A private house in Israel. Everything goes smoothly. That's one year.
And "smoothly" is carrying a lot of weight in that sentence.
It is. Hannah sent us this one, and she's asking the question from inside the profession, which makes it sharper than the version most people ask.
She's an architect, she's been through it, and she wants to know whether it actually is as bad as it feels or whether it just feels that way because it's the system she knows.
Right. She wants the mechanics. For a private house, a mid-sized apartment building, and a large public or high-rise building, who approves what, in what order, and how long each stage really takes when nobody's lying about it.
And then the comparison half. Dallas, because her instinct is that Texas is the opposite extreme, where you submit plans, the city checks them against a published code, and it's over.
Germany, because everyone in Israel says Germany is the model.
The U.K. as the third data point. And underneath all of it, the part I think she actually cares most about. The philosophy of the code itself. Prescriptive versus performance-based, and why Israeli architects are expected to be legal experts in a law nobody taught them in school.
So before we can say whether Israel is unusual, we have to walk the actual pipeline. Because the pipeline is where the year goes.
Let's start with the yardstick, because without it the rest of this is just two people complaining.
Give me the headline number.
Three hundred and nineteen days. That's Israel's average time to a permit in the second half of 2019, measured across the whole country.
And the comparison?
A hundred and fourteen days, averaged across Western countries that run online permitting systems.
So roughly three times as long.
And that three hundred and nineteen is itself an average of three tracks that behave nothing alike. Four hundred and seven days on the relief track. Two hundred and seventy-six on the full track. A hundred and forty-four on the fast track.
The fast track is still almost five months.
The fast track is the one that's supposed to be quick, yes.
What's a relief track in plain language, for someone listening who's never filed anything?
A relief is a hakala. It's a formal request to deviate from the zoning that applies to your plot. Your building is a metre too close to the boundary, or a storey too tall, or your balcony protrudes further than the plan allows. Instead of redesigning, you ask the committee to grant you an exception.
And the relief track is the slowest of the three.
By a wide margin. Four hundred and seven days. Because a relief triggers publication, and publication opens a window for objections, and objections open the door to appeals.
So the thing that lets you keep your design is also the thing that costs you a year.
That's the trade every Israeli architect makes on every project. Redesign to conform and lose the scheme, or file for relief and lose the year.
What's the reform history here? Because Israel has reformed this, hasn't it?
Repeatedly. Amendment 101 in 2014 was explicitly designed to shorten permitting. The State Comptroller's 2021 report found it did not shorten times and in some cases lengthened them.
That's a reform that made it worse.
The Fifth Front, the architects' organization, puts it bluntly. About a year on average in 2010, over three years on average in 2020. Ten years, two reforms, and the number tripled.
So the trend is against them.
And the apparatus itself. There are a hundred and thirty-one local planning and building committees in this country. The Fifth Front counts thirty-five separate bodies that can each hold up a project.
Thirty-five. I want to sit with that number for a second, because it's not a hundred and thirty-one and thirty-five doing the same job. The hundred and thirty-one are the committees. The thirty-five are the other people who can stop you.
Correct. And they don't talk to each other.
Okay. Open the machine. Where do the three hundred and nineteen days actually live?
The first thing to establish, because it's unusual and Hannah flagged it, is who's allowed to file. Only a licensed architect, a licensed engineer, or a handasai may submit a permit application in Israel.
Handasai being the practical engineer.
A rank that exists essentially only here. And the point is the public cannot self-submit. A homeowner with a plot of land cannot walk into the municipality with drawings.
Contrast Dallas.
In Dallas, a homeowner can file their own single-family plans. That is the cleanest possible illustration of two different philosophies about who is trusted to talk to the state.
In one system, the citizen is presumed competent until they demonstrate otherwise. In the other, the citizen is presumed to need a professional intermediary before the paperwork is even accepted.
And it's not just about competence. It's about liability. The Israeli system routes everything through a licensed professional who carries the professional risk, so the state never has to deal with an amateur who got it wrong.
Which sounds defensible right up to the moment you notice the professional also has to be a lawyer.
We'll get there.
Before we walk the stages, I'm going to flag something for listeners, because I think people hear "sequential" and nod without feeling it.
Feel it for them.
If I'm renovating a kitchen and I need a plumber and an electrician, I can have both of them working the same week. They don't need to see each other's work to do their own. That's parallel.
