I have long struggled with the logic of applying the aggregator model to price-insensitive goods like luxury, and it was only a year or so ago that I understood the source of my confusion. (The reason is so simple that I—like T. H. Huxley when first grasping the dynamics of Darwin’s theory of evolution—thought to myself: How extremely stupid not to have thought of that!)

Essentially, my realisation was that luxury economics work in reverse to normal economics. While this is what makes them powerful, it also suggests that any aggregator model applied to those economics might also work in reverse.

Useless Veblen goods and the honest signalling of wealth and status

As Thorstein Veblen realised and wrote about in his treatise, The Theory of the Leisure Class, luxury turns everything that we thought we knew about supply and demand on its head: Instead of becoming less desirable as the price goes up, luxury items become more desirable (and often the more so the more expensive they are). I think this is because luxury is synonymous with excess (i.e. the opposite of necessity), so that ‘luxury’ is that which nobody needs. As a result, the consumption of luxury goods says something about the consumer, meaning that the consumption of luxury products serves a signalling function. 

In biology there’s a concept known as ‘signalling theory’, where signals are thought to be either honest or dishonest. Honest signals are the most valuable since they can only be shown by those organisms able to afford it. Examples of such honest signals would be the gaudy males of some sexually dimorphic animal species, where traits like the male peacock’s tail have evolved because females like gaudy males. The reason why females like gaudy males is because the extravagant tail makes the male conspicuous. This conspicuousness further becomes an honest signal of health, since only strong and healthy males could maintain the tail to a high standard and afford to be so conspicuous to predators. (Another example is the warning colouration of venomous animals, where the more venomous an animal is, the more honestly conspicuous it can afford to be.)

When it comes to humans and luxury, the conspicuous consumption of luxury goods communicates to others that the conspicuous luxury consumer has both an elevated social status and that they enjoy superior access to material resources. This is because social status and material wealth are both metrics that humans find attractive and that many people, therefore, want to maximise and flout. As such, the value of a luxury good lies less in the value of the good itself and more in the signalling function that access to such goods provides. Therefore, while only the richest people (at least historically) have been able to afford luxury, the desire for such consumption is near-universal. 

This desire (combined with the efficiency of the signalling), however gives rise to dishonest signalling (which, in the biological world, is known as mimicry), where people without the wealth or resources required to afford luxury good procure these (or something like them) to gain the implied social status. Dishonest signalling however cheapens the signalling function of associated objects, like luxury brands, and for this reason brand equity is best protected by keeping prices high. This is probably also the reason why luxury goods are Veblen goods (i.e. becoming more desirable the more expensive they are), because the more expensive the goods are, the more honest the signals become. 

The world’s luxury brands are therefore likely the monetary beneficiaries of the human need (and demand for) the conspicuous signalling of wealth and status. The desire for such goods creates a supply-demand gradient that the luxury brands can profit from selling into—up to a point, and the CEO of Ferrari China [link] explains the conflict of managing a luxury brand well when he says:

The protection of the residual value, the pre-owned market, for us, is a priority. As the founder of the company always said—and I want to repeat it, because it very clearly explains the philosophy of the company—we should always sell one car less than the demand; we will never sell one car more than the demand. … We never push. Ever. For us, the most important thing is to always have the correct amount of cars in the market. Even if we can sell more, we refrain from selling, because we have to protect the exclusivity of our clients … This is a key element of the business model.

In other words, luxury brands can amplify the demand gradient that they feed on by artificially reducing supply so that it doesn’t meet demand, and to so keep prices (and demand) high. This dynamic is unique to Veblen goods as they are the only goods to see demand scale proportionally with price (rather than inversely). This is because their function is price-dependent in a way that no other types of good are: If a luxury item wasn’t expensive, then it wouldn’t be a luxury anymore. Or, put differently, the point of a luxury item isn’t the item itself, it is signalling power of that item as a function of the price. The higher the price is, the greater the impact of the signalling. As the price goes up, the power of the signalling goes up too. This is the exact dynamic required for reverse price-sensitivity to work, namely, that the price is the real product. This means that the highest-quality ‘products’ are the most expensive. 

Ultimately, the power of the economic curation that is engaged in by luxury brands is rewarded through the erection of deep moats (built from sales discipline and investments in brand equity), all with the intent to keep prices artificially high and to ensure the longevity of demand. When done well, this strategy results in the emergence of outlier (‘star’) brands that have less to do with traditional ideas of luxury (e.g heritage, tradition, artisanship, etc.) and instead exist more as purely cultural phenomena with sufficient momentum to create the positive feedback loop of fame that drives luxury economics: Where the price is high enough so that one’s possession of the item implies a specific degree of material wealth and that others (not just the conspicuous consumer) know just how much must have been paid.

