In brief
A5 is the label the entire English-speaking wagyu market is sold on. It appears on menus, on packaging, in advertising, and in almost every article written about Japanese beef. It is also, as it is usually explained in English, wrong.
Not wrong in the sense of being made up. A5 is a real grade, issued by a real institution, and it does sit at the top of Japan’s grading scale. It is wrong in the sense that the two things people believe it means — this beef will taste the best, and this beef is extremely rare — are not what the grade measures, and in the second case are contradicted by Japan’s own published grading data.
This page explains what A5 is, using the grading standard and the grading statistics themselves. We publish Japan’s wagyu restaurant selection and take no money from the restaurants or producers we cover, so we have no reason to make the grade sound more impressive than it is.
We should say our position plainly at the start, because it shapes everything below. We do not think marbling is a mistake, and we are not trying to talk anyone out of eating A5. Our view is narrower and, we think, harder to argue with: a single number has become the only thing anyone can see, and everything that single number does not capture has become invisible — including to the people paying for it. The grade is not the problem. Being the only visible thing is the problem.
Japanese beef carcasses are graded by the Japan Meat Grading Association (JMGA), a third-party body, under the Beef Carcass Trading Standards. Every carcass receives two independent scores, which are then printed side by side.
So “A5” means: highest yield band, highest quality band. Two thresholds cleared. That is the entire content of the label.
What it does not contain is any assessment of how the beef tastes. No one eats the carcass to grade it. There is no cooked sample, no tasting panel, no aroma evaluation. This is not a criticism of JMGA — the system was built to make carcass trading fair and consistent between producers and wholesalers, and at that job it works well. The problem is what happens when a trading standard is exported abroad as a promise about flavour.
Most claims about grading are hard for a reader to verify. This one is not, because the yield grade is a published formula:
Yield estimate = 67.37 + 0.130 × ribeye area (cm²) + 0.667 × rib thickness (cm) − 0.025 × cold carcass weight (kg) − 0.896 × subcutaneous fat thickness (cm)
(+2.049 for designated beef breeds)A grade = 72 or above | B grade = 69 to under 72 | C grade = under 69
Four inputs: an area, a thickness, a weight, and another thickness. There is no variable for flavour, aroma, tenderness or fat quality anywhere in the equation, because the formula is not trying to describe eating. It is trying to predict butchering yield — a commercial question of how many kilograms of sellable meat come off the frame.
This means the most common way A5 gets explained in English — that the letter is a measure of quality and the number is a measure of grade, or some variation of the two being “quality scores” — is not a matter of interpretation. It is arithmetically incorrect. The letter is an economic measure that happens to be printed next to a quality measure.
The quality grade has four sub-scores, and the carcass receives the lowest of them. But the four are not evaluated with equal precision. Three of them — colour, firmness and texture, fat colour and quality — are assessed against reference standards in broad bands. Only marbling has its own fine-grained numeric scale: the Beef Marbling Standard (BMS), running from No.1 to No.12.
| Meat quality grade | BMS number |
|---|---|
| 5 | No.8 – No.12 |
| 4 | No.5 – No.7 |
| 3 | No.3 – No.4 |
Because the other three sub-scores rarely fall low enough to be the binding constraint on good wagyu, the number in “A5” is, for practical purposes, a marbling threshold. A5 tells you the carcass hit BMS 8 or higher — and nothing about where in that range it landed. BMS 8 and BMS 12 are both “5”.
The full series behind this — every year from 1996 to 2026, with the head counts and the source files — is set out separately in How Rare Is A5 Wagyu?
Here is the number that changes the picture, and the one that is almost entirely absent from English-language coverage.
JMGA publishes its grading results monthly, broken down by breed and sex. Those files are the primary record of how much Japanese beef receives which grade. According to them, as of May 2026:
Three out of four steer wagyu carcasses are A5. And this is a recent development, not a permanent state of affairs:
| Year | Share of steer wagyu graded A5 |
|---|---|
| 2001 | 11.8% |
| 2018 | 40.5% |
| 2022 | 59.7% |
| May 2026 | 76.6% |
The standard itself has not been loosened. Japanese breeding got extremely good at producing marbling, generation after generation, because marbling was what the market paid for. The threshold stayed put and the population moved past it.
Compare that with the figure repeated across English-language wagyu marketing, which typically states that only 1–2% of Japanese cattle achieve A5. That claim circulates widely, including in articles that rank at the top of search results for this exact question. Measured against JMGA’s own numbers it is wrong by well over an order of magnitude — more than thirtyfold even if you take the more generous 2% end of it. As far as we have been able to determine, almost no English-language source cites the JMGA grading data directly — the figures get passed between retailers rather than checked at source, and nearly all of those retailers sell the beef in question.
The market has noticed even if the marketing has not. The carcass price gap between A5 and A4 had narrowed to roughly 6.5% at the Tokyo market as of May 2024, and in January 2026 A5 wagyu hit its lowest price in eleven years. A label held by three quarters of the category cannot command much of a premium indefinitely.
The honest way to state this is not that A5 has gotten worse. It is that A5 has lost its power to distinguish. It used to be a filter. It is now closer to an attendance record.
The qualities researchers associate with how wagyu actually eats are, almost without exception, not graded at all.
Work by Motoyama, Sasaki and Watanabe at Japan’s National Agriculture and Food Research Organization documents that Japanese Black fat averages around 52.9% oleic acid with a melting point near 20–21°C — which is why good wagyu fat can feel light rather than waxy. The same body of work shows that the characteristic wagyu aroma compounds are barely present at slaughter and develop during storage through oxygen exposure, becoming most pronounced around 80°C — precisely the temperature range of sukiyaki and shabu-shabu.
