the Israeli case's most frustrating feature. The full requirement was never disclosed.
Never disclosed. Which means the analysis may have been excellent. It may have been thorough. It may have addressed all of the counter-arguments the Finance Ministry raised, including the drone question. We just don't get to see.
There's a reasonable argument that we shouldn't see, because the requirement is classified for a reason.
There is. And there's a reasonable argument on the other side, which is that a billion shekels of development funding for Arab communities is being redirected, and the public has a legitimate interest in knowing why. Both of those arguments are real.
The legal opinion flagged a third concern, which is that the decision binds whoever takes office after the election. That's a specific structural problem with emergency allocations near a vote.
I'll say, because I think it's the sharpest criticism in the whole story, a senior officer said what happened here was more reminiscent of the conduct of a third-world country than of a nation insisting on calling itself advanced. That's a serving official describing his own government's process.
Strong words.
I don't think he's wrong that something went wrong. I just don't think the thing that went wrong was the requirement. It was the funding mechanism, and the timing.
It's worth saying that the analysis did work as designed in one respect. It surfaced a genuine disagreement about doctrine, between people who have all the information, and the disagreement was real, not manufactured.
That's right. The Finance Ministry's professionals believe attack helicopters are a sunset capability. The Air Force's believe the specific thing a helicopter does cannot be replaced. That's a legitimate strategic argument, and it should be fought out. Where it gets ugly is when it gets fought out in a budget negotiation two weeks before an election.
Because in a budget negotiation the winner is whoever has leverage, not whoever has the better argument.
Leverage is a function of timing, and timing is a function of the Numerator rule, which is a function of an accounting convention nobody outside the Finance Ministry has ever heard of.
The entire business turns on a rule about when money has to be named.
It turns on a rule about when money has to be named. A billion dollars of capability is decided by a cash-flow convention.
There's a broader observation in that, which is that the most consequential parts of procurement are almost never the parts the public sees.
The public sees the price tag and the platform. The decisions that matter are the requirement document, the quantity, the funding mechanism, and the schedule. All four of those are technical, and all four of them get made by people whose names nobody knows.
The platform gets the headlines because it's the thing you can photograph.
You can photograph a helicopter. You cannot photograph a Numerator rule.
Which is why this is a good episode for the show. The interesting machinery is all invisible.
It's invisible by default, not by conspiracy. It's just not the kind of thing that makes a headline.
I'd like to test the capability-based planning idea against the Israeli case once more before we move on.
Go ahead.
If capability-based planning means buying things useful across a wide range of scenarios, then the Apache has to justify itself against that standard. A helicopter is useful in some scenarios and useless in others. A tank is the same. A drone is the same. Nothing is universally useful.
The whole point of CBP is not to find the universally useful platform, because it doesn't exist. It's to build a portfolio that covers the range without over-investing in any single scenario.
The right question isn't is the Apache obsolete. It's does the portfolio still need a manned attack helicopter in it, given what else is in the portfolio.
Which is a much harder question to answer, and the Israeli Air Force appears to have answered it yes, and the Finance Ministry's professionals appear to have answered it no, and both answers are defensible.
The process that should have resolved it was a classified requirement document that we can't see.
That's the frustration. The disagreement got surfaced to the public, and the analysis that should have settled it stayed secret. So we get the argument without the evidence.
The political layer gets to pick without being accountable to the analysis, because the analysis is classified.
Which is arguably the strongest argument against classifying requirements this way, at least when the money is being moved from somewhere visible.
The Arab community funding is what makes this visible. If the billion had come out of a contingency line, this would have been a page-three story.
It would. And the fact that it came out of a named development plan is the reason we're discussing it at all.
Which raises a question I don't have a clean answer to. Should the source of the funds change how we judge the decision?
I think it changes the politics of it and it doesn't change the military analysis. The question of whether Israel needs twelve Apaches is the same question regardless of where the money comes from. The question of whether this is the right thing to spend this particular billion on is a different question, and it's political.
We have two questions bundled together and the public conversation keeps swapping between them.
Constantly. And that's not dishonesty, it's just that both questions are real and they're both being asked at once.
The JCIDS process Daniel asked about is designed to answer the first question cleanly so that the second question can be argued on its own terms.
That's the aspiration. And it gets you part of the way there. What it can't do is make the political choice go away.
Because the choice is legitimate.
Somebody has to decide whether the billion goes to helicopters or to roads. That's a real decision and there is no analytical answer to it.
The analysis does the work it's supposed to do, and the politics does the work it's supposed to do, and the failure mode is when the two collapse together and nobody can tell which one made the call.
In this case they collapsed, because the legality of the allocation is itself being questioned.
That's the sharpest thing you've said.
I'll take it.
Daniel asked a version of this and I want to make sure we answer it directly. If rigorous analysis sets the menu but politics picks the dish, is the analysis doing any work at all.
My answer is yes, and here's why. Without the analysis, the political actor picks from a menu that doesn't exist, which means the pick is unaccountable. With the analysis, the pick is a choice between defined options with defined costs, and someone can ask why this one rather than that one.
