Japan's Two-Yearly Car Inspection and When Your Car Stops Being Worth the Sha-ken
A personal breakdown of why cheap used cars in Japan lose money faster than foreign buyers expect. The sha-ken cost is only half the trap—the other half is the exit price nobody talks about.
5 min read · Category: buying
I paid ¥500,000 for a beautiful Honda kei
. Eighteen months later, kaitori offered me ¥30,000.
The sha-ken bill itself is rarely what breaks a cheap-used-car purchase. What breaks it is the exit—kaitori chains stop paying meaningful money for cars past a certain age, and by the time your first sha-ken renewal arrives, the car you paid ¥500,000 for may be worth ¥30,000 to anyone who's offering. Foreign buyers coming from the US or Europe usually see only the sha-ken cost when they think about ownership. The exit price is the invisible half of the math, and it's where the real loss shows up.
What actually happened with my Honda
A couple of years ago I found a Honda N-One GL that had been imported to Sapporo from Honshu. About ten years old, roughly 100,000 km on the clock, no rust worth mentioning, and a fresh sha-ken. It looked, drove, and felt like a great deal. The price was ¥500,000, which was below the market for the condition it was in.
For a year and a half, the car was exactly what I hoped it would be. It ran well, took care of me through a Hokkaido winter and a summer, and never surprised me. On paper, I was winning.
Then I decided to sell it.
The kaitori chains—the same ones every Japanese person defaults to when they want to move a car quickly—offered me between ¥30,000 and ¥50,000. Not one of them treated the car as if it had any meaningful value. I posted on Jmty. I tried Facebook Marketplace. I tried the local community boards. Nothing worked. In the end, a friend was willing to pay ¥250,000 for it, and I took the deal because the alternative was watching it slide toward the kaitori floor.
The honest way to read that number is that without a friend willing to help, I would have lost roughly ¥450,000–¥470,000 on that car in eighteen months. Even with the friend, I lost ¥250,000. On a ¥500,000 purchase.
That is the trap this article is about. It has almost nothing to do with the sha-ken bill you dread and almost everything to do with how the Japanese used-car market prices cars at the exit.
Why kaitori quotes look the way they do
The first thing to understand is that kaitori chains—Gulliver, IDOM, Big Motor, Apple, and the smaller regional players—are not trying to pay you what your car is worth to a driver. They are trying to pay you what they can get for it at the next weekly auction, minus their operating margin.
Every one of those chains resells the cars they buy through auction houses like USS or JU. The bidders at those auctions are dealers, exporters, and other kaitori operators, and they price cars based on a small handful of signals: year of first registration, model, grade, condition grade, and remaining sha-ken. Kilometres matter, but less than most foreign buyers assume. Once a car is past about ten years old, the auction floor collapses regardless of how carefully the previous owner treated it.
So the kaitori quote you get is a back-solve. It starts from "what will this move for at Tuesday's auction," subtracts transport, inspection, and margin, and lands on the number you're offered. For a well-kept ten-year-old kei, that back-solve often produces something between ¥30,000 and ¥80,000. That is not because your car is worthless. It is because the buyer pool the kaitori chain is selling into does not pay meaningful money for cars past a certain age.
The gap between what you feel your car is worth and what the market will actually pay is the entire problem.
Sha-ken is the trigger, not the villain
Most articles about the cost of owning a used car in Japan focus on sha-ken as if it were the main enemy. It isn't. The sha-ken bill is the moment you find out how bad the situation already is.
Here's how sha-ken cost tends to scale with car age, in rough bands. Actual numbers vary by vehicle, region, and shop, but this is close enough to plan around:
| Vehicle age | Typical sha-ken cost (compact / kei) | Typical sha-ken cost (mid-size / larger) |
|---|---|---|
| 3–4 years | ¥60,000–¥90,000 | ¥90,000–¥130,000 |
| 5–7 years | ¥80,000–¥120,000 | ¥120,000–¥170,000 |
| 8–10 years | ¥100,000–¥150,000 | ¥150,000–¥200,000 |
| 11–13 years | ¥130,000–¥200,000 | ¥180,000–¥250,000 |
| 14+ years | ¥150,000–¥250,000+ | ¥200,000–¥300,000+ |
The reason the older-car numbers rise isn't that the mandatory fees change much. Jibaiseki insurance, weight tax, and the inspection fee itself are close to fixed within a vehicle class. What changes is the repair pile. On an older car, rubber has aged, brake pads and rotors are due, suspension bushings are due, exhaust components are due, and the shop has to actually pass a real inspection—not sign off a car that's been kept new.
