1

Consumer Choice, Utility, and Revealed Preference

2

Agenda Consumer Utility Consumer Choice Revealed Preference

3

Consumer Theory There are two important pillars that

consumer theory rests upon: The utility function The budget constraint

4

Utility Function A utility function is a function/process that

maps a bundle of goods into satisfaction where the satisfaction can be ranked for each bundle of goods. In essence, it allows us to rank the desirability of

consuming different bundles of goods. A bundle of goods is a particular set of goods

you choose to consume. This is also referred to as a consumption bundle.

5

Mathematically Representing Utility In Economics, we tend to define the utility

function as the following: U = u(x1, x2, …, xn) where

U is the level of satisfaction you receive from consuming the bundle of goods consisting of x1, x2, …, xn

u( ) is the function that maps the bundle of goods into satisfaction. (Think of this as the mathematical representation of your brain.)

xi for i = 1, 2, …, n is the quantity of a particular good consumed from a bundle of goods, e.g., x1 may be 5 pancakes.

6

Example of Representing Utility in Mathematical Terms Suppose you have a utility function that is

based on three goods pizza, soda, and buffalo wings. We can represent your utility function as U = u(x1, x2, x3). x1 = amount of pizza consumed x2 = amount of soda consumed xn = x3 = amount of buffalo wings

consumed

7

Example of Representing Utility in Mathematical Terms Suppose you have a choice of

consuming one of two different bundles of goods. Bundle 1 has a large pepperoni pizza, a

two liter bottle of coke, and 20 spicy buffalo wings.

Bundle 2 has a large pepperoni pizza, a two liter bottle of coke, and 40 spicy buffalo wings.

8

Example of Representing Utility Cont. For bundle 1

x1 = a large pepperoni pizza x2 = a two liter bottle of coke xn = x3 = 20 spicy buffalo wings

For bundle 2 x1 = a large pepperoni pizza x2 = a two liter bottle of coke xn = x3 = 40 spicy buffalo wings

9

Example of Representing Utility Cont. We can further write the following:

U1 = u(a large pepperoni pizza, a two liter bottle of coke, 20 spicy buffalo wings)

U2 = u(a large pepperoni pizza, a two liter bottle of coke, 40 spicy buffalo wings) Remember u( ) is just a function that transforms bundles

of goods into satisfaction. (Think of it as your brain.) What might be inclined to say that U2 > U1.

Why?

10

Some Notes on Utility It was once thought that utility could be

broken down into units called utils. Total utility is defined as the utility you

receive from consuming a particular bundle of goods.

For analytical tractability, total utility can be given a number value. In our previous example we could say that U1 =

100 and U2 = 200.

11

Some Notes on Utility Cont. There are two views of utility:

Cardinal Utility Cardinal utility is the belief that utility can be measured

and compared on a unit by unit basis. E.g., A utility measure of 200 is twice as big as a utility

measure of 100.

Ordinal Utility Ordinal utility is where you rank bundles of goods, but

cannot say how much greater one bundle is to another. Ranking is the only thing that matters when dealing with

ordinal utility.

12

Utility from Consuming One Good When examining utility, we sometimes would

like to examine how one good affects our utility.

This can be looked at by assuming that we have only one good in our utility function, i.e., U=u(x).

A more realistic outlook is that we examine only one good in our utility function, holding all the other goods constant.

13

Utility from Consuming One Good Cont. Mathematically, we can represent

holding things constant in the following two manners for one good: U = u(x1; x2, x3, …, xn) or U = u(x1 | x2 , x3, …, xn)

Where it is understood using this notation that goods x2 through xn are held at some constant level.

14

Marginal Utility (MU) Marginal utility is defined as the change

in total utility divided by a change in the consumption of a particular good.

In mathematical terms, marginal utility of good i is defined as du(x)/dx, which is just the first derivative of the utility function with respect to good x.

15

Calculating Marginal Utility Suppose Dr. Hurley only like to consume

chocolate. Also, assume that Dr. Hurley’s total utility

function can be represented as U(x) = 10x-x2 This implies that MU(x)=du/dx= 10 - 2x

Assuming that chocolates were free, how many chocolates would Dr. Hurley consume?

16

Total Utility and marginal Utility of U(x)=10x-x2

Utility and Marginal Utility

-15-10-505

1015202530

0 2 4 6 8 10 12

Number of Goods

Utils U(x)

MU(x)

Chart2

11

22

33

44

55

66

77

88

99

1010

1111

U(x)

MU(x)

Number of Goods

Utils

Utility and Marginal Utility

9

8

16

6

21

4

24

2

25

0

24

-2

21

-4

16

-6

9

-8

0

-10

-11

-12

Sheet1

1133.16624790361.5075567229

1234.64101615141.443375673

1336.05551275461.3867504906

1437.41657386771.3363062096

1538.72983346211.2909944487

16401.25

1741.23105625621.2126781252

1842.42640687121.178511302

1943.58898943541.1470786694

2044.721359551.1180339887

2145.82575694961.0910894512

2246.90415759821.0660035818

2347.95831523311.0425720703

xU(x)MU(x)

198

2166

3214

4242

5250

624-2

721-4

816-6

99-8

100-10

11-11-12

12-24-14

Sheet1

U(x)

MU(x)

Number of Goods

Utils

Utility and Marginal Utility

Sheet2

Sheet3

17

Deriving Total and Marginal Utility from a Function Suppose you could represent your

utility function for drinking Mountain Dew as the following: U = u(x) Where u(x) = 10x½

What is Total Utility and Marginal Utility? Graph both.