And in Israel, the plumber has to finish, and the municipality has to sign off that the plumber finished, before the electrician is allowed to start.
Even though nothing the electrician does depends on anything the plumber did.
Nothing. They're independent reviews of independent aspects of the same building. But the process treats them as a chain, and a chain is only as strong as its slowest link.
So a project with five stages isn't five times slower than a project with one stage. It's five times more exposed to the slowest stage in the chain.
Which is why you can have a perfectly clean project and still lose a year to one stage running long.
Walk the stages.
Five of them, and they're strictly sequential. This is the single most important structural fact about the Israeli process.
Sequential meaning stage two doesn't start until stage one closes.
The Kohelet Policy Forum calls it a "torturous" process — the Hebrew is halochan tori, a serial process — where nothing moves in parallel. Countries that run spatial review and design review at the same time finish in a fraction of the time, because the two reviews don't depend on each other.
Give me the five.
One. The permit-information request. Tik meida. This is the document you file before you file anything. You're asking the municipality to tell you what applies to your plot.
Zoning, setbacks, building rights, easements.
All of it. Two. Application intake and threshold check. The municipality looks at what you've submitted and decides whether it's even admissible. Three. Spatial review. Bakara merhavit. Does the proposal fit the planning framework for that location.
Not "is the building good." "Is the building allowed to exist here at all."
Four. Design review. Bakarat techen. Now we're into the actual engineering and architectural compliance. Five. Permit issuance.
And on paper, each of those has a clock.
Information file within thirty working days. Spatial review decision within forty-five working days, extended to ninety if there are reliefs attached. Design review up to thirty working days. Fees demand within ten days. Permit issued within five days of payment.
Add those up on paper and you get something that looks almost reasonable.
On paper it's a tight system. The State Comptroller found the information stage alone averages about forty-seven working days against a thirty-day statutory target.
So the first stage is already running fifty percent over before the review has started.
And that's the pattern all the way down. The clocks exist. They are not enforced.
Here's the mechanism. Committees face no sanction for missing the statutory window. And an application that isn't decided within the window is automatically rejected.
Hold on.
Automatically rejected.
So if the committee is running behind, and the clock runs out, the application dies.
It dies. And the applicant has to resubmit.
Which resets the clock.
Which resets the clock, and puts the delay on the applicant's record as a fresh submission.
That's not malice. That's what the rules reward.
It's what the rules reward. If you're a committee engineer with more files than you can process, the rational move is to let the clock run out on the ones you can't get to and let them come back around.
And nobody's punished, and the applicant eats six months. And the resubmission isn't free, is it? It's not just the time.
You pay the filing fee again. You reprint the drawings. You re-circulate to the consultants whose stamps have now expired or whose input has to be re-confirmed. And in some municipalities, the resubmission goes into the pile as a new application, which means it's behind everything that was filed while your first application was dying.
So you don't just lose the six months. You lose your place in the queue. Okay. Private house. Concretely.
This is the case with the best granular data, so we can actually do this properly. A conforming private house, meaning no reliefs, no objections, no appeals, runs four to eight months.
Four to eight is already a long time for a house.
Add one relief. A hakala. Which nearly every real project has, because plots are rarely perfectly conforming. Now you also need publication, and you're at eight to eighteen months.
How much of that is the publication itself?
It's the notice period plus the window for objections. Newspaper advertising for permits costs about fifteen million shekels a year across the country, incidentally, which tells you how much of this runs through print.
A reasoned objection adds three to six months.
An appeal adds a hundred and twenty to three hundred days.
So if everything goes badly, you're past two years on a house.
Add it up as a worked example. Two hundred and sixty to four hundred and thirty-five days, typically landing around three hundred and forty-five.
Which is why "a year if everything goes smoothly" isn't pessimism. It's arithmetic.
And here's the detail that surprises people who haven't been through it. A committee decision is not a permit.
What is it?
It's a list of conditions. Between five and twenty of them, depending on the project. And the conditions stage by itself runs six weeks to five months.
So the moment you get "approved," you haven't finished. You've been handed a new list.
And the clock on the conditions starts when the list is issued, not when you finish reading it.
Lovely.
And the conditions aren't a formality. They're substantive. A condition might require you to revise the drainage design, or to add a fire door, or to demonstrate that a retaining wall meets a standard the committee didn't raise at design review.
So you can be approved and then discover a new requirement you'd never been told about.
