Consider the following price-determination theory, which was suggested in a recent experimental economics class: "rank all buyer values from high to low and find the median value. Then rank all seller costs from low to high and find the median cost. The price should be the average of the median value and the median cost." By giving buyers a lot more units than sellers, it is possible to create a design where the median cost and the median value are both equal to the lowest of all values and costs. Use this idea to set up a supply and demand array where the competitive price prediction is considerably higher than the prediction based on medians. (If your design has a range of possible competitive prices, you may assume for purposes of discussion that that the average price prediction is at the mid-point of this range.