Three Exercises to Demonstrate Markets and Agents


  1. You are at a party. Everyone in the room identifies two other people. Call one your wing-person and the other your unwanted admirer.
  2. Your goal is to always keep your wing-person between you and your admirer.
  3. Move around

Part II

  1. This time, you are the wing-person and your former wing-person is the target. Now your job is to keep yourself between the admirer and the target.

Market for Chairs

Our start up is moving to new offices. We've acquired furniture from a supplier. There are all different kinds of chairs. With arms, without. With wheels, without. Soft and padded, firm and rigid. Brown, black, blue. Adjustable and fixed. We'll boil these down to three characteristics each of which ranges between 1 and 5. Each of us has our preferences; I might, for example, prefer a chair that is 5-5-5. Right now, though, I have one that is 1-3-5. I can measure my dissatisfaction by computing the difference between what I have and what I want. In this case it is (5-1) + (5-3) + (5-5) = 6. If you have a 3-3-5 (which would have a dissatisfaction score of 4 for me) we could trade (if my 1-3-5 is better for you).

  1. Compute your level of dissatisfaction with the chair you have been given. Report this number to the clerk of the exercise. After all numbers are reported we will calculate the total and average levels of dissatisfaction. Note that if everyone had exactly the right chair, the total and average dissatisfaction would be 0. If everyone had the most wrong chair the worst the average dissatisfaction would be would be 4.0 (since people would want either 1 or 5 and have either 5 or 1).
  2. Let's mill about and consider chair swapping.
  3. When you meet someone, compare your chairs and your chair preferences and compute your dissatisfaction level with your partner's chair. If you'd both rather have the other person's chair, trade. If one of you would be indifferent (no better, no worse) but the other would be happier, make the trade. If both of you are indifferent or both would be less happy, just move on.
  4. After 5 minutes we'll re-report our individual levels of dissatisfaction and tally up the aggregate and average levels.

Self Organizing Neighborhoods

Once we have settled on chairs, it is time to pick our offices. Now our numbers mean something different. The first number has to do with your noisiness and tolerance of noise. The second has to do with personal habits like smoking. And the third has to do with how social and chatty you are. In order to make a happy workplace, we have decided to let people choose their office based on a desire to minimize the frictions with neighbors based on noise, smoking, and chattiness.


Again, the most similar neighbors will have difference scores of 0 and the most dissimilar will have scores of 12. But you have to manage multiple neighbors. You evaluate the suitability of an office based on your similarity/difference with your three nearest neighbors (next door and across the hall).

  1. First we settle everyone into a provisionary office. Then you look around you and see how much you differ from your current neighbors. Compute differences from each and total.
  2. If the average total mismatch with your neighbors is 2 or less, then you are happy and can stay where you are. If any one of your neighbors is more than 9 different from you or if the average is 6 or higher then you will move.
  3. When the next round is called, if you will move, step out of your office and into an empty one
  4. Repeat.

File nameFile typeSize
chairs-endowments.docxNo description57.15 kBInfo
chairs-preferences.docxNo description58.09 kBInfo
markets-exercise-data.xlsxNo description58.97 kBInfo
Office-Preferences-Cards.docxNo description62.54 kBInfo
Office-Preferences-Master.docxNo description118.12 kBInfo