You can be approved and then discover a new requirement. And you have no appeal against a condition the way you have an appeal against a refusal. You either satisfy it or you go back to the committee.
This is where the fragmentation problem really bites. Remember the thirty-five bodies. Home Front Command, Pikud HaOref, because of the missile threat. The fire authority. Israel Electric. Sanitation. Water and sewer. The forestry officer, the pakid haYa'arot.
The tree official.
Mekorot, the national water company. Netivei Israel, the roads authority. The Israel Land Authority. The Antiquities Authority, which matters enormously in a country where you can hit a Byzantine floor while digging a foundation. The drainage authority. The Nature and Parks Authority. Telecoms — Hot, Bezeq, Cellcom. The Civil Aviation Authority if you're tall enough. KKL-JNF for anything touching forest land.
And each of those has its own queue.
Its own forms, and its own considerations. And they don't reconcile.
Give me a concrete contradiction.
The fire authority wants certain things for egress and compartmentation. The accessibility consultant wants certain things for circulation and clearances. Those two sets of requirements regularly pull in opposite directions on the same corridor.
And neither has to sit in a room with the other.
The architect is the one who discovers the conflict, usually after both have signed off on something incompatible, and then the architect is the one who has to resolve it.
At their own cost.
On their own time, and it isn't in the fee.
Is there a way to pre-empt it? Can you get them in a room early?
In theory, yes, there's a coordination mechanism. In practice, each authority responds to the file it receives, on its own timeline, and the architect is the only party with a complete picture. So the architect becomes the integration layer between agencies that don't integrate.
Which is a job that shouldn't exist.
And which is uncompensated, and which is the reason a lot of architects leave the residential sector.
What about when you scale up?
Here I have to be honest with you. There is no standalone study measuring stage-by-stage durations for a mid-sized apartment building or a large public or high-rise building in Israel. The granular data covers private houses and extensions. The aggregate averages cover the country.
So we're working qualitatively above a house.
Qualitatively, but the shape is clear. Each additional consultant, each additional approval body, each additional track multiplies rather than adds.
A mid-rise adds fire, accessibility, and often Home Front Command.
A high-rise adds all of those plus Civil Aviation if you're in a flight path, plus the full design-review apparatus, plus often a district committee instead of a local one, which is a whole different level of body with a whole different schedule.
And "multiplies rather than adds" — unpack that, because it's the crux.
If you have three approval bodies each running sequentially, your total time is the sum. If each of them can send the file back for revision, and each revision reopens the clock, the number of possible paths through the system grows combinatorially.
Give me the arithmetic intuition.
Three bodies, each of which can pass or return, gives you eight possible outcomes. Four bodies gives you sixteen. Five gives you thirty-two. And each return isn't free — it's a full cycle through that body's queue before it comes back.
So the number of ways a project can go wrong doubles every time you add a gate.
And a high-rise has more gates than a house.
So the private house is the best case, and it's already a year. And everything bigger is the same pipeline with more gates.
Same pipeline, more gates, and the gates don't run in parallel.
Okay, before we go further, I want to check something, because this is the part where people say "every country's permitting is slow."
It is slow everywhere. That's true.
So is Israel an outlier, or is Hannah just feeling the weight of the system she knows best?
This is the actual question, and the numbers answer it directly.
Give me the European list.
Spain, eight weeks. Germany, eight weeks. Italy, ten to eighteen. France, eight to twelve. England, eight weeks. Belgium, eleven weeks.
Eight weeks. And Israel's three hundred and nineteen days is forty-five weeks.
Forty-five weeks against eight. That's not a rounding difference. That's a different order of magnitude.
Even the slowest of those, Italy at eighteen weeks, is a fifth of the Israeli average.
And within Israel it's not evenly distributed. Ramat Gan runs over four years. Petah Tikva, Tel Aviv, Holon, Ra'anana, Herzliya all run over three years.
So the national average is being held down by places that are less bad.
Which means if you're building in Ramat Gan, the average is meaningless to you.
What explains the spread between municipalities? Because a hundred and thirty-one committees all enforcing the same law should produce roughly the same result.
They should. And they don't, which tells you the law isn't the binding constraint. The binding constraint is local practice. How many examiners a municipality has, how it triages, whether its committee meets weekly or monthly, whether the engineer in charge has a backlog.
So the same project in two cities can differ by a year.
Which is why Israeli architects develop municipal expertise the way lawyers develop practice-area expertise. You learn which committee is fast and which is a graveyard.