For a foreign buyer who paid ¥500,000 for a ten-year-old car, the first sha-ken quote often lands between ¥130,000 and ¥180,000. That's not the disaster. The disaster is what comes next: you look at that quote, decide to sell rather than renew, and discover the exit is worse than the bill.
The math foreigners don't do
Sha-ken forces a decision every two years, so the honest window to compare two cars is a two-year hold—from one sha-ken to the next. That's how ownership actually works in Japan. Nobody plans four years ahead. Everyone plans to the next inspection.
Consider two ways of spending money on transportation over two years.
Scenario A: The "cheap" old car. Buy a ten-year-old kei or compact for ¥500,000 with fresh sha-ken. Drive it for two years. No sha-ken renewal in the window—you sell just before the next one is due, because the quote to renew is close to what the car is worth. Realistic unplanned repairs across two years on a car ageing from ten to twelve sit around ¥50,000–¥100,000. Sale proceeds at year twelve, through kaitori, are typically ¥30,000–¥80,000.
Scenario B: The moderate-newer car. Buy a three-year-old compact for ¥1,200,000 with fresh sha-ken. Drive it for two years. Again no sha-ken renewal in the window—same logic, you sell before the next inspection. Repairs across two years on a car ageing from three to five are typically zero to ¥30,000. Sale proceeds at year five, through kaitori, are typically ¥750,000–¥900,000 for a mainstream compact in average condition.
The two-year total-cost-of-ownership math looks like this:
| Cheap old (¥500k kei, buy at year 10) | Moderate-newer (¥1.2m compact, buy at year 3) | |
|---|---|---|
| Purchase price | ¥500,000 | ¥1,200,000 |
| Sha-ken cycles in window | 0 (sell before renewal) | 0 (sell before renewal) |
| Realistic repairs (2 yrs) | ¥50,000–¥100,000 | ¥0–¥30,000 |
| Sale proceeds at exit | ¥30,000–¥80,000 | ¥750,000–¥900,000 |
| Net 2-year cost | ~¥470,000–¥570,000 | ~¥330,000–¥480,000 |
| Per year of driving | ¥235,000–¥285,000 | ¥165,000–¥240,000 |
The cheap old car isn't cheaper. It costs more per year to drive, and you spent those two years in an older, rougher car with a much worse exit. The newer car costs less per year, drives better, is safer, and sells to any kaitori chain in a fifteen-minute phone call.
My own N-One story sits right on top of this table. The model predicts ¥470,000–¥570,000 lost over two years on a cheap old kei. My actual loss on the N-One was ¥250,000 in eighteen months—and that was with a friend willing to pay ¥250,000 for it. Without that friend, my loss would have been closer to ¥450,000–¥470,000, which is exactly what the table predicts. The model is not pessimistic. It matches what actually happens.
The cheap-car math only works if you buy a car for ¥100,000–¥200,000 and treat it as fully disposable from day one. In that case, you're not really buying transportation—you're renting time until the sha-ken forces the decision. That's a legitimate strategy for someone who's certain they'll be in Japan for a short window and doesn't mind driving something rough. It is not the strategy most foreign buyers think they're pursuing when they spend ¥500,000 on a car that looks nice on Yahoo Auctions.
Why the kei/economy trap hits especially hard
The N-One story is not a coincidence of one bad car. Older kei and economy cars lose value faster at the exit than most foreign buyers realise, and the reasons are structural.
The buyer pool for used keis in Japan skews toward first-time drivers, older rural drivers, and people buying a household's second car. Almost all of them prioritise the same thing: cheapest possible car with fresh sha-ken. A ten-year-old kei with sha-ken running out has none of what that buyer wants. They'd rather pay ¥400,000 for a five-year-old N-Box with two years of sha-ken remaining than ¥100,000 for a ten-year-old N-One where they'll immediately face a ¥130,000 inspection.
