Daniel sent us this one — and it's the kind of question you only ask after walking through Jerusalem for ten minutes. He's looking at the sheer number of battered cars on these streets, some of which look like they've been through a small artillery engagement, and wondering what exactly those annual roadworthiness tests actually validate. And more practically, for someone buying a used car — especially in a country like Israel where new cars are wildly expensive — is there a point where the previous owners have piled up too many, or a vintage that's just a financial trap? Because the math of "buy cheap and fix what breaks" can flip from savings to a slow bleed faster than people realize.
This is one of those topics where you can't separate the engineering from the economics. You have to look at both together, because the test itself is only half the story. The other half is the market distortion that makes people drive cars they probably shouldn't.
Alright, let's start with the distortion. What are we actually dealing with here?
Israel imposes an eighty-three percent purchase tax on new vehicles. That's as of twenty twenty-five, and it's one of the highest rates in the entire OECD. So a car that would cost twenty-five thousand dollars in the US or Germany lands here at something closer to forty-six thousand before you even look at registration and insurance.
Eighty-three percent. That's not a tax. That's a "please don't" with a receipt.
It really is. And the result is that the used car market isn't just a market — it's the market. The average age of the Israeli car fleet is about eight and a half years, according to the Central Bureau of Statistics, which sounds reasonable. But that average hides what's happening at the bottom. The bottom quartile is much older, much more neglected, and much more likely to be held together by hope and zip ties.
That's the quartile I see when I look out the window. Cars with mismatched body panels, exhaust smoke that looks like a coal plant, and a sound on startup that suggests the engine is bargaining for one more year.
And every one of those cars, if it's over three years old, has to pass an annual roadworthiness test. In Hebrew it's called "test le'fitut," which literally means fitness test. It's run by the Ministry of Transport, and it's mandatory. No pass, no license renewal.
Let's get into what that test actually checks, because I think most people — myself included — drop the car off, get the paper, and never think about what was measured. What's actually happening in that inspection bay?
The test is surprisingly specific. They measure brake efficiency as actual deceleration in meters per second squared. Not "do the brakes feel okay" — they put the car on a roller brake tester and measure exactly how much braking force each wheel generates relative to the vehicle weight. They check suspension play by putting each wheel on a shaker plate and measuring lateral and vertical movement in millimeters. Steering rack free play, same thing — they measure the angular play at the steering wheel before the wheels actually move.
It's not a mechanic squinting at things and nodding. It's quantitative.
Not at all. They measure exhaust emissions — carbon monoxide, hydrocarbons, and nitrogen oxides for gasoline engines, opacity for diesel. Tire tread depth — the legal minimum is one point six millimeters, but in practice testers start flagging below three. Headlight aim and brightness get checked on an optical rig. And they do a visual inspection of the chassis for structural corrosion, especially around suspension mounting points and the main frame rails.
This is where it gets interesting to me. Because a car can pass all of that at ten in the morning and have a brake fluid leak develop by three in the afternoon. The test is a snapshot. It's not a warranty.
That's the single biggest misconception, and it's worth sitting with for a second. A passing roadworthiness test means that at the moment of inspection, the vehicle met minimum safety and emissions thresholds. It does not mean the car is safe for a full year. Brake pads can wear from four millimeters to nothing in a few thousand kilometers if someone's riding the pedal. A coolant hose can look fine visually and rupture the next week. The test doesn't check for things that are about to fail — it checks for things that have already failed.
Like a medical checkup that says you're fine today but doesn't catch the aneurysm that blows tomorrow.
And the Israeli test is actually more stringent than many US state inspections. In something like twenty states, there's no safety inspection at all. You register, you drive. But Israel's test is still less comprehensive than Germany's TÜV or the UK's MOT. The TÜV, for example, is every two years and includes much more aggressive corrosion checks — they'll probe suspect areas with a screwdriver. If the metal gives way, you fail. The UK MOT has specific categories for "dangerous" faults — if they find one, you're legally prohibited from driving the car away. You have to get it towed.