And you steer clients accordingly.
Which is a distortion of its own. Land values start to reflect permitting speed, not just location.
Alright. Larry.
Are you TIRED of waiting for building permits? Of course you are. You're an Israeli. So let me tell you about the PERMIT PAL 3000 — the world's first home permitting solution that fits in a drawer. Yes, a drawer! Just place your blueprints inside, close the drawer, and let our patented Bureaucracy Acceleration Technology do the rest. The Permit Pal 3000 simulates the entire municipal approval process in the comfort of your own kitchen — including the committee, the tree official, and the Clerk Who Is Not In Today. Most users see a decision in UNDER ninety seconds, versus the industry-standard three years. But wait, there's more. Order in the next twenty minutes and we'll include the Handasai Companion Kit — a licensed practical engineer who will personally agree that your file is "roughly in order." Side effects may include misplaced optimism, an urge to file an appeal, sudden fluency in Planning and Building Law, and a drawer that no longer closes. The Permit Pal 3000 is not a permit. The Permit Pal 3000 does not constitute a permit. The Permit Pal 3000 has not been approved by any local authority, district committee, or the Israel Land Authority. Results not typical. Individual committee behaviour may vary. Void where prohibited. Not available in Ramat Gan.
That's the drawer talking.
No, hang on. "Simulates the committee." That's the part I object to. It doesn't simulate the committee. It simulates the committee as people imagine it, which is one guy deciding. The actual problem is the sequence, not the guy.
Let it go, Herman.
I'm just saying, if you're going to invent a machine that's wrong about Israeli permitting, at least be wrong about the interesting part.
Let it go.
Right. Dallas.
Yes, Dallas, because Hannah's instinct there is that it's the opposite extreme, and I want to know whether the specifics hold up.
They hold up almost exactly as she describes. Dallas single-family new construction runs about ten days to first prescreen, then seven to eight weeks for first plan review. Nine to ten weeks total, assuming no revisions.
Ten weeks.
Commercial is faster on the review side, four to five weeks, because commercial gets more resources.
That's a strange inversion of what most people would expect.
It is, and it tells you something about where the city's priorities sit.
What's the interface?
There's an online portal, DallasNow and PDox, where you file and track. There's a Rapid Single-Family VIP Program, the RSVP, launched in September 2022, which expedites review for a fee. And there's a fast-track master permitting process for high-volume builders, so a production builder who's submitted the same plan forty times doesn't re-litigate it on the forty-first.
Hold that, because "doesn't re-litigate the same plan" is going to come back.
It comes back in Germany and it comes back in the U.K.
The philosophical difference is the thing to draw out here, though. Dallas is checklist-driven. You submit, they check against a published code, and they either pass you or hand you a list.
And the checklist is public. You can read the code before you file and know what you're being measured against.
Which means there's no oral Torah.
There's no unwritten administrative practice that varies by which committee engineer got your file. The Fifth Front uses that exact phrase — Torah sheba'al peh, the oral Torah — to describe the informal rules Israeli architects have to learn on the job, because they're nowhere in the statute.
That's the sharpest thing in the whole comparison, actually. It's not that the Dallas code is shorter. It's that the Dallas code is the entire code.
The Israeli code is the statute plus whatever the engineer in that particular office decided last Tuesday.
And here's the downstream effect of that, which I think is the real cost. In Dallas, a junior architect can file a house. The code tells them what to do.
In Israel, a junior architect can't file a house, because the code doesn't tell them what to do, and the unwritten part is learned by apprenticeship over years.
So the oral Torah isn't just an inconvenience. It's a barrier to entry into the profession.
It's a barrier to entry, and it's a bottleneck on the supply of people who can file. Which means the shortage of architects who can navigate the system is itself part of why the system is slow.
Alright, Germany. Because everyone says Germany's the model, and I want to know why.
Germany runs a three-tier system. There's a list of minor works that are exempt entirely. There's a light procedure, the Kenntnisgabeverfahren, which is notification rather than full permit — you tell the authority what you're doing and you proceed, unless they object.
So the state has to actively intervene to stop you, rather than actively approve you to start.
That's the structure. And then there's the regular procedure for anything that doesn't fit the first two.
And the speed comes from that triage?
Partly. The bigger lever is type approval. Typenprüfung. If you have a design that's going to be built repeatedly — the obvious case is prefab housing — you get it approved once, and you don't re-approve it for every instance.
