Market experiment design

Design document for disclosure market experiment
Purpose: To study the effect of disclosure regulations on market outcomes including price complexity
and shrouding, consumer and producer surplus, and volume of goods sold.
Experimental overview: We propose a series of studies in which participants will act as buyers or sellers.
Sellers will choose the pricing structure for their product and are compensated based on profits
(quantity sold x price). Buyers will choose to buy a good from one of the sellers. They are compensated
based on the product’s quality (which is the same across products) and the price paid. Markets in
different experimental conditions will vary in terms of disclosure rules, that is, the information that is
automatically provided.
Our proposed experimental conditions are as follows:
Full information condition (baseline)

Price-1 condition

Comprehension condition

Performance condition

Sellers will choose values for each of 10 prices. All of those
prices will be revealed. Each buyer will take a
“comprehension” quiz, which requires them to recall those
prices. Buyers who perform well on the quiz will receive a
small bonus.
Sellers will choose values for each of 10 prices. Price #1 will
be revealed. Buyers can search for information on the other
prices for 10 tokens each; if they search, the prices will be
revealed in order. Each buyer will take a “comprehension”
quiz, which requires them to recall those prices. Buyers who
perform well on the quiz will receive a small bonus.
Price-1 condition, plus: For each buyer that does not
successfully recall at least 5 fees, the corresponding seller is
fined 50 tokens. The buyer will lose the purchase.
Price-1 condition, plus: Each seller has a 25% chance of being
“audited” each round. Sellers whose products have a total
price above a given value are penalized 3 times the
difference from the value.

Timeline
Repeated 4 times, with different
order of disclosure conditions for
each session
Informed
General
Condition Practice 10 game
consent
instructions
specific
rounds
rounds
[In person, (applicable instructions
(as
from
to all
buyer
contractor] disclosure
and
conditions)
seller)

Survey,
CFPB
Payment
including, risk
disclosure
[In
preferences,
(including
person]
numeracy,
metaand basic
experiment)
demographics [In person]

Experimental detail:
 Disclosure conditions: We will run three markets per session. In each session, each market will
experience all four disclosure conditions. We will track the order of disclosure conditions across
sessions, making sure that all 24 combinations are played once.



Market:
o Each market will be comprised of 2 buyers and 2 sellers. Participants will be randomly
assigned to a market group each round.
o Additionally, the role of buyer and seller will be randomly selected each round.
o All participants will be anonymous.

General Instructions (Applicable to All Disclosure Conditions and Roles)
Introduction
Values may be modified to allow for problems that arise during pretesting.
Welcome to our research study.
In this study, you will be playing 40 rounds of a game. Each round, participants will be assigned one of
two roles: a buyer or a seller. Sellers will decide on prices for an object they are trying to sell. Buyers
will decide which seller’s object they want to buy. You will be randomly reassigned to being a buyer or a
seller each round.
The game is divided into four parts, each of which lasts ten rounds. There are different rules for each
set of 10 rounds. We will provide more instructions about each of the parts as we reach them.
Paperwork Reduction Act
According to the Paperwork Reduction Act of 1995, an agency may not conduct or sponsor, and a person is not required to
respond to a collection of information unless it displays a valid OMB control number. The OMB control number for this
collection is 3170-0048. It expires on 12/31/2017. The time required to complete this information collection is estimated to
average approximately 1 hour per response. The obligation to respond to this collection of information is voluntary. Comments
regarding this collection of information, including the estimated response time, suggestions for improving the usefulness of the
information, or suggestions for reducing the burden to respond to this collection should be submitted to the Consumer
Financial Protection Bureau (Attention: PRA Office), 1700 G Street NW, Washington, DC 20552, or by email to [email protected].

