THE TWO TOMATO CONJECTURE
Tomatoes have been on sale in the street markets of the Mediterranean world for two centuries or more. Solon has observed that it is quite common for the stalls that are selling them to have tomatoes on sale at two prices. Solon has also observed that it is quite common for the two piles to be replenished from the same set of baskets, or nowadays boxes; usually with some degree of dissimulation. He further notes that the higher price pile often sells rather faster than the one with the lower price. Solon does not challenge such fascinating phenomena, he observes them. There seems to be two standard responses on the rare occasions when customers question what is going on. One is “Which do you want?” “If you don’t want to buy tomatoes, - - - - - - off.” The other is “We keep some at a special price for our regular customers.” That often draws the comment that they are available for everybody. The stallholders respond “Of course we want everybody to become regular customers, don’t we.” That non sequitur baffles the questioner.
What is going on, of course, is that tomato sellers have realised that a worthwhile number of their customers judge the quality of tomatoes partly by their price. Selling some more expensively generally shifts more tomatoes at a better margin. Because the use of price as one of the indicators of relative quality in what is for sale is a widespread habit,[1] similar things occur in all sorts of transactions across the world. Sometimes the consequences are extreme. Various bemusing instances are quoted in the footnotes of the literature of marketing; and one case in his personal experience amused Solon a few decades ago. He lived for a year or more close to a shop selling second-hand odds and ends cheaply - a run-down junk shop. One object permanently in its window caught Solon’s eye because he could not imagine what its purpose was. It was large, ceramic, ornamented and ugly. The shop closed and reopened selling antiques. The mysterious object reappeared in the window, washed and with a new ticket at nearly twenty times the old price. It sold in a few weeks.
One consequence in the tomato market is that from time to time a shopper comes home [2] and says. “I will have to buy the tomatoes tomorrow, dear. There were only the cheap ones left today.” In the particular case he will probably pay more tomorrow for the tomatoes he rejected today. In the general (static) case, the market does not clear; and there is no reason to expect it to clear when quality is partly assessed according to price.[3] As implied by the case of the mysterious object which crossed Solon’s path, in the presence of judgement of quality by price, the market need not clear even when there are no two piles of the same goods at different prices.
Modern, as opposed to traditional, retailing apparently has not realised the scope for exploiting the phenomenon. Solon has sometimes marvelled at the quantities of apparently potentially saleable perishables that appear in the refuse containers of modern supermarkets. He was enlightened about at least part of that waste when standing by the fresh fish counter of a very modern hypermarket. A large stock of trout was displayed, which caught Solon’s eye because he has a Mediterranean taste for good, fresh fish. He noted that it was priced much below other varieties of fish. An elderly couple who appeared far from rich approached the trout. In a short conversation they convinced one another that if the trout was that cheap, it could not be as good as it looked. They bought other, more expensive fish that Solon judged was of poorer quality. Before they had finished their purchase, another similar couple speaking a different language approached the trout. The conversation between the second couple was on the same lines as the first. They also bought more expensive, less fresh fish of another variety[4]. The hypermarket was going to have a lot of trout on its hands at the end of the day; and an even cheaper “ Special Offer” would not dispose of it.
The two right responses for the hypermarket’s fish counter manger were, of course, either to take all the trout off display for an hour and bring it back at a substantially higher price; or to split the display of trout on the two tomato principle. Solon has encountered such double displays of fish on fishmarket stalls, but that level of flexibility and sophistication appears to be beyond the scientific retailers.
To be fair to the scientific retailers, they have responded to price being used as a judge of quality in at least part of their business. They had introduced own brands when the realised that they could obtain supplies from many manufacturers in bulk under the store’s own label, and sell these cheaper than the manufacturers’ brands while spending much less on promotion. Many of them have now split their “own brands” into very cheap lines (often labelled Basic) and Premium lines priced only slightly below the main brands and presented very similarly. The Unit judges that in many cases the actual Basic and Premium goods in the own brand wrappers are so similar in quality that if, by mistake, they were presented in each others wrappers, sales would not be affected.
The scientific retailers were not quick about adopting this tactic. Detergent manufacturers had started introducing Economy brands beside their main (now Premium) brands decades earlier; partly in response to the first significant retailers’ own brands. The difference in manufacturers’ costs between the Economy and Premium brands is almost wholly in the lower promotion expenses of the Economy versions. Brand managers are under a perpetual tension to find ways of attracting customers through lower effective prices without sacrificing the quality associations of a higher ticket price. This is the origin of the “Free X% more” , the “3 for the price of 2” and many similar offers. Car manufacturers’ discounts in place of price reductions are a less advanced attempt to achieve the same end.