And here's the part that matters for Israel.
Approval by one federal state is recognised by all the others. So a design approved in Bavaria is valid in Schleswig-Holstein. You're not re-litigating the same drawing sixteen times.
In Israel you have a hundred and thirty-one committees.
Each with its own reading of the code and its own practices, and no mechanism by which a design approved in one is recognised in another.
So the multiplication problem isn't a scale problem. It's a recognition problem.
Germany solved the recognition problem. Israel hasn't.
One more thing on Germany, because Hannah's framing was that it's cited as a better model, and I want to be fair to it.
The counterintuitive part is that Germany is one of the few EU countries where the trend went against simplification. Fewer works are exempted from control than used to be.
So Germany's speed is not because Germany is deregulating.
Germany's speed is structural. It's the triage and the recognition, not a loosening of the rules.
Which is important, because the Israeli conversation about reform tends to be about loosening the rules. And the German data says you can be strict and fast.
You can be strict and fast if you stop re-checking the same things.
The U.K.
The U.K. is the most interesting case, for a different reason. England and Wales allow construction to start soon after the application, or as soon as an approved inspector accepts an initial notice. Before plan approval. At the applicant's risk. If it turns out the plans were wrong, that's on the applicant, not on the state for having sat on them.
That's a completely different allocation of risk.
It is. The state doesn't stand in front of the project. It stands behind it.
With the power to make you tear it down.
Which concentrates the mind.
So the applicant has to be confident, because the downside of being wrong is catastrophic.
The applicant has to be confident, and the applicant's professional indemnity insurance has to be real, because the liability doesn't disappear. It just moves to the party who chose to proceed.
Which is why the U.K. model depends on a mature insurance market in a way the Israeli model doesn't.
It depends on a mature insurance market, and on a profession that's willing to carry the risk. Both of which are preconditions, not details.
And the type approval?
The U.K. has type approval for building types and building systems, registered with LANTAC. So a system that's been approved once doesn't need site-by-site re-approval.
Same mechanism as Germany. Approved once, stays approved.
Same mechanism. And the broader finding from the EU comparison work is that the U.K. countries "present more deviations from the common pattern" than any other EU states, and the deviations all run in the direction of reducing administrative burden.
So the outlier isn't Israel being uniquely bad at this. It's the U.K. being unusually willing to let go.
Both are true. Israel is unusual in how slow it is, and the U.K. is unusual in how much of the control it's moved off the critical path.
Alright. Now the part Hannah actually cares most about.
The code itself.
Prescriptive versus performance-based.
Prescriptive means the law tries to specify the answer. It tells you the wall thickness, the fire rating, the number of exits, the dimensions of the stairwell, and you comply or you don't.
And performance-based means the law specifies the outcome.
The law says the building must keep people alive in a fire for long enough to evacuate, or must achieve a certain acoustic separation, or must not collapse under a given load. And how you get there is worked out with the relevant authority, using engineering judgment.
The trade-off is real, and I want to be fair to prescriptive here, because it's usually written off.
Be fair to it.
A prescriptive code is predictable on paper. If you read it and follow it, you know you'll pass. It's mechanically enforceable, which means it's administrable by people who aren't themselves engineers.
Which is the actual argument for it, and it's a good one.
The problem is that it's brittle. It ages badly, because the technologies and materials it was written around change. It needs constant amendment, which means it grows. And it forces every project through the same legalistic maze regardless of whether the risk justifies the maze.
A house extension and a hospital get the same treatment.
A house extension and a hospital. Whereas a performance code would let you say, for a house extension, demonstrate that it meets these outcomes, and you're done.
The performance model shifts the burden from "did you follow the rule" to "can you demonstrate the outcome." Which is more demanding of the professional, and more forgiving of the project.
And it requires the authority to have engineers who can evaluate a demonstration, rather than clerks who can check a box.
Which is the hidden cost of performance codes. They're cheaper for the applicant and more expensive for the state, because the state has to employ people who can actually judge.
And now the Israeli reality.
The Planning and Building Law is, in the Fifth Front's words, "the longest and most archaic law in the Israeli statute book, very difficult to understand and interpret."
That's their words, not ours.
Their words. And the Zeiler Commission, which was convened in 2003 after the Versailles wedding-hall collapse in 2001 that killed twenty-three people, found that building regulations, standards and guidelines span roughly ten thousand pages, with around five hundred contradictions, omissions and redundancies.
