In response to my entry Why have retailers changed clothing sizes?, Alaina asks:
Can you explain why/how early misses sizes were described a “size 14 years?” surely that was not a coincidence with the sizing system you describe in paragraph 5 (which, I agree, sounds completely plausible)
In the absence of my familiarity with your source material, I would reason that this was the beginning of the push back; when retailers began to wrest control of sizing from manufacturing. That it coincides with the advent of children’s sizing at retail is significant.
Previously, the divisional drafting system (“scale”) adopted for pattern making for production was limited to adult clothing. It only worked for figures that were mature because the scale relies on divisions of 8. [As an aside, don’t ask me about the imperial measuring system, specifically the inch unless you want a lengthy discourse. I prefer metric but for clothing drafting, no measurement increment has proven of greater utility than the inch]. The key landmarks of infants and children’s figures do not divide evenly by 8. Infant figures are divisible by 4, children by 6 etc. Thus, the divisional scale drafting system couldn’t be used for physically immature bodies.
Now this was okay considering the marketplace of the era because the making of children’s attire was not a widespread commercial activity at that time. Most children’s clothing was made by mothers, often cut down from adult clothing (again, it’s no coincidence that children were dressed no differently from adults once they left infancy). However, once the economy expanded with increasing specializations in both occupations and lifestyle, the former niche production of children’s clothing became a significant market segment all its own. It is possible that the reference you cite serves as a commercial indicator to differentiate children’s clothing vs scale derived adult’s wear. It is likewise likely that it was due to this confusion (children’s clothing sizes were or are largely indicated by age) that adult size numbers derived from scale were gradually abandoned.
It was at that time, when production of children’s attire became an appreciable commercial activity, that greater recognition of needing to manage sizes but more specifically size breaks, was realized. It would be helpful to keep in mind that this development was largely ad hoc. Without precedent, nobody knew what to do; everybody looked to their closest competitor to see what they were doing, to rationalize it to see if it was a logical system standard they could also use. There was no grand plan or overarching system anyone could follow. They made it up as they went along, no different than today. It is purely reactionary, reacting to the market. Consumers miss the point entirely if they think we lead anything.
The awareness of the need to manage sizing didn’t reach crisis proportions until after WW2. Again, that sequence of events is not coincidental (divisional scale was adopted and refined circa the civil war, necessary for uniform production etc) and had much to do with expansion of the commercial sector post war and indirectly, also impacted by Eisenhower’s initiative of what became the Interstate Highway system because for the first time, clothing production expanded beyond regionalism. [It is to our continuing detriment that the impact of retailing beyond a “natural” market scale (I’m at a loss to define, a continuing quest) is not recognized for its influences and often, impediments in addressing the needs of consumers better. No need to thank me for abrupt abandonment of the influence of economic geography in the development of commercial markets, I’ll move on now.]
Most of what I can derive about post war mass retailing sizing practices was from Disher’s American Factory Production of Women’s Clothing. Conflicts were readily apparent at that time with a much broader range of sizing than we have today. Sizes were sorted by category types and numbers. Some categories were “Precision size, Tween size, Miniature Miss, Diminutive, Short Cuts, Happy Mediums, Mid-size, Average Size, Petti-size, and last but not least, P.A.T (Perfect American Type)”. The numbers designating sizes particular to each category were likewise different such as 12A-20A, 10D-20D, 10S-20S, A.A.12-A.A.20, AH8-AH18, 12B-18B and 10+ to 20+.
My point being, as non-descriptive as sizing is today, it used to be much worse. There is no doubt that retail wielded influence to simplify sizing if only to reduce the range of inventory they were required to carry, to reduce consumer confusion as well as the cost of fixtures. I do not think consumers understand the considerable costs of retail sizing fixtures and its influence on needing to simplify sizing. That we no longer remember the complexity of antecedent sizing systems does not mean we did not once have greater problems.
You did not ask me for my opinions on how to resolve these problems but I believe the solution can only come from indirect cooperative strategies between manufacturers, retailers and consumers.