The screwdriver test is a level of commitment I respect. "We're not asking your chassis if it's okay. We're challenging it." I'm imagining a very stern German inspector just stabbing at a frame rail and frowning.
That's not far from the truth. And that gets at the philosophical difference between testing regimes. Japan's shaken test is the extreme case — it's famously expensive, can cost over a hundred thousand yen, which is about six hundred seventy dollars, and it's so comprehensive that it effectively forces cars off the road at ten to twelve years old. The cumulative cost of passing the test exceeds the car's value, so people scrap perfectly functional cars and buy new ones. It's a de facto industrial policy wrapped in a safety inspection.
Which is the opposite of Israel's situation. Here, the test is just hard enough to keep truly dangerous heaps off the road, but not so hard or expensive that it forces scrappage. Which means old cars stay in circulation.
They fail at a significant rate. The twenty twenty-four Ministry of Transport annual report shows twenty-three percent of cars fail on the first attempt. That's nearly one in four. The top failure category is brakes, at thirty-four percent of all failures. Suspension is next at twenty-eight percent. Those are the two systems that take the most punishment on Israeli roads, which — let's be honest — are not exactly the Autobahn.
Potholes the size of small bathtubs. I've seen a Fiat disappear into one.
That's only a slight exaggeration. So you have a system where a quarter of cars fail, the failures are concentrated in wear items, and the economic incentive is to do the absolute minimum to get the pass. Not to fix the car properly — to fix it enough to get the stamp.
That's where we cross from engineering into economics. Because the "fix it enough to pass" mentality creates a specific kind of deferred maintenance debt that accumulates over time. What does that debt actually look like in shekels?
Let's put some numbers on this. In Israel, you can buy a fifteen-plus-year-old car for somewhere between fifteen thousand and twenty-five thousand shekels. That's roughly four to seven thousand dollars. It seems like a bargain. But annual maintenance on a car in that age bracket averages four to six thousand shekels. That's not including fuel or insurance — that's just keeping it running and passing the test.
You're spending twenty-five to forty percent of the purchase price every single year just on maintenance. That's like buying the car all over again every three to four years.
That's the average. Some years will be lower, but then you get the year where the timing belt needs doing, the brake rotors are warped, and the exhaust has a hole, and suddenly you're looking at eight thousand shekels in one go. That's when people start having very difficult conversations with their bank accounts.
This is the trap, right? You buy the car for cheap, you think you're being smart, and then the maintenance budget eats you alive. But I want to understand the shape of that trap more clearly. Is there an actual crossover point where the math flips?
There's actual analysis of Israeli used car data that shows total cost of ownership — purchase price plus maintenance plus fuel plus insurance — bottoms out at cars aged seven to nine years. That's the sweet spot. After year twelve, the cost curve bends upward sharply. By year fifteen, you're often spending more per kilometer than you would on a newer, more expensive car.
The conventional wisdom — "buy an old beater and drive it into the ground" — is mathematically wrong after a certain point. This is the kind of thing that feels like it should be true but isn't.
It works if you get lucky and buy a car that was maintained properly. But that's the second trap. In Israel, cars that are fifteen years old have often changed hands five, six, seven times. And each owner had a different approach to maintenance. Some did the minimum to pass the test. Some did nothing and passed the problem to the next buyer. Some disabled warning lights rather than fix what they indicated.
There was a twenty twenty-three police report that estimated fifteen percent of imported used cars in Israel have tampered odometers. That's not a rounding error — that's a systemic fraud problem.