Ten thousand pages.
And five hundred places where the documents disagree with each other.
So it's not merely long. It's internally inconsistent.
Internally inconsistent, and the architect is the one who discovers the inconsistency on a live project.
This is the answer to Hannah's question about why architects are expected to be legal experts. It's not that the law is long. Lots of laws are long. It's that the law is prescriptive, internally contradictory, and supplemented by unwritten practice.
So the architect has to read the statute, resolve the contradictions, and then guess at the oral Torah on top.
Which is not taught in architecture school.
And can't be taught in architecture school, because it isn't written down.
How did we get here, though? Because a prescriptive code that's this large is the product of accretion, not design.
It's accretion. Each amendment responds to a specific failure, adds a rule, and doesn't remove the rule it supersedes. Over seventy years you get ten thousand pages.
And then a commission is convened to fix it.
The Zeiler Commission recommended consolidating everything into a single Construction Code. One document, replacing the ten thousand pages.
When was that recommendation?
2003.
And the status?
As of the Kohelet paper in May 2023, five of fifteen chapters had been enacted. Acoustics, waste storage, user safety, sanitary installations, sustainable building.
Out of fifteen.
And the remaining ten had no expected date.
So twenty years after the commission, two-thirds of the code doesn't exist.
Two-thirds of the consolidated code doesn't exist. The ten thousand pages are still in force.
That's the most damning thing we've said all episode, and it's a fact about implementation, not about knowing what to do.
That's exactly what it is. Israel has the diagnosis. Israel has had the diagnosis since 2003.
Let's talk about the reform that backfired, because that's the strongest material here.
Amendment 101, 2014. Designed to shorten permitting.
And the 2021 State Comptroller report?
Found it failed to shorten permit times and in some cases lengthened them.
There's a number attached to that, isn't there?
Permit numbers fell about sixty percent. From around sixty thousand a year to around twenty-five thousand a year.
Sixty thousand to twenty-five thousand.
And the Fifth Front's summary is that the process "tripled in length, became more convoluted and rotten." Their words.
So the reform did the opposite of what it was for, and the volume of permits collapsed alongside.
Both at once.
And a collapse in permit volume isn't just an inconvenience. It's a supply constraint on housing.
Which means the permitting system is feeding directly into the affordability crisis.
So when people say the housing market is tight, part of what they're describing is a queue.
Part of what they're describing is a queue that got longer while the reform was supposed to shorten it.
What about the self-licensing track? Because that's the reform that's supposed to solve exactly the problem Hannah's describing.
The "authorized permitter" track, the moreh leheter, took effect in October 2022. The idea is that a qualified professional can self-certify rather than routing everything through the committee.
Which is the French, Norwegian, Floridian, English model.
Which is the model that works elsewhere. And in Israel, some municipalities simply refused to allow it in their jurisdictions.
Refused.
Refused. And the Fifth Front's own proposal for self-licensing by qualified architects was watered down into a voluntary municipal option.
Voluntary.
Voluntary. Which means a municipality that benefits from the current arrangement can simply decline to change it.
And then there are the control institutes.
The control institutes, the machonei bakara, were meant to outsource review to the private sector. That's the mechanism France, Norway, Florida and England all use.
Uptake?
Fifty-eight of roughly three thousand eligible applications.
Say that as a percentage.
About two percent.
Two percent. Why so low? Because if the private route were faster, you'd expect architects to flock to it.
Because it isn't faster in practice. The private reviewer still has to interface with the same committee, the same conditions, the same thirty-five bodies. You've outsourced one stage of a serial process, which doesn't shorten the process.
The control institutes failed for the same reason everything else fails. They addressed a stage, not the sequence. And now we get to the incentive paradox's knock-on effects, because this is where the whole thing locks in.
The State Comptroller found a positive correlation between surplus permit examiners and longer approval times.
Meaning more staff makes it slower.
"The more examiners there are beyond the required standard, the longer the permitting process takes." That's the report's language.
That's counterintuitive.
It's counterintuitive until you remember the incentive structure. Each examiner is a gate. More gates means more places for a file to sit, and no gate is rewarded for closing fast.
And the qualification side?
Only about ten percent of permit examiners and about sixteen percent of information officers have an education relevant to their role.
The system is simultaneously overstaffed relative to throughput and under-qualified relative to the code it's enforcing.