All these offers and other manoeuvres by scientific marketers to exploit the habit of judging quality in part by price are surprisingly lacking in theoretical foundation. How to do it is certainly taught in Marketing courses, but not much is said about why and when one should exploit the phenomenon. Solon was delighted to encounter a paper which begins to establish an adequate framework for and underpinning of these tactics[5]. He confidently expects it and its successors to be the established wisdom of marketing schools in 20 years time, and somewhat later to affect the behaviour of hypermarket fish counter managers. He trusts that it will never influence the traditional market stallholder.
The theoretical development that Unit economists would really like to see goes deeper. It involves giving up the idea of competition to supply a market efficiently in favour of the concept of competition to extract the rents in a market most efficiently. The basis of this thought is that for markets to work efficiently in the classical model, they must be capable of clearing. Markets where quality is judged partly by price are not necessarily, or often, capable of clearing. The classical model therefore does not apply. However competition does clearly operate in these markets, driving down profit towards the acceptable minimum. The nature of part of this competition is clearly competition to extract the rents available because buyers judge quality (in part) by price.
The Unit regards it as inevitable that buyers faced with a wide choice will use price as one of the indicators of probable quality. The cost of assembling good information on all the possibilities in a wide range of choice are high. Price is usually cheap, easily available information. It is rational to use it as a filter to hold down total costs of gathering information. A clear, though to Solon surprising, example of this is readily available on the internet. Many internet real estate agents offer a very wide range of properties. They offer clients the option of filtering the choice with a minimum price as well as a maximum price. Solon would never use that option; there are large possible gains from filtering on the other factors until you find a fully suitable property which if you are lucky will be at a price that is unexpectedly low. Nevertheless, the option is offered and used, presumably because the habit of using price as an indicator of quality is so common.
In the case of routine purchases where the range of choice may not be wide, buyers need to minimise the recurrent costs of decision. In the absence of other immediately available quality indicators, it is again rational (if not necessarily optimal) to use price in all areas where the buyer judges that price has some positive correlation with quality. Market stall tomatoes are a case in point.
Brands are attempts to establish immediately available quality indicators in order to extract the rent available from buyers’ need to hold down their costs of purchase decision. All Unit members agree on the usefulness of brands to them in this role[6]. Most brand managers struggles to avoid an image of low price and therefore low quality for their brand are noted above. For luxury brand managers one question is always “Is my advertised price high enough to adequately support my image of high quality (and of conspicuous consumption) while my actual discriminating prices maximise profitable sales?” For economy brand managers, the question is “How far is it profitable to use promotion to add an image of quality to my brand, so increasing sales?”
These are questions about extracting available rent most effectively from a market. The introduction of a new model or new variant of a good or a service is another exercise towards the same end. The aim is to create a rent and then extract it. The feats in rationalising the chain of supply by Dell and Wal-Mart are exercises in identifying rents and extracting them more efficiently. The creation of financial derivatives is aimed at identifying a rent and extracting it from a market which might have been deemed fully efficient and rent free in a classical analysis. The Rotterdam spot market for petroleum is a very efficient arbitrager, but a great deal of the profits of its participants come from extracting the rent in bringing fringe supplies to market. And so on through the whole gamut of market competition. The phenomena of the market are, it seems to the Unit’s economists, much more comprehensible if considered from this viewpoint.
The Two Tomato Conjecture is therefore formulated as:
“Rents are always present in markets where some buyers use price as an indicator of quality, and it appears that rents are usually possible in a great variety of other markets. The efficiency of market competition is therefore to be found in the competition to extract actual and possible rents from markets most effectively. Most actual markets cannot clear in the sense of eliminating rents and therefore have no static equilibrium; but competition to extract rents does lead to a dynamic equilibrium. Competition to extract rents is therefore the preferred way to analyse markets”
The first lemma, also a conjecture, is:
“Equilibrium in competition to extract rents will generally take the form of a Nash equilibrium.”
The second lemma is:
“Given the first lemma, the general description of the form of market dynamics will include a cloud of evanescent Nash equilibria as the market evolves.”
The Unit expects that it will prove intrinsically impossible to fully identify one of these evanescent equilibria in practice. A sort of uncertainty principle will operate. But the Unit also suspects that the cloud as a whole will have interesting and describable characteristics. A Nobel Prize may await the economists who first formulate that description.
* * * *
Should you wish to comment, an email to solon@use-solon.org may draw a response.
1.The earliest respectable academic discussion of the phenomenon that the Unit has chanced upon is “Some Consequences of the Habit of Judging Quality by Price” , Tibor Scitovsky (on whom be peace), 1941, Review of Economic Studies, Vol. 12, no. 32.
6.Most of them also agree with Solon that the recent success of brand managers in getting buyers to display their logos and brand names on articles of clothing is going too far. Solon is explicit: he may display brand names on his person for brand managers who pay him sufficiently to do so. He will not pay them a penny for that anti-privilege.