Fat composition, melting point, aroma development, bloodline, feeding length, handling and ageing: none of these appear in the grade. And the gap this creates is not theoretical. Japan’s own Ministry of Agriculture has published data showing carcasses at the same marbling score (BMS 9) ranging from 35.3% to 64.6% crude fat — close to a twofold difference in how fatty the meat actually is, under an identical label.
This is the practical consequence for anyone buying or ordering. A5 is a real signal about two things and silent about everything else, which is why the buyers we deal with do not treat it as a ranking. In our experience of the trade, an experienced buyer will readily choose an A4 carcass over an A5 one when the bloodline, the fat or the handling is better — and restaurants that buy on those grounds often serve better beef at a lower price than restaurants buying the letter and number alone.
We should be precise about what we are not saying. Marbling is not the enemy, and there is good research indicating that intramuscular fat contributes to eating quality. What we would question is the assumption hidden inside a 1-to-5 scale: that the relationship runs in a straight line, so that more is always better.
That is not how it works at the table. Fat carries flavour up to a point, and past that point it starts crowding everything else out — and where that point sits depends on the cut, the thickness, how it is cooked, and how much of it you are being served. There is an appropriate amount, not a maximum amount. A grading ladder that only counts upward cannot express that, which is why the most marbled carcass in a room is not automatically the one a good chef wants. In our experience, the beef that people remember is rarely the fattiest thing they ate that year.
If you want a single piece of evidence that the grade is not the top of the hierarchy in Japan, it is sitting inside the country’s most famous beef brand.
Matsusaka beef has a top tier called Tokusan Matsusaka Ushi — “special Matsusaka beef” — which accounts for only about 4% of Matsusaka beef and is registered as a protected Geographical Indication by Japan’s Ministry of Agriculture. Here is its definition:
Read that list again and notice what is absent. There is no grade requirement anywhere in it. Not A5, not A4, no BMS floor. Japan’s most storied beef designation is defined entirely by where the animal came from and how long it was raised — by lineage and by time.
This is our central point, made for us by an institution far older than any grading system: when Japanese producers wanted to mark out their very best beef, they reached for bloodline, not for a grading letter. The grade came later, and it measures something else.
It also explains something that confuses a lot of visitors. Regional brand names carry less information than people assume, because the best sires get bought by farms all over the country — two differently branded steaks can be closely related. The brand is a postcode. The bloodline is the thing.
It would be easy to read the above as a contrarian foreign opinion. It is not. It is closer to current Japanese policy.
The Ministry of Agriculture, Forestry and Fisheries states in its own published materials that beef-cattle production aligned with consumer needs requires attention to factors beyond marbling — naming feed efficiency, fertility, fat quality and fine-grained marbling. Alongside that:
There is also a conservation argument that has nothing to do with taste. Because breeding concentrated so heavily on marbling, use of a small number of sires became concentrated too: the inbreeding coefficient of registered Japanese Black females rose from 4.8% (born 1986) to 9.7% (born 2020), and the calves of the top 15 sires account for roughly half of market transactions. MAFF documents the associated risks — stillbirth, infertility, poor growth. A single measure, followed for decades, has narrowed the genetic base of wagyu itself.
Not a different number. Replacing A5 with another single score would reproduce the same failure — and several competing alternative indices already exist without any of them becoming a standard.
What we do instead is refuse to collapse it into one figure, and keep the separate things separate. These are the axes we actually judge on, and what we have come to believe about each.
Our committee dines unannounced and pays its own way, and no restaurant can buy a place in our selection. The four criteria we judge on, and how we deal with what the grading system leaves out, are set out in The Standard. The same gap explains why the grade is no help if you are asking whether wagyu is healthy: it tracks how much fat there is, not what that fat is made of.
None of the above means avoiding A5. It means treating the grade as one variable rather than the ranking. In practice the useful question at the table is not “what is the highest grade here?” but “what do I want this to taste like?”
There is a simple test for whether a restaurant is buying on substance: ask where the animal came from. A kitchen that chose its beef can usually tell you the region, the producer and the bloodline. “It’s A5” is an answer about a certificate, not about an animal.
A5 means a carcass cleared the top yield band and the top quality band, as assessed by a third-party grader who never tasted it. The letter is a yield calculation containing no flavour term. The number is, in practice, a marbling threshold with a wide range inside it. And 76.6% of steer wagyu now clears the whole thing.
None of that makes A5 beef bad. Plenty of it is superb. It means the label is a floor, not a ranking — and that if you are choosing where to eat wagyu in Japan, the letters on the certificate are the least informative thing in the room.
There is one more reason we write these pages, and it is the least commercial one. Taste is partly a matter of resolution: the more distinctions you are able to make, the more of what is in front of you you can actually perceive. Someone who knows that the fat should feel light, that the animal came from a particular line, that this cut is being served at this thickness for a reason, is not just better informed — they get more out of the meal. That is the honest case against reducing all of this to one letter and one digit. Not that the grade lies, but that it stops you looking. Knowing more about what you are eating makes it taste better, and no two-character label was ever going to do that job for you.
Related reading: what the word wagyu actually covers, if you want the category before the grade; wagyu vs Kobe beef, and what Kobe beef is — whose certification, notably, does not require an A5; Matsusaka beef, defined by virgin females and 900 days rather than by any grade; olive wagyu, prized for exactly the fat qualities the grade does not measure; Japanese wagyu vs. American and Australian wagyu; and why wagyu is structurally expensive.
Where we’d actually send you to eat. Our committee dines unannounced, pays its own way, and takes no money from any restaurant. See the selection of Japan’s best wagyu and yakiniku restaurants.