Alice in Wonderland and the reverse aggregator model

Because luxury brands are used for signalling, it also becomes obvious that the better-known a brand is, the more effective the signalling will be. This makes a well-known brand or a logo pretty much required. The Internet can further help to amplify this dynamic, as it’s a medium as-if-built for maximising fame by amplifying reach; allowing previously local brands to become global superstars. Therefore, even if it’s superficially counterintuitive, the cultural relevance and fame of the existing star luxury brands might actually increase once the Internet is thrown into the mix. If so, rather than the Internet acting as the ‘great equaliser’ it has been in every other industry, in the luxury industry it might actually help the pre-existing star brands to grow more and more famous; further cementing their outlier status and allowing them to gobble up more and more of the industry money flows. These are classic free-market dynamics in action, where the rich get richer and where the big (or famous) get even bigger and better-known. 

In every other industry, the Internet has of course powered exactly the opposite development, where generalists and aggregators have scaled to never-before-seen levels of success at the expense of the brands they are aggregating. So, what gives? Of course, the missing piece of the success-equation comes from the upside-down dynamics that characterise Veblen goods: While generalists have thrived in commodified industries (where the lowest common denominator is price and since everyone loves a bargain, the generalists can leverage economies of scale to lower prices and to so win sales), the luxury industry (which, famously, does not thrive on bargains) cannot make use of this dynamic since luxury—by definition—cannot compete on price. In other words, in the luxury industry, the economic benefits of increasing scale (as afforded by the Internet) are likely to accrue to the players that are the best able to compete on fame (rather than price). 

Now, if we were to accept the premise that fame (and not price) is the driver of growth in the luxury industry and then run the resultant argument to its logical conclusion, the Internet could—counterintuitively—make the luxury industry less competitive: Because technological innovation typically focusses on lowering prices (by making more of something, or making it cheaper), the luxury industry will be uniquely hard to disrupt. Instead, disruption in the luxury industry will likely feed on fame (i.e. being able to build it faster and more sustainably or better and more efficiently), but since fame also (famously) begets fame, the luxury disruptors might—provocatively—actually be the incumbent brands themselves as the Internet has magnified their pre-existing autocatalytic potential. If so, the winner-take-all dynamic would likely not accrue to a marketplace or other aggregator (because their power over the brands would be limited), but would, instead, accrue to the star-brands themselves.

Consider, for example, quotes like the following from a transcript of a presentation at Kering’s 2019 capital markets day [link; p. 6]:

When we look at luxury … we do not see the emergence of digital native brands as much as we’ve potentially seen in other industries. When you look at the ranking of the top luxury brands, these are the same brands. … So, less disruption on the brand side … ¶ … net-net, the luxury industry is subject to that digital revolution as well. Probably not as massively as other industries, and probably driven by the fact that we’re creating unique products that are impossible to replace, and also we are controlling our distribution. ¶ … ¶ What we see in [the] online market really mimics what’s happening in offline retail

In other words, in the Alice-in-Wonderland world of luxury, the luxury conglomerates themselves might actually fulfil the role that Amazon and kin has played everywhere else.

Presupposing everything above, the opportunity available to luxury ‘disruptors’ could therefore look very different from the (otherwise generally defensible) copy-paste ‘platform + Internet = success’ arithmetic. As a result, there is the potential for aggregators in this upside-down industry to actually become more reliant on the brands that they are aggregating, rather than less (as has been the case in every other industry with more ‘normal’ economics). If so, the bigger a luxury aggregator becomes, the greater the aggregator’s needs to actively disintermediate the luxury brands would be, almost like an aggregator model running in reverse.

Given this, the most successful strategy for hopeful luxury disruptors might not actually be that of an aggregator, but that of an anti-aggregator (e.g. becoming a brand). For the luxury industry, this would mean that wannabe aggregators would need to give up on the misleading ‘platform = disruption’-playbook to instead roll their sleeves up and acquire the portfolio of brands required to play the game as the game is currently being played. Such a strategy could allow the aggregators to claw back some of the power that they’re otherwise potentially helping third-party brands accrue (which actively, but unintentionally, undermines their own competitive position).