Why Clothing Sizes Are Inconsistent

Three garments, three different labels, one body that has not changed between the changing rooms. Anyone who has carried a size 10, a size 14 and a medium into the same cubicle already knows why clothing sizes are inconsistent is a reasonable question. The answer is not that a rule was broken.

There is no rule. In the United States a size label is a marketing decision, not a measurement, and no law obliges a brand to mean anything in particular by the number it prints.

What makes this interesting is the assumption underneath the frustration — that a standard once existed and was somehow lost. A standard did exist. It was voluntary from the day it was written.

The standard was never a rule

Between the 1940s and the early 1980s, American clothing sizing was shaped by documents called Commercial Standards, developed with the National Bureau of Standards, the body now known as NIST.

They look official — designations, tables of measurements, committee membership lists, all published by the Department of Commerce. The documents themselves are candid about their status. The text of one such standard states plainly that the adoption and use of a Commercial Standard is voluntary, and describes the purpose as providing uniform bases for fair competition rather than binding anyone to anything.

The committees that drew them up were made up of the trade. The membership lists name garment manufacturers, department stores, underwear industry bodies and the Mail Order Association of America, alongside a handful of academics. This was industry coordinating with itself, hosted by a government agency and published on its letterhead.

That origin explains the demand. Mail order was the driving force, because a catalogue sale fails without a shared vocabulary of size — there is no changing room in a postal order.

Where the numbers came from

The measurement data behind the women’s standard came from a Depression-era survey of American women’s body dimensions, later distilled into the 1958 commercial standard for misses’ sizes. Volunteers were paid to be measured, which shaped who turned up: the survey recorded the women willing to undress for a small fee in the 1930s, not a cross-section of the country.

Two problems were built in from the start. The sample was not representative of the population it would be used on, skewing towards younger, white, and comparatively slim volunteers. And the system derived a whole garment from a small number of dimensions, assuming bodies scale proportionally.

They do not. Two people sharing a bust measurement can differ substantially at the waist and hip, and no amount of grading a pattern up or down fixes that — grading changes the size of a shape without changing the shape itself.

So the standard described a shape that few people actually had, and manufacturers who followed it precisely produced garments that fitted a minority of their customers. Deviating from it was not defiance of a rule but a straightforward commercial correction, and since the document bound nobody, there was nothing to stop them.

1983, when the pretence ended

The standard was downgraded over time and then dropped. As the record has it, use of US standard clothing sizes by manufacturers as official guidelines was abandoned in 1983.

Nothing replaced it with force of law. Standards still exist — ASTM publishes sizing tables that any company may adopt, and revises them periodically — but adoption remains optional, and most companies no longer use them at all.

The word “standard” survived while the thing it named quietly did not.

What the drift actually looks like

The clearest way to see the effect is to hold a body measurement constant and watch the label move around it.

The same body, unchanged, travels from a 14 to a 0 across seventy-odd years of catalogue and retail practice. Nothing happened to the woman in that row. Everything happened to the label above her.

This is what people mean by vanity sizing, and the direction is consistent: the number attached to a given set of measurements falls over time, because a smaller number is easier to sell.

It also creates a one-way ratchet. Once a competitor relabels, holding the old numbering makes a brand look as though its clothes run small, so the pressure only ever points downward.

Flattery is not the whole explanation

Vanity is the popular account and it is real, but it is not sufficient on its own, and the other reasons are more mundane.

Every brand fits its garments to a house model — a specific set of measurements, sometimes a specific person — and then grades sizes up and down from that base. Two brands with different house models produce different garments at the same label, without either intending to mislead.

Target customer matters too. A label aimed at teenagers and one aimed at women in their fifties are fitting genuinely different average bodies, so identical numbering would serve neither well.

Then there is segmentation. Inconsistent sizing raises the cost of switching brands, because a customer who knows their size at one retailer has to rediscover it everywhere else. That friction is not obviously unwelcome to the brand that already has them.

The cost lands on returns. Online, where nothing can be tried on, size uncertainty produces a return rate the industry treats as a permanent expense, and a sizing system nobody can rely on is a large part of why. Shoppers respond rationally by ordering the same garment in two or three sizes, which is expensive for everyone and entirely the retailer’s own doing.

What the label is legally required to say

Here is the contrast that makes the position absurd, and it is worth stating precisely.

A garment sold in the United States must carry a permanent care label. Under the federal care labelling rule that label must be attached so it will not separate from the product and must remain legible for the useful life of the garment.

The required content is specific, and the specificity is the point:

• Whether the item is washed by hand or by machine
• A water temperature, in terms such as cold, warm or hot
• Whether it is dried by machine or by some other method, and at what temperature
• Either a washing instruction or a drycleaning instruction — every garment needs one or the other

So federal law prescribes, in detail, how hot the water may be.

It says nothing at all about what the number on the size label means. The one piece of information a shopper uses to decide whether the garment will fit is the one piece the regulation does not touch.

Even the measurements are not measurements

Menswear looks like it escaped the problem. Trousers are labelled in inches, waist by inseam, which appears to replace an arbitrary index with a physical quantity that cannot drift.

In practice the same pressure applies, because nothing compels the printed number to match the garment. A stated waist is a description of what the maker has decided to call it, and the gap between the label and a tape measure laid across the finished waistband tends to run in the flattering direction.

Attaching a unit to a label does not make it a measurement unless somebody is checking.

Why the fix keeps failing

Attempts to restandardise are made regularly and go nowhere. The reason is not technical difficulty or cost, but that the parties who would have to agree are the same parties currently benefiting from the disagreement.

Body scanning has removed the technical excuse. Large-scale scanning surveys can now describe a population’s dimensions far better than a mid-century tape measure ever could, so the data problem that defeated the 1958 standard is solved.

The obstacle is commercial. A brand’s fit is part of what it sells, and converging on a shared numbering would hand customers an easy comparison and remove a reason to stay.

Voluntary standards fail for a familiar structural reason. A published table that binds nobody influences behaviour only while participants find compliance useful, which is the same weakness that makes the Wassenaar arrangement behind how export controls work a coordination forum rather than a rule.

Some markets have gone further. European standards bodies have pushed sizing built on stated body measurements in centimetres, which at least anchors the label to something physical and checkable. Adoption there is uneven too, and a garment sized to those tables still has to be cut by a manufacturer with its own view of the customer.

What a size label is actually telling you

Treated as a measurement, the number is close to useless. Treated as a brand-specific index, it is informative — it tells you where a garment sits within one company’s own range, and nothing beyond that.

The measurements printed in a size guide are the real content, and they are the only part of the system anchored to a body rather than to a marketing decision. Garment measurements, where a retailer publishes them, are better still, because they describe the object rather than the wearer.

This is also the strongest practical argument for a small, familiar set of brands. The case for a capsule wardrobe is usually made on aesthetics and cost, but the underrated benefit is informational: within a handful of known labels, the numbers regain a meaning they do not have across the market as a whole.

The frustration in the changing room is not a failure of the system. It is the system working exactly as its incentives dictate, in the absence of anyone with the authority or the desire to change them.