[next]
First, let’s review what will happen. In this game, each participant will be assigned one of two roles: a
buyer or a seller. All roles will be reassigned each round.
Every time you play as a seller, you will be trying to sell an object worth 500 tokens. You will
choose prices for the sale of this object.
Every time you play as a buyer, you will choose which seller’s object you want to buy. You will
also be asked to remember the prices for the object that you bought.
You will be paid based on the choices that you make in this study.
Sellers will earn more money when they sell a greater number of objects at a higher price.
Buyers will make more money when they buy an object at a lower price.
[next]
You have just been assigned to your group for the first round. We will describe the rules for this round
on the next few pages. There will be new instructions after 10 rounds.
[next]

Seller instructions
Contains some condition-specific text. This text should be displayed for each participant as appropriate.
Imagine that you are playing as a seller. Each round, you will be placed into a group with two buyers
and one other seller. You will be asked to sell an object that has a value of 500 tokens. The other seller
in your group will be selling an identical object with the same value.
Your job is to choose prices for this object. You have the chance to set 10 prices for this object. Each
price can be set between 0 and 100. An example is shown in the table below.
Price 1
Price 2
Price 3
Price 4
Price 5

26
64
72
0
99

Price 6
Price 7
Price 8
Price 9
Price 10

25
13
0
83
17

Full information condition: Buyers will see all of the prices that you have assigned. If a buyer chooses to
buy your product, they will be asked to report the values that you set for all 10 prices.
Price-1 condition: To start, buyers will see only the price that sellers have assigned to Price 1. For
example, if you used the prices above, buyers would know that price 1 for your object was 26. In order
to see the other prices, buyers will have to pay a fee of 10 tokens per price. If a buyer chooses to buy
your product, they will be asked to report the values that you set for all 10 prices.
Comprehension condition: To start, buyers will see only the price that sellers have assigned to Price 1.
For example, if you used the prices above, buyers would know that price 1 for your object was 26. In
order to see the other prices, buyers will have to pay a fee of 10 tokens per price. If a buyer chooses to
buy your product, they will be asked to report the values that you set for all 10 prices.
There is a 50% chance that the buyer’s report will affect their purchase. Specifically, in 50% of the
rounds, the buyer’s report will get checked. If they have not exactly report at least 5 of the prices on
this report, you will lose 50 tokens and they will not buy the object. If they exactly report 5 or more of
the prices, they will purchase the object and you will be paid as normal.
Performance condition: To start, buyers will see only the price that sellers have assigned to Price 1. For
example, if you used the prices above, buyers would know that price 1 for your object was 26. In order
to see the other prices, buyers will have to pay a fee of 10 tokens per price. If a buyer chooses to buy
your product, they will be asked to report the values that you set for all 10 prices.
In addition to this report, there is a 25% chance that your pricing scheme will be “audited.” In the case
of an audit, if the sum of your prices is above 500, you will lose three times the difference between your
price and 500. For example, if the sum of your prices was 600 tokens and you were audited, you would
lose 300 tokens [300 = 3 * (600 – 500)]. If the sum of your prices was below 500, there will be no effect.
All conditions: For each object you sell, you will receive the sum of the prices you set. For example, if
you sold to one buyer using the prices above, you would receive 399 tokens (399 = 26 + 64 + 72 + 0 + 99

+ 25 + 13 + 0 + 83 + 17). If you sold to both buyers, you would receive 798 tokens. If you did not sell any
objects, you would receive 0 tokens.
[next]
Seller comprehension check
Q1. Correct answers are: Full information condition = “All of the prices” and all other conditions = “Only
Price 1”
If the answer is correct, move on to the next question. If the answer is not correct, display the
appropriate error message. Specifically:
Full information = “In this game, buyers will see all of the prices that you set.”
All other conditions = “In this game, buyers will automatically see only price 1. They can see other prices
by searching for those prices.”
SellerQ1. What information will buyers automatically see from you?
___ All of the prices
___ Only Price 1
___ I don’t know
[next]
Q2. Correct answers are: Comprehension = “I will lose 50 tokens + I will not sell them the object” and all
other conditions = “Nothing.”
If the answer is correct, move on to the next question. If the answer is not correct, display the
appropriate error message. Specifically:
Comprehension = “In this game, buyers must report at least 5 prices correctly. Otherwise, you will lose
50 tokens and will not make the sale.”
All other conditions = “In this game, buyers’ reports have no effect on making a sale or your payoffs.”
SellerQ2. What happens if a buyer who tries to purchase my object reports only 3 out of 10 prices
correctly? (Mark all)
___ Nothing
___ I will lose 50 tokens
___ I will not sell them the object
[next]
Only display this question before round 1 – do not display before rounds 11, 21, or 31.
Q3. Correct answer is “0 tokens”
If the answer is correct, move on to the next question. If the answer is not correct, display “Remember
that you only receive tokens if you sell an object. If you do not sell to any buyers, you will receive 0
tokens.”
SellerQ3. How many tokens will you receive if you don’t sell an object?