Consumers need a greater understanding of how manufacturers develop sizing particular to their niche. Just because today’s consumer has the means to purchase goods from a far flung enterprise does not hold that the goods were produced with their local sizing needs in mind. The larger the entity, the more generic their sizing must be because along with larger market share, comes the financial need to appeal to the broadest swathes of the market. Vertical manufacturers will have the most generous sizing because it is the nature of their enterprises. Expecting another outcome is no different from admonishing a cat to refrain from hunting mice.
The best solution for consumers is to frequent smaller retail outlets which carry more specialized (sizing refined) product lines. It also holds that it is unreasonable for consumers to expect smaller producers to price goods commensurate to larger multinational firms (the inflation adjusted cost of apparel continues to drop). Economy of scale is only one facet; smaller enterprises endure proportionately greater costs in refining and targeting their sizing too. They do a lot more testing and sampling; the minimal cost of sample sewing -not including pattern development- is three times the cost of production.
Retailers have an unenviable difficult task because they often cannot purchase goods from smaller companies. These reasons are due to too small lot sizes available from smaller manufacturers, standardization that isn’t congruent with their private label brands (as explained in the entry you commented on), corporate purchasing policy, the need to cross merchandize (ibid), the lack of sophistication on the part of their suppliers as to the need of professional packaging, fulfillment and EDI and lastly, their payment policies to manufacturers generally suck. Retailers are very slow pay, typically 6-9 months so one needs to be factored in order to sell to them. The costs of factoring are so high that most manufacturers need to scale up, reduce costs (more generic sizing) and move off shore -it’s a double edged sword.
Manufacturers could do a better job of communicating their sizing parameters to consumers with a specific hang tag and legally required descriptions, no doubt of it. Many don’t because they don’t want to eliminate possible sales or they know their sizing so intimately or think it is “standard” that there is no need to do it. People lose sight that manufacturers are people too; meaning they can be just as delusional as the next guy. I know of one designer who sincerely believes she’s a size 6 (all her sizing is based on her figure) and she can’t weigh an ounce under 180lbs. And then, if consumers cannot agree on what constitutes a given size, manufacturers cannot either. Tragically, some designers don’t even know how to measure garments, no one is required to have minimal competencies to become a manufacturer.
In summary, consumers don’t realize just how complex this is. They remember mom sewing at the kitchen table and think there’s nothing to it. The apparel industry is like peeling an onion, it becomes increasingly complex the more you get into it. Its variables and inter-related dependencies are far more complex than is given credit for, requiring a great deal more mastery and intelligence than is realized. If the issue of sizing were as simple as everyone believes -reduction to a simple sound bite- then the popularity of the “vanity sizing” cultural myth would not be so prevalent.
All right! Next question, perhaps for your next post: can you explain why in the 1920s the most common size range advertised was 14-20, in the 1940s it was 10-20, in the 60s it was 8-18, in the 70s it was 6-16, and in the 80s it was 4-14?
No, I cannot.
Your source material is highly problematic in that it cannot be considered representative of the market. Yours was a survey of advertisements from one month each year from one magazine -and today while it’s unarguably the biggest or most influential one but it wasn’t true of the era. In any event, its advertisers could not be construed as representative of the market then anymore than advertisers today would be considered representative of the market now.
Perhaps this can be illustrated with two anecdotes that I learned in statistics classes:
The first was that a telephone poll in 1932 predicted that Herbert Hoover would win the presidential election that year. FDR won handily – what could have gone wrong? The clear answer was that people who owned telephones in 1932 had different voting habits than those who didn’t. Another similar story: supposedly Pauline Kael once said that she didn’t know how Nixon had won (in 1972) because nobody she knew had voted for him.
Just off the top of my head, I can think of a number of problems with using advertising in Vogue as a proxy for actual sizes. First, there is Vogue’s demographic and how it may have shifted over the years. Several magazines have come and gone over that period, from McCalls (of Seven Sisters Fame) to George, each with an appeal to some demographic. Second, the advertisers: not all of them advertise sizes, so you are only selectively sampling from them. Third, the advertisers’ motives: were they trying to dump stuff, appeal to a niche, or just stating what they had? Any of these sampling issues is subject to the same problems as telephone polling in 1932, or sampling the friends of some New York Democrat in 1972. All of them together spells statistical nightmare.
So what do we do when the smallest size (australian 6) is still far too big?