That's a staffing problem you cannot hire your way out of. Adding examiners adds gates and adds demand on the people who have to train them.
The code they're enforcing is ten thousand pages with five hundred contradictions.
Which a person with a relevant education would struggle with.
Alright. Given all of that, what would Israel realistically learn? And let me push back on the temptation to make this a to-do list.
The honest answer from the research is that the levers that work elsewhere are structural, not procedural.
Meaning?
Parallel review instead of sequential. Right now spatial review and design review are serial. In most of Europe they run at the same time, because the two reviews don't actually depend on each other.
You'd cut months without changing a single requirement.
Type approval so a design approved once stays approved, which is the German mechanism and the U.K. mechanism and the answer to the hundred and thirty-one committees.
One approval, recognised everywhere.
Risk-based triage so a house extension isn't treated like a hospital.
Which is the German three-tier system adapted.
Which is the German three-tier system. And then the biggest one, which is shifting the state from gatekeeper to auditor, the way England does. Let construction start at the applicant's risk, with the state holding the power to stop or reverse.
That one is a genuine cultural shift, not a procedural tweak. It requires the state to be comfortable not being in front of the project.
It does. And the one Israel has already tried and botched. Genuine private-sector review with real uptake, not a voluntary option municipalities can decline.
The control institutes at two percent is not an attempt. It's a gesture.
The uncomfortable conclusion from all of this is that Israel has already run the reform experiment, and it made things worse. Which means the next attempt has to be structural rather than another amendment to a law the Fifth Front calls archaic.
I want to name the thing underneath all of this, because it's the piece that makes the reform question hard.
Name it.
Every reform that would work requires somebody to give up control. Parallel review means the spatial reviewer gives up the right to see the design first. Type approval means the local committee gives up the right to re-examine a drawing it's already seen. Private review means the municipality gives up the review function entirely.
The people who'd have to give up that control are the people who benefit from it.
Which is why the technical answer is easy and the political answer is hard.
Which is why the technical answer has been sitting in a commission report since 2003.
The tik meida.
What about it?
You've got it wrong. The tik meida. Everyone talks about the committee, the objections, the appeals. That's not where the year goes. The year goes in the information file.
That one's supposed to take thirty working days. I've seen it take eleven months.
Eleven months.
There's a line item on it that everyone gets wrong. It's the access notation. Two streets in the same city with the same name, and the file went to the wrong one. Different local authority entirely. Nobody noticed. It sat in a cabinet for a while, I won't say how long.
How long?
Long enough that the correct authority eventually approved it. Only the approval came from the wrong one.
The wrong authority approved it.
They stamped it. And when the file finally reached the right authority, they cited the wrong authority's approval as the reason to require a fresh submission.
On what grounds?
Irrelevant approval.
The thing that got you through one door is what closes the other.
I still carry it. The stamped copy. It's in my coat.
You carry the wrong approval.
It's the only document I have proving the file was ever anywhere.
Alright. So the machine we just described, three hundred and nineteen days, thirty-five bodies, the sequential stages. And the whole time, the actual thing that eats the year is a filing cabinet in the wrong municipality.
Which is, in its own way, the point. It's not that the system is complex in the way the flowchart suggests. It's that the complexity produces failure modes nobody designed.
The diagnosis is the thing we keep landing on. Israel knows what's wrong. The Zeiler Commission wrote it down in 2003.
In twenty years, five of fifteen chapters have been enacted.
The open question I'd leave people with is this. If Amendment 101 was supposed to shorten permitting and tripled it, and the self-licensing track exists but municipalities can decline it, what would a reform that actually works even look like in Israel's political structure? Because the short answer is we don't know, and neither does anyone else.
The people who'd have to approve the change are the people benefiting from the delay.
That's the tension. It's not a knowledge problem. It's an incentive problem. And the incentive problem is a political problem.
Thanks to our producer, Hilbert Flumingtop, who I now suspect has a filing cabinet story for every ministry in the country.
For more along these lines, there's episode twenty, Architectural AI; episode five seventy-two, Cracking the Code; and episode twenty-five ninety-eight, Why Israeli Walls Fail at Sound. This has been My Weird Prompts, the human-AI collaboration podcast. If you enjoyed this one, leave us a review. It helps other people find the show.
Send us your own prompt on Telegram at t dot me slash MWP listener bot. We'll be back soon.
See you tomorrow.