18

Total Utility and marginal Utility of U(x)=10x½

Utility and Marginal Utility

0

10

20

30

40

50

60

0 5 10 15 20 25

Number of Goods

Utils U(x)

MU(x)

Chart1

11

22

33

44

55

66

77

88

99

1010

1111

1212

1313

1414

1515

1616

1717

1818

1919

2020

2121

2222

2323

U(x)

MU(x)

Number of Goods

Utils

Utility and Marginal Utility

10

5

14.1421356237

3.5355339059

17.3205080757

2.8867513459

20

2.5

22.360679775

2.2360679775

24.4948974278

2.0412414523

26.4575131106

1.889822365

28.2842712475

1.767766953

30

1.6666666667

31.6227766017

1.5811388301

33.1662479036

1.5075567229

34.6410161514

1.443375673

36.0555127546

1.3867504906

37.4165738677

1.3363062096

38.7298334621

1.2909944487

40

1.25

41.2310562562

1.2126781252

42.4264068712

1.178511302

43.5889894354

1.1470786694

44.72135955

1.1180339887

45.8257569496

1.0910894512

46.9041575982

1.0660035818

47.9583152331

1.0425720703

Sheet1

xU(x)MU(x)

1105

214.14213562373.5355339059

317.32050807572.8867513459

4202.5

522.3606797752.2360679775

624.49489742782.0412414523

726.45751311061.889822365

828.28427124751.767766953

9301.6666666667

1031.62277660171.5811388301

1133.16624790361.5075567229

1234.64101615141.443375673

1336.05551275461.3867504906

1437.41657386771.3363062096

1538.72983346211.2909944487

16401.25

1741.23105625621.2126781252

1842.42640687121.178511302

1943.58898943541.1470786694

2044.721359551.1180339887

2145.82575694961.0910894512

2246.90415759821.0660035818

2347.95831523311.0425720703

Sheet1

U(x)

MU(x)

Number of Goods

Utils

Utility and Marginal Utility

Sheet2

Sheet3

19

Lessons Learned from Example As long as marginal utility is positive,

total utility will increase. Total and marginal utility can be

negative. When marginal utility is zero, total

utility is maximized. This graph demonstrated the Law of

Diminishing Marginal Utility.

20

Law of Diminishing Marginal Utility This law states that as you consume

more units or a particular good during a set time, at some point your marginal utility will decrease as your consumption increases. Note: Beware of laws in economics, they

are not like physical laws. This is equivalent to saying that the

derivative of MU is negative.

21

Utility from Consuming Two Goods When examining utility, we sometimes would

like to examine how two good interrelate in our utility function.

This can be looked at by assuming that we have only one good in our utility function, i.e., U=u(x1, x2).

A more realistic outlook is that we examine two goods in our utility function, holding all the other goods constant.

22

Utility from Consuming Two Goods Cont. Mathematically, we can represent

holding things constant in the following two manners for two goods: U = u(x1, x2; x3, …, xn) or U = u(x1, x2 | x3, …, xn)

Where it is understood using this notation that goods x3 through xn are held at some constant level.

23

Isoutility To this point we have examined utility

based on the consumption of one good. Realistically, our consumption bundles

have more than one item in them. When more than one good exist in our

consumption bundle and utility function, it becomes important for us to examine the idea of isoutility.

24

Isoutility Isoutility is a concept where differing

bundles of goods provide the same level of utility.

From the idea of isoutility comes the graphical representation--an indifference curve. An indifference curve is the graphical

representation of differing bundles of goods giving the same level of utility.

25

Demonstrating an Indifference Curve Suppose we have two goods in our utility

function—corn dogs (C) and sandwiches (S). To obtain an indifference curve, we want to

hold our utility constant and look at the different bundles of consumption of each good that provide the same utility.

Suppose we know that we have the following utility function U = u(C, S) = C*S

26

Demonstrating an Indifference Curve Cont. To derive the indifference curve, we

want to keep U constant. In this case we will assume U = 10

What bundles of goods will give us this level of satisfaction?

What if U = 20?

27

Increasing Utility

Chart1

22

33

44

55

66

77

88

99

1010

Utility = 10

Utility = 20

Number of Corn Dogs

Number of Sandwiches

5

10

3.33

6.66

2.5

5

2

4

1.7

3.4

1.4

2.8

1.25

2.5

1.1

2.2

1

2

Sheet1

2345678910

Utility = 1053.332.521.71.41.251.11

Utility = 20106.66543.42.82.52.22

28

Observations The two utility curves do not cross each

other. Why?