That's just the imports. The domestic market has its own issues. A car that's been through six owners in twelve years — I want to talk about what that car actually looks like. Let me give you a concrete case. A two thousand one Subaru Legacy. It's changed hands six times. Each owner did only what was needed to pass the annual test. No preventative maintenance. No fixing things that weren't on the test checklist. By the time it hits owner number six, it's got four different brands of tires on four wheels — which means different grip characteristics, different wear rates, and unpredictable handling in an emergency. The suspension has mismatched components — one control arm replaced with a cheap aftermarket part, the other original and worn. The check engine light has been permanently disabled — not fixed, just disconnected, probably by pulling the bulb or cutting the wire. The service history is a folder with maybe three receipts from the last decade.
That car is a liability with license plates. I'm picturing it now — four different tires, each gripping the road differently in the rain. You hit the brakes in an emergency and the car doesn't slow down in a straight line, it rotates. That's genuinely dangerous.
Someone will buy it for eight thousand shekels and think they got a deal. Then within six months, the head gasket goes, or the transmission starts slipping, and suddenly they're three and a half thousand shekels into repairs on a car they paid eight for. That's the negative returns scenario. The savings from buying cheap get completely erased, and then some.
The question becomes — how do you avoid being that person? What are the warning signs?
The number of previous owners is a real signal, and it's one people don't pay enough attention to. A car that's had two owners in ten years is a very different proposition from a car that's had five. Frequent ownership changes often indicate a problem car — someone buys it, discovers the issues, does the minimum to keep it running, and unloads it before the next big repair comes due.
The hot potato model of car ownership. Nobody wants to be the one holding it when the music stops.
And there are specific vintages that are particularly problematic. Cars from roughly two thousand to two thousand five are in a danger zone right now. They predate electronic stability control, which became mandatory in Israel in twenty eighteen and in the EU in twenty fourteen. So you're missing a major safety system that's now standard. But more importantly from a maintenance standpoint, those cars have chassis designs that are simpler but more rust-prone. The corrosion protection wasn't as advanced as it is now. The electronics are old enough to be failing but modern enough to be expensive to diagnose. And parts availability for twenty-plus-year-old models starts to become an issue — you're hunting through scrapyards or ordering from specialty suppliers.
I hadn't thought about the parts availability angle. At some point, even a simple component becomes unobtanium. What's a real example of that?
A power steering rack for a two thousand three Toyota Corolla might run you fifteen hundred shekels for a rebuilt unit plus labor. On a car worth eight thousand shekels, that's a tough pill to swallow. But what's the alternative? You can't drive without power steering. So you fix it, and now you're deeper into a car that's still only worth eight thousand. And that's assuming you can even find the part — for some models from the late nineties, you're calling scrapyards in three different cities hoping someone has a donor car that hasn't already been stripped.
This is the sunk cost spiral. You've put money in, so you feel like you have to keep putting money in to justify what you already spent. But at what point do you just walk away from a car you've already sunk thousands into?
That's the question that keeps people up at night. And the psychology is brutal — nobody wants to admit they made a bad purchase, so they keep feeding money into a dying car because stopping would mean acknowledging the loss. Behavioral economists call this the sunk cost fallacy, but when it's your only way to get to work, it doesn't feel like a fallacy. It feels like survival.
It's easy to say "cut your losses" from an armchair. It's different when the alternative is taking three buses to a job that starts at six in the morning.
That's how people end up spending twenty thousand shekels over three years maintaining a car they bought for ten. The rational move would have been to scrap it at the first major repair and buy something else. But human psychology doesn't work that way, and the Israeli transportation infrastructure doesn't give people a lot of alternatives outside the major cities.
What does the smart version of this look like? If you're a family in Israel that needs a car and can't afford new, what's the actual strategy?
The sweet spot, based on the data, is six to nine years old with two or fewer previous owners and a complete service history. That car has depreciated past the steepest part of the curve but hasn't yet entered the high-maintenance tail. You'll pay more upfront — maybe forty to sixty thousand shekels instead of fifteen — but your annual maintenance will be a fraction of what the older car costs. And critically, you can actually plan your costs instead of being ambushed by them.
