The Psychology of a Price Tag

AOrthodox economics treats a price as a piece of information and the buyer as somebody who weighs it against the satisfaction on offer. The literature gathered under the label of behavioural pricing starts elsewhere, from the observation that most shoppers have no settled idea of what a given item is worth to them and must therefore judge a figure by setting it beside something else. That something else is called a reference price, and the interesting question, in my view, is not whether references matter but which reference a seller can persuade a buyer to adopt. Sellers have grown adept at supplying one. The mechanisms involved are neither mysterious nor especially numerous, and three of them account for most of what retailers actually do. Whether they amount to a serious departure from the older account, or merely to a set of footnotes attached to it, is a question worth holding in mind while the evidence is examined.

BThe most robust of the three is the sense of fairness that a price arouses, which turns out to depend on the reason a buyer supposes to lie behind it. Rosalind Achterberg, of the Meesen School of Commerce, put a series of scenarios to 2,100 respondents in 2019 and found that 79 per cent judged an increase acceptable when it followed a rise in the seller's own costs, while 21 per cent accepted an identical increase attributed to a shortage of stock. The two increases were the same size and reached the buyer in the same way. Achterberg argues that buyers hold sellers to an implicit bargain under which the seller may protect a customary margin but may not enlarge it by exploiting circumstance. The pattern holds, she notes, only for goods a customer buys regularly, where a habitual price exists to be departed from; for unfamiliar purchases the judgement of fairness has nothing to fasten itself to.

CSellers can also change the shape of a price without changing its size. Splitting a total into a headline figure and a set of additional charges reliably lowers the sum a buyer remembers, because attention settles on the first number while the remainder is discounted rather than added. In a trial run across 96 online sellers over 14 weeks, listings that separated delivery from the item price attracted 14 per cent more orders than listings quoting a single inclusive figure, although the sum eventually paid was identical. It is here, rather than in the more famous experiments, that psychological pricing shades into something the ordinary use of language would call deception. Sellers defend the arrangement on the ground that separating a charge makes the charge visible, which is true, and that buyers are free to compare the components, which they seldom do. The practice has accordingly attracted regulators, whose interest tends to arrive some years after a technique has become standard.

DThe third mechanism concerns the company a price keeps. Presented with two options, buyers divide unevenly between them; presented with a third and conspicuously expensive one, they migrate towards the middle, which now looks moderate rather than costly. A chain of 38 restaurants that introduced a premium dish in 2017 found that orders of the second most expensive item rose by 26 per cent within a month, while the new dish itself was chosen by 3 per cent of diners. The dish, in other words, was not there to be eaten. Effects of this kind are strongest where comparison is difficult, which is to say in restaurants, on aeroplanes and in shops selling goods nobody buys twice in a year. Where a buyer can set two sellers side by side, the middle option loses most of its magnetism. Achterberg maintains that the mechanism is best understood as a shortage of comparison rather than as a defect of reasoning.

ENot everyone believes these findings deserve the attention they have received. Ivo Papadakis, an economist who has spent much of his career studying retail margins, contends that the profits credited to psychology are in the main the ordinary returns to segmentation and to the trouble buyers face in searching. His analysis of 1,900 product lines between 2015 and 2018 found that a seller's margin was predicted far better by the number of competitors visible to the buyer than by any feature of the way the price was displayed. He dismisses the more theatrical demonstrations as artefacts of the laboratory, where a participant has nothing at stake and no alternative supplier to turn to. Papadakis concedes that his own data are drawn from online sales in two categories only, cover nine months of trading, and cannot capture what happens in a shop where the buyer sees no comparison at all, and the limitation is not a small one.

FRegulators have begun to insist that the figure first shown to a buyer be the figure eventually paid, a requirement that is easier to write than to police. Rules of this kind address the shape of a price rather than the conditions that make its shape effective. The evidence, taken together, supports a modest conclusion rather than a dramatic one: the effects are real, considerably smaller in shops than in laboratories, and largest wherever a buyer cannot easily compare. That last clause seems to me the important one. If it is right, the most useful regulation would not prohibit particular devices, which sellers replace as fast as they are banned, but would attack the conditions that give the devices their power, by making comparison cheap and unavoidable. Whether any regulator has the appetite for so indirect a remedy is another matter entirely.