___ 0 tokens
___ 500 tokens
___ It depends on the prices I set
[next]
Only display this question before round 1 – do not display before rounds 11, 21, or 31.
Q4. Correct answer is “It depends on the prices I set”
If the answer is correct, move on to the next question. If the answer is not correct, display “Remember
that you will receive tokens according to the prices you set. In the case of a sale, the higher the price,
the more tokens you will receive.”
SellerQ4. How many tokens will you receive if you sell an object?
___ 0 tokens
___ 500 tokens
___ It depends on the prices I set
[next]
Buyer instructions
Contains some condition-specific text. This text should be displayed for each participant as appropriate.
Now imagine that you are playing as a buyer. Each round, you will be buying an object from one of the
two sellers in your group. Both sellers are selling an identical object that is valued at 500 tokens.
However, they may have chosen different prices for that object.
Full information condition: You will see all of the prices from each seller. When you buy the object, you
will be asked to report the prices from the seller whose object you bought. If you correctly report at
least 5 of these prices, you will receive 10 tokens.
Price-1 condition: To start, you will see only Price 1 from each seller. In order to see the other prices,
you will have to pay a fee of 10 tokens per price. When you buy the object, you will be asked to report
the prices from the seller whose object you bought. If you correctly report at least 5 of these prices, you
will receive 10 tokens.
Comprehension condition: To start, you will see only Price 1 from each seller. In order to see the other
prices, you will have to pay a fee of 10 tokens per price. When you buy the object, you will be asked to
report the prices from the seller whose object you bought. There is a 50% chance that this report
affects your purchase.
Specifically, in 50% of the cases, if you correctly report fewer than 5 of the prices, you will lose the
object and receive 0 tokens. If you report at least 5 prices, the purchase will occur as usual. In the other
50% of cases, if you do not correctly report at least 5 prices, there will be no effect on the purchase.
Performance condition: To start, you will see only Price 1 from each seller. In order to see the other
prices, you will have to pay a fee of 10 tokens per price. When you buy the object, you will be asked to

report the prices from the seller whose object you bought. If you correctly report at least 5 of these
prices, you will receive 10 tokens.
All conditions: For each object you buy, you will receive 500 tokens minus the total of the prices. For
example, if the prices were set at 350, you would receive 150 tokens.
[next]
All except full information: An example is shown below.
Imagine that the following prices
were chosen…
Seller 1
Seller 2
Price 1
26
15
Price 2
64
35
Price 3
72
86
Price 4
0
45
Price 5
99
92
Price 6
25
74
Price 7
13
22
Price 8
0
10
Price 9
83
0
Price 10
17
84

Price 1
Price 2
Price 3
Price 4
Price 5
Price 6
Price 7
Price 8
Price 9
Price 10

Then you would see…
Seller 1
Seller 2
26
15
[reveal next price for 10 tokens]

___ I want to purchase the object from Seller 1.
___ I want to purchase the object from Seller 2.
[next]
Please report the prices that you saw from [Seller 1 / Seller 2].
Price 1
Price 2
Price 3
Price 4
Price 5
Price 6
Price 7
Price 8
Price 9
Price 10

[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]

[next]
If comprehension condition AND at least 5 of the prices did not exactly match: The prices you reported
do not match the seller’s prices. You do not receive the object this round. You receive 0 tokens.

Otherwise: You purchased the object for a total of [#] tokens [and passed the quiz for an additional 10
tokens]. You receive [#] tokens this round.
[next]
Practice Rounds
Assign participants such that everyone plays three rounds as a buyer and three rounds as a seller. Keep
them in the same role continuously for these rounds.
First we will do a few practice rounds. You will not gain or lose any tokens this round based on your
decisions.
[Show same screen as the Game Rounds below].
Game Rounds
Both buyers and sellers. Repeat for 10 rounds.
Throughout, there will be text that indicates that you are waiting for the other players to finish making
their choices. These text screens should disappear automatically.
Waiting for other players to begin the game…
Seller game screen
[0] represents an open text box that is pre-populated with 0.
This round, you are a seller. Please enter your prices below.
Price 1
Price 2
Price 3
Price 4
Price 5
Price 6
Price 7
Price 8
Price 9
Price 10

[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]
[0]

