Data are collected by means of an economic experiments. The description of the experiment is as follows: We ran two experiments: a posted offer market experiment with three treatments (Experiment 1) and an individual choice experiment with two treatments and with a posted offer market frame (Experiment 2). A posted offer market setup corresponds to the reality of retail markets where sellers post prices and buyers simply decide whether and how much to buy at the given price. Both experiments involved the same two pairs of products (S1 and C1 or S2 and C2), two trial periods using an example product and four phases; each phase had 10 independent trading periods. In Experiment 1, subjects were randomly assigned to the role either of seller or buyer while in Experiment 2 all subjects were buyers. They were handed instructions, questionnaires, and consent forms. After they read the instructions they answered the questionnaire and if they had any doubts they could ask for clarification. When all the participants were ready, after they did the two trial periods, the experiment started. In the trial periods we employed an example product, which was the same across sessions. The reason why we used an example product is two-fold. Firstly, we did not want to disclose any information regarding both lotteries. Secondly, we wanted to avoid any possible anchoring effect to the outcomes occurred in the trial phase that could have affected buyers’ decisions. In both experiments we used ‘points’ as the experimental currency (the conversion rate being 975 points to one pound). Experiment 1 involved 3 treatments: B (Baseline), IS1 (Informed Seller with one product on sale), and IS2 (Informed Seller with two products on sale simultaneously). The key difference relative to Experiment 1 was that this was an individual choice experiment. Since there was not a seller, prices were randomly generated from a uniform distribution. Half of the subjects faced high prices ranging between 75 and 95, the other half faced low prices ranging from 45 to 65.5 Subjects knew that prices were randomly generated, but they knew neither the range of the distribution nor that the distribution was uniform. Buyers could buy any quantity they desired at the stated price. Experiment 2 involved two treatments: IC1 (Individual Choice with one product on sale each period) and IC2 (Individual Choice with two products on sale simultaneously). The IC1 treatment was an individual choice treatment with only one product on sale in each period. As in the B and IS1 treatments, subjects faced either the simple product in phases 1 and 2 and the complex product in phases 3 and 4 or vice versa. The IC2 treatment was an individual choice treatment with both products (S1 and C1 or S2 and C2) on sale throughout the session. As such, it was the counterpart of IS2 treatment. This is an economic experiments that has the purpose of understanding whether product complexity affects competition and consumers in retail markets. We are unable to detect a significant effect of product complexity on prices, except insofar as the demand elasticity for complex products is higher. However, there is qualified evidence that complex products have the potential to induce consumers to buy more than they would otherwise. In this sense, consumer exploitability in quantities cannot be ruled out. We also find evidence for shaping effects: consumers’ preferences are shaped by past experience with prices, and firms may in principle exploit this to sell more.

Economic Experiment