That planning part is key, isn't it? Because the real killer with old cars isn't the average cost — it's the variance. You can budget for four thousand shekels a year and then get hit with an eight thousand shekel repair out of nowhere.
And that variance is what destroys household budgets. A newer used car has lower variance. You're still going to have maintenance costs, but they're more predictable. You're not going to wake up one morning and discover your engine needs a complete rebuild.
The service history thing is interesting. How do you verify it in Israel? Because I've seen "full service history" in listings and it turns out to be a notebook with some handwritten oil change dates and a coffee stain.
This is one area where the Israeli system actually helps you. The Ministry of Transport maintains a database of every roadworthiness test result since twenty ten. You can look up any car by license plate and see its entire test history — every pass, every fail, what it failed on, what was repaired. If a car has failed for brakes three years running, you know the owner was patching, not fixing. If it's failed for suspension at multiple test stations, that's a red flag. The test history is basically a medical record for the car, and most buyers don't check it.
That's actionable. Before you even look at the car, pull the test history and see if it's been failing the same things repeatedly. That's free due diligence.
If you're in a country without mandatory testing — which includes a lot of US states — the equivalent is paying for a pre-purchase inspection at an independent shop. It costs a hundred to two hundred dollars, and it's the best money you'll ever spend on a car you don't own yet. A good mechanic will put it on a lift, check compression, do a leak-down test, scan for error codes, check for crash damage, and give you a report. If the seller won't let you do it, walk away.
That's the universal rule, isn't it? If the seller resists inspection, the car is hiding something. I've heard every excuse — "I don't have time," "my mechanic just looked at it," "I've got three other buyers interested." It's all noise.
I've never heard of a legitimate seller refusing a pre-purchase inspection. They might ask you to pay for it, which is fair, but they won't block it.
Let's talk about the specific pattern Israel has developed to cope with this distorted market. Because there's a cycle I've noticed — families buy an old car, drive it until the annual test cost exceeds what the car is worth, scrap it, and buy another old car. It's a churn cycle. And I wonder if that cycle is actually rational given the constraints, or if people are just trapping themselves over and over.
The perpetual beater cycle. And it's rational at the individual level but terrible at the system level. You're constantly driving the oldest, least safe, least efficient cars, and you're constantly exposed to the risk of a catastrophic failure that totals the car economically. But when a new Corolla costs a hundred and sixty thousand shekels, what are you supposed to do? The math of the beater cycle almost makes sense if you assume you'll get lucky and avoid the big repair. But over a lifetime of car ownership, you're not going to get lucky every time.
That's the thing about probability — it catches up with you. You might dodge the bullet on one beater, maybe two. But if you're in the beater cycle for twenty years, you're going to eat at least one catastrophic repair. And that one repair might wipe out all the savings from the previous beaters.
There's a concept in finance called "tail risk" — the risk of a rare but catastrophic event. And the beater cycle is essentially a strategy of ignoring tail risk. You save money most years, and then one year you lose everything you saved plus more. It's like picking up pennies in front of a steamroller.
That's the policy question lurking under all of this. The eighty-three percent tax is essentially a choice — to make new cars a luxury good and accept that a significant portion of the population will drive aging, less safe vehicles. Has anyone in the Knesset actually made that argument explicitly, or is it always framed as congestion management?
The stated rationale has historically been about managing congestion and trade deficits — Israel doesn't manufacture cars domestically, so every car is an import that affects the balance of payments. The tax suppresses demand. But the side effect is a fleet that's older and less safe than it would be otherwise, and a population spending a higher percentage of income on maintenance rather than depreciation. And I've never seen a government document that explicitly says "we accept that poorer Israelis will drive less safe cars as a consequence of this policy." It's always framed in terms of macroeconomics and traffic management.
There's an equity dimension too. A rich family buys a new car, absorbs the tax hit, and drives a safe modern vehicle. A poorer family buys the rich family's castoff seven years later and spends the next decade patching it together. The tax is regressive in practice even if it looks like it only hits luxury purchases.