Kei cars also don't hold value at auction the way mid-size cars do. Auction bidders exporting used cars overseas are generally not buying old keis—the export markets that pay premium for used Japanese cars want SUVs, minivans, and performance models, not aged economy cars. So the kaitori chains have almost nowhere profitable to send an older kei except to scrap or to the roughest end of the domestic used market, and their offers reflect that.
The compact market (Fit, Vitz, Note, Demio) sits slightly better than kei but follows the same curve. Once a compact crosses about twelve years, the same collapse happens. Family-sized cars (Alphard, Voxy, Serena) hold value longer because there's a stronger export market for them, but even they hit a wall eventually.
None of this is a reason to avoid kei or compact cars. They're often the right choice for a foreign buyer in Japan. The point is to buy them at the right age, and to sell them before the exit collapses.
The ownership window that actually saves money
If you plot depreciation against age for almost any mainstream Japanese used car, the curve has three phases. Steep loss in the first three years. A flat middle stretch from roughly year three to year six or seven. A second decline after year seven, accelerating past year ten.
If you plot sha-ken and maintenance cost against age, the curve has two phases. Roughly flat through year seven or eight. A steep rise after that as consumables and rubber age out.
The intersection of those two curves is the window where total cost of ownership is lowest. Buy at year two or three, sell at year five or six. You skip the steepest part of the depreciation curve because someone else already ate it. You exit before the maintenance curve turns on you. And—most importantly—you sell while kaitori chains are still bidding real money, because your car is still inside the age band their auction buyers actually want.
In practice this looks like:
- Kei: Buy a three-year-old N-Box, Tanto, or Spacia at around ¥900,000–¥1,100,000. Sell at year six for ¥500,000–¥700,000.
- Compact: Buy a three-year-old Fit, Vitz, Note, or Solio at around ¥1,000,000–¥1,300,000. Sell at year six for ¥550,000–¥800,000.
- Family / larger: Buy a three-year-old Freed, Sienta, Voxy, or Serena at ¥1,600,000–¥2,200,000. Sell at year six for ¥900,000–¥1,400,000.
The three-year losses on those exits, spread over three years of ownership, work out to somewhere between ¥100,000 and ¥250,000 per year depending on model and condition. Compared to the N-One story—¥250,000 lost in eighteen months, and that with a friend's help—the newer-car window is often cheaper per year of driving, not more expensive.
There's more depth on why age matters more than mileage in Japan in the companion piece Why a 10-Year-Old Japanese Car with 10,000 km Is Still Cheap. The short version: the sha-ken schedule and the auction market both index on age, not on odometer, and that shapes almost everything about how Japanese cars depreciate.
A simple decision rule for current owners
If you already own a car and you're staring at a sha-ken quote wondering whether to renew or sell, the honest test is short:
If your next sha-ken quote is greater than what a kaitori chain would give you today, sell now.
Kaitori quotes are free and take fifteen minutes. Call two or three chains (Gulliver, IDOM, Apple) and get real numbers on your current car's exit value. Compare that to the sha-ken quote you've been given. If the sha-ken costs more than the exit price, renewing makes almost no financial sense—you're paying to keep a car that's already worth less than the maintenance bill.
There are legitimate exceptions. If you'll be in Japan for exactly one more year, renewing sha-ken and running the car into the ground can still be cheaper than buying and reselling another one. If the car has meaningful sentimental value, that's a personal decision, not a financial one. But for the typical case, the rule holds. Sha-ken is a forcing function that tells you the market has already moved past your car.
Exit strategy: how to actually sell for real money
If you buy inside the 2–3 to 5–6 window, your options at exit are much better than they were in my N-One story. In order of typical proceeds:
- Private sale to a friend or coworker. Highest realised price if you can find a buyer. Requires trust on both sides and some paperwork help. Not something you can plan on—consider it a bonus if it happens.
- Jmty, Facebook Marketplace, and local community boards. Works reasonably well for cars inside the sweet-spot window with fresh sha-ken. Almost never works for older cars, as I learned. Budget four to eight weeks of patience.
- Dealer trade-in. Convenient if you're buying your next car from the same dealer. Typically produces less than a kaitori chain would give you, because the dealer is baking the trade-in margin into your new-car price.
- Kaitori chain (Gulliver, IDOM, Big Motor, Apple). Convenient and fast. Not the highes
…