The analysis is doing accountability work rather than decision work.
Most of the time. And occasionally it does decision work, when the analysis says something is impossible, or when it says a requirement is already met by something else.
Which is the case that the Finance Ministry is arguing. That a cheaper option covers the requirement.
The Air Force is arguing the opposite, that the cheaper option doesn't cover the specific thing. And the document that would tell us which one is right is classified.
The process works, and the transparency doesn't.
That's the cleanest way to say it. The process is real, and it's opaque.
A deal that survives a two-year staff study, a Nagel committee, and agreement across three institutions, and then dies over an accounting rule, and gets revived by moving money out of a development plan for a fifth of the population. That's not a story about analysis. That's a story about plumbing.
Which is the part everyone ignores.
Herman, when you were describing the squadron-sized buy, you talked about establishment like it was fixed. It isn't. It depends on the establishment the last war produced.
My brother-in-law did freight for a defense outfit in Haifa in the late seventies. He's not reliable on anything you could look up, but he's reliable on how numbers get picked. They ordered seventy-two of a thing because the storage shed took seventy-two crates and the shed was already there.
The shed set the quantity.
The operational requirement made it in as a footnote. Then the shed burned down six weeks before delivery and they had to rent a field in Ramla for eleven months. Seventy-two crates in a field with a tarp over them.
What was in the crates?
Cables.
Seventy-two crates of cables.
They had enough cable for thirty years and eventually sold the surplus to a kibbutz that needed irrigation line. I don't know how the requirement got signed off. I know what the shed took.
That's actually a cleaner illustration of the point than the DOTMLPF discussion.
It's not a lesson, Herman. It's a shed.
What happened to the field.
Nothing. It's a field. The tarp went into the shed that replaced the shed.
The number was never a requirement.
The number was a shed.
There's something useful buried in that. When you read twelve in a press release, someone at some point chose twelve, and the reason may be so mundane it never makes the file.
That mundane reason can outlast the entire analytical process, because the analysis produces a range and the number comes from outside it.
A range is defensible, a number is procurable.
That's the lot. The analysis can tell you that somewhere between eight and twenty is reasonable. It cannot tell you fourteen versus sixteen. Something else has to close that gap.
A factory slot. A training pipeline. A shed.
In the Israeli case, twelve reflects a squadron-sized buy and a favorable Boeing production slot. That's the public framing. And I suspect the real answer sits inside a classified document that nobody outside the ministry will ever read.
Which takes us back to the last question Daniel asked directly. How many of these decisions come down to analysis and how many come down to something else entirely.
My honest answer is that the analysis narrows the field, and then something else closes it. That something else might be legitimate, like doctrine. It might be structural, like a shed or a slot. It might be political, like the election calendar. But it's always there, and anyone who tells you the number is an output of the analysis is overselling the analysis.
In this case the something else was a Numerator rule and a vote eleven days before an election.
Which is why the deal fell apart and came back. The requirement never changed. The closing mechanism did.
The requirement was real enough to survive two years of staff work and the Nagel committee, and then it turned on two hundred and eighty-eight million shekels of accounting, and it was saved by taking money from a five-year plan for Arab communities. What do we make of that.
I think it means the analysis was doing its job and the plumbing was not. The requirement had been established. The disagreement about doctrine was real and legitimate and was being worked through. What broke was the mechanism for turning a decision into a contract.
The mechanism is invisible. A hundred and forty-three billion shekel defense budget, seventeen percent of total spending, and the deal turns on a rule about when a shekel has to be named.
Somebody has to decide whether a billion goes to helicopters or to roads. That's a real decision. The problem is that this one was decided by a rule that was never meant to decide anything.
The analysis narrows the field. Something else closes it.
The something else is usually the least interesting part of the system, and it's the piece that ends up deciding.
One last thing worth holding onto. There is no public Israeli equivalent of a JCIDS document. The process is real, and it's opaque. So we have the argument without the evidence, and we're judging the outcome without seeing the work.
Which is the opposite of what the American system is designed to produce, and a reminder that the most elaborate rulebook in the world doesn't help you if nobody publishes the result.
The one thing I'd take from this. When you read a number like twelve in a headline, that number was closed by something, and the something was probably not the analysis. The analysis produced a range. A shed, a slot, or a deadline picked the number.
The moment you start asking what closed the number, you understand procurement better than most of the coverage does.
Thanks to Hilbert Flumingtop, who has been producing this from a desk the whole time, and who has now told us about a shed we will not be able to forget.
For more along these lines, there's episode four seventy-four, The Price of Autonomy; episode fourteen oh eight, Israel Wartime Readiness Field Guide; and episode five forty-eight, Israel's Security Tax. This has been My Weird Prompts.
If you want to say something about this, leave us a review on whatever app you're listening on. And if you want to send us a prompt, we take them on Telegram at t dot me slash MWP listener bot.
We'll be back soon.
See you tomorrow.