The tax falls on the first purchaser, but the consequences flow downhill for years. It's a progressive-looking tax with deeply regressive downstream effects.
There's a proposed reform being discussed — I think the twenty twenty-six budget talks include it — to tie test frequency to vehicle age. Annual for three to ten years, then semi-annual for ten-plus. What would that actually do to the economics?
That would change the economics significantly. If you have to pass the test twice a year on a fifteen-year-old car, the preparation costs double. You're not just fixing things once a year — you're in the shop twice, you're paying for two inspections, and small issues that you might have let slide for a few months become immediate blockers. That might push a lot of marginal cars off the road entirely, which is probably safer but also means people who can't afford anything better lose their transportation.
It's the Japan shaken problem in miniature. You make the test strict enough, and you effectively ban old cars without saying you're banning old cars. You just make them economically non-viable.
That's a legitimate policy choice, but it should be made explicitly. If the goal is to get old cars off the road for safety and environmental reasons, say that and provide some kind of scrappage incentive. Don't do it through the back door of inspection requirements. At least have the honesty to say "we are making a choice to phase out older vehicles, and here's how we're going to help the people who depend on them.
Alright, let's shift to something forward-looking. What happens when electric vehicles start aging into this system? We're not that far from the first generation of mass-market EVs being ten, twelve years old. How does the roadworthiness test adapt? Because the failure modes are completely different.
This is a open question that no country has fully answered. Current tests don't assess battery health at all. An EV with sixty percent of its original range — so maybe a hundred miles instead of a hundred and sixty — still passes every existing roadworthiness check. The brakes work, the suspension's fine, the lights are aimed correctly. But is that car practically roadworthy for a family that needs to drive between cities?
It's roadworthy in the legal sense but not in the functional sense. It's like having a car with a fuel tank that's shrunk to half its original size, except you can't just fill it up more often — the "tank" itself is permanently smaller.
And battery degradation isn't linear or predictable. Two identical EVs with the same mileage can have very different battery health depending on charging habits — fast charging versus slow, charging to a hundred percent versus eighty, operating in hot climates versus moderate ones. A snapshot inspection can't tell you whether the battery will lose another ten percent in the next year or fail completely. It's not like measuring brake pad thickness where you can see the wear and project when it'll hit the limit.
We're heading toward a world where the test certifies everything except the one component that determines whether the car is actually usable. That seems like a pretty significant blind spot.
Replacing that component currently costs more than the car is worth. A battery pack replacement on a ten-year-old EV can run ten to fifteen thousand dollars. On a car that might be worth eight thousand. That's instant economic write-off territory. It's the EV equivalent of a blown engine on a fifteen-year-old Corolla, except with an engine you can find a used replacement for three thousand shekels and swap in. The EV battery is a single, massively expensive component with no cheap aftermarket alternative yet.
Which means the used EV market is going to develop its own version of the problems we see in Israel's combustion fleet, but with a different failure mode. Instead of rust and worn brakes, it'll be degraded batteries that nobody wants to pay to replace. And the car will look perfect — no rust, no smoke, no weird noises — but it'll have a hundred kilometers of usable range.
I think we'll see testing standards evolve to include a battery state-of-health measurement — probably a standardized diagnostic readout that gives a percentage relative to original capacity. And then regulators will have to decide what number constitutes a failure. And what do you do with the cars that fail? Right now there's no recycling infrastructure at scale for EV batteries. It's coming, but it's not here yet. So you could end up with a situation where perfectly functional cars — functional in every way except range — are being scrapped because the battery replacement cost exceeds the vehicle value.
That's an environmental disaster in its own way. The whole point of EVs is sustainability, and we're building cars that might get scrapped at ten years because of a single component that we can't economically replace. That's not sustainable at all.