[next]
Please wait for buyers to make their decisions…
Buyer game screen
If more than 45 seconds passes, display text saying, “Please make your decision in the next 30 seconds.”
This round, you are a buyer. Please wait for sellers to enter their prices…

Price 1
Price 2
Price 3
Price 4
Price 5
Price 6
Price 7
Price 8
Price 9
Price 10

Seller 1
Seller 2
[show values as appropriate]
[reveal next price for 10 tokens]

___ I want to purchase the object from Seller 1.
___ I want to purchase the object from Seller 2.
[next]
Please report the prices that you saw from [Seller 1 / Seller 2].
Price 1
Price 2
Price 3
Price 4
Price 5
Price 6
Price 7
Price 8
Price 9
Price 10

[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]
[enter price]

[next]
Seller summary screen
Personal summary:
# of buyers who tried to purchase from you: [0 / 1 / 2]
# of prices reported correctly from your buyers:
[Not applicable – No buyers]
Buyer 1: [X] of 10 – [Successful purchase]
Buyer 2: [X] of 10 – [Unsuccessful purchase]
Total profits this round: [#] tokens
[next]
Buyer summary screen

Personal summary:
# of prices you reported correctly: [X] of 10 – [Successful purchase]
Total profits this round: [#] tokens
[next]
All summary screen
Round summary:
This round, sellers used the following prices:

Price 1
Price 2
Price 3
Price 4
Price 5
Price 6
Price 7
Price 8
Price 9
Price 10
Total price:

Seller 1
26
64
72
0
99
25
13
0
83
17
399

Seller 2
15
35
86
45
92
74
22
10
0
84
463

[#] buyers purchased from Seller 1.
[#] buyers purchased from Seller 2.
Price reports:
Buyer 1: [X] of 10 – [Successful purchase]
Buyer 2: [X] of 10 – [Unsuccessful purchase]
[next]
At the end of 10 rounds, re-randomize players into groups. Return to the seller and buyer instructions
for the next disclosure condition.

Survey
Thank you, the games are now complete. Before we finish, we have a few questions to help us better
understand how you make decisions and tradeoffs.
RiskQ1
If participant choses “50% chance…”, administer RiskQ2
If they chose “100%...”, administer RiskQ3
This question measures risk preferences.
RiskQ1. Hypothetically, which would you prefer: a 50% chance of getting $35, or a 100% chance of
getting $15?
___ A 50% chance of $35
___ A 100% chance of $15
[next]
RiskQ2 display is contingent on RiskQ1 = “50%...”
For any response, continue to NumeracyQ1
This question measures risk preferences.
RiskQ2. Hypothetically, which would you prefer: a 50% chance of getting $35, or a 100% chance of
getting $17.50?
___ A 50% chance of $35
___ A 100% chance of $17.50
[next]
RiskQ3 display is contingent on RiskQ1 response = “100%...”
For any response, continue to NumeracyQ1
This question measures risk preferences.
RiskQ3. Hypothetically, which would you prefer: a 50% chance of getting $35, or a 100% chance of
getting $10?
___ A 50% chance of $35
___ A 100% chance of $10
[next]
RiskQ4. How willing are you to take risks in your life, in general?
o
Not at all willing
to take risks

o

o

o

o
Very willing to
take risks

[next]

Subjective numeracy scale, as published in Fagerlin et al (2007).
For these questions, display a 6-point scale with anchors 1 = “not at all good” to 6 = “extremely good.”
NumeracyQ1. How good are you at working with fractions?
[scale]

NumeracyQ2. How good are you at working with percentages?
[scale]
NumeracyQ3. How good are you at calculating a 15% tip?
[scale]
[next]

Experience level
CoursesQ1. Please indicate if you have taken college-level coursework in any of the following fields.
(Mark all)
___ Microeconomics or game theory
___ Marketing or business
___ Law
___ None of the above
ExperienceQ1. Approximately how many in-person, laboratory experiments have you completed
before today? If you’re not sure, please make your best guess.
___ 0
___ 1 to 3
___ 4 to 6
___ 7 to 9
___ 10 or more
[next]
GenderQ1. What is your gender?
___ Male
___ Female
[next]
Thank you for participating in today’s experiment. Please wait at your desk until your participant
number is called. You will then receive your payment from the experimenter.
Your number is: [participant number]