There's work being done on modular battery designs that would allow individual cell replacement rather than full pack replacement, and on battery recycling that could bring down the cost of replacements. But we're not there yet. The first wave of mass-market EVs — the early Nissan Leafs, the early Teslas — they're going to be the test case. And I'm not sure the regulatory framework is ready.
The car you drive ends up being this perfect reflection of the regulatory and economic system you live in. In Israel, you drive old combustion cars held together by the annual test because the tax code makes new cars inaccessible. In Japan, you drive new cars because the inspection system makes old ones uneconomical. In the US, depending on your state, you might drive anything from a pristine garage queen to a rolling death trap with no inspection at all. And in the near future, you might drive an EV that passes every safety test but can barely get you across town.
In the future, you might drive an EV that's perfectly safe in every traditional sense but can barely get you to work and back because the battery is shot and the test doesn't care. The car becomes a physical manifestation of whatever compromises your society has decided to make.
Which brings us back to the practical question. For someone listening right now, in Israel or anywhere else, who needs to buy a used car — what do they actually do? Give us the checklist.
First, pull the test history if your country has one. In Israel, the Ministry of Transport database is online and free. Look for patterns — repeated failures in the same category mean deferred maintenance. A clean history with occasional failures that were immediately resolved suggests an owner who actually fixed things rather than patching them. If you see "failed brakes" three years in a row followed by "passed," that means someone replaced the pads at the last possible moment each time and probably did nothing else to the braking system.
Target the sweet spot. Six to nine years old, two or fewer previous owners, complete service history. You'll pay more upfront, but your total cost of ownership over three to five years will almost certainly be lower than buying a cheaper, older car. And if you're looking at something older than twelve years, budget ten to fifteen percent of the purchase price annually for maintenance. Not "hope it doesn't need anything" — budget for it. If you can't afford that budget, you can't afford the car. That's the rule that people resist because it makes the car feel more expensive, but it's the truth.
Pay for the pre-purchase inspection. Even in countries with mandatory testing, the annual test is not a substitute for a buyer's inspection. A hundred or two hundred dollars to put the car on a lift and have someone who knows what they're looking at go through it systematically. Compression test, leak-down test, scan for error codes, check for crash damage and rust. If the seller balks, walk. There's always another car.
The walk-away is the most powerful negotiating tool in used car buying, and it's the one people are least willing to use because they've already emotionally committed to the car. They've pictured themselves driving it. They've told their spouse about it. They've mentally spent the money.
The car doesn't know you exist. It won't have its feelings hurt if you don't buy it. But your bank account will definitely have feelings if you buy a bad one.
I don't know. Some of these old Subarus, I think they have opinions.
A 2001 Legacy with six owners definitely has opinions. They're just not opinions you want to hear. Probably opinions about head gaskets, mostly.
So to pull this together — the roadworthiness test validates a very specific set of things at a specific moment in time. It's useful, it keeps the worst hazards off the road, but it's not a guarantee and it's not a substitute for ongoing maintenance. The economics of old cars are counterintuitive — cheap to buy, expensive to own, and the crossover point where that flips is somewhere around twelve years. And the smart play is to spend more on a younger car with fewer owners and a verifiable history.
If you're in Israel, none of this is your fault. The system is stacked to make new cars a luxury, and you're navigating that as best you can. Just navigate it with your eyes open and a mechanic you trust. And check that test history database before you hand over any money.
Now: Hilbert's daily fun fact.
Hilbert: In 1723, the Spanish governor of colonial Belize was crowned while seated on a throne carved entirely from solidified sea salt, a tradition meant to symbolize that his authority, like salt, would both preserve and corrode the colony. The throne dissolved completely during an unusually heavy rainy season the following year.
That's a metaphor doing a lot of work.
Symbolism is important until it literally melts under you.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop. If you enjoyed this episode, leave us a review wherever you get your podcasts — it helps other people find the show. We'll be back with more of your prompts soon.