Utility is increasing as you move away from the x and y-axis.

As you consume more of one of the goods, the consumption of the other good has relatively more value to you.

Any other observations?

29

Notes on Indifference Curves Indifference curves usually are not

considered thick. Why.

Indifference curves tend to not be upward sloping. Why?

30

Example of a Thick Indifference Curve

x2

x1

106

7

••

31

Marginal Rate of Substitution (MRS) When examining indifference curves, it is

important to look at the tradeoff between consuming one good versus the other. This is called examining the marginal rate of

substitution. The marginal rate of substitution can be defined

as the rate at which the consumer is willing to trade one good for another.

Why is this important to examine?

32

Mathematical Representation of MRS MRS of good xi for good xj

=dxj /dxi

33

Example of Calculating MRS From our previous example, we had the

following utility function: U = u(C,S) = C*S Setting U = 10, what is the MRS of

corndogs for sandwiches at 2 sandwiches and 5 corndogs, i.e., dS /dC?

34

Implication of Indifference Curve Due to indifference curves, we have the

following relationship for differing bundles of goods: (dxj / dxi) = -(MUxi /MUxj)

Which implies (dxj * MUxj) = -(MUxi * dxi) Which implies (dxj * MUxj) + (MUxi * dxi) = 0

This implies that the loss in utility from consuming less of good xj is just matched by the gain in utility from consuming more of xi.

35

Budget Constraint This constraint defines the feasible set of

consumption bundles you are able to consume.

It is dependent on three things: The prices of the goods in your consumption

bundle. The quantity of each good in your consumption

bundle. Your income.

36

Budget Constraint Cont. The budget constraint for two goods is

defined as the following: m = p1*x1 + p2 * x2

m = income p1 = the price of x1 p2 = the price of x2 x1 = the quantity consumed of good 1 x2 = the quantity consumed of good 2

37

Budget Constraint Cont. Assume that we want to graph are budget

constraint and choose x2 as the good we want to put on the y-axis.

m = p1*x1 + p2 * x2 can be rewritten as: x2 = (m/p2) – (p1/p2) * x1

With this equation we can draw the budget constraint on a graph.

Note that the slope of the budget constraint is equal to the negative of the price of good 1 divided by the price of good 2.

38

Budget Constraint Example Suppose you have $100 to spend on

chips and soda. You know that the price of chips is $1 and the price of soda is $2. Draw your budget curve assuming that chips will be on the y-axis.

39

Budget Constraint Example Budget equation:

$100 = $2 *x1 + $1 *x2 ⇒x2 = 100 – 2*x1

40

Budget Constraint Cont.# of Chips

# of Sodas50

100x2 = 100 – 2*x1

Feasible Set

Not Feasible

41

Consumer Equilibrium Consumer equilibrium is comprised of

two concepts: The utility function The budget constraint

Consumer equilibrium can be defined as a consumption bundle that is feasible given a particular budget constraint and maximizes total utility.

42

Consumer Equilibrium Cont. If there was no budget constraint, a person

would consume each good to the point where marginal utility of consumption for each good is zero. Why?

Given a budget constraint, the consumer maximizes total utility by consuming a bundle that is feasible. A feasible bundle is one that lies either on or

inside the budget constraint.

43

Consumer Equilibrium Cont. In graphical terms, consumer

equilibrium is defined as the point where the highest utility function touches the budget constraint.

44

General Utility Maximization Model

∑=

≥n

iii

nxxx

xpm

xxxuMaxn

1

21,...,,

:subject to

),...,,(21

45

Utility Maximization with One Good

0

0

)()(),(

:subject to

)(

≥−=∂Γ∂

=∂∂

=⇒

=−∂∂

=∂Γ∂

−+=Γ≥

pxm

pMU

pxu

pxu

x

pxmxuxpxm

xuMaxx

λ

λ

λ

λλ

46

Utility Maximization with One Good Cont. When your maximizing utility with respect to

one good, there are two scenarios that occur: You have more money than you need to maximize

your utility, so you do not spend all of your money. This implies λ = 0 and m > px.

You run out of money before you reach the highest amount of utility you can achieve. This implies λ > 0 and m = px.

To solve this problem, you need to check which scenario provides the highest utility while satisfying the constraint.

47

Consumer Equilibrium Example for One Good Suppose we have the following utility

function: U = u(x) = x½

Suppose the price of good x is $5 and the income available is $125.

What is the optimal utility and how much money do you have left?

48

Consumer Equilibrium Example for One Good Cont. Now, Suppose we have the following

utility function: U = u(x) = 10x-x2

Suppose the price of good x is $5 and the income available is $125.

What is the optimal utility and how much money do you have left?

49

Utility Maximization for Two Goods Suppose we have the following utility

function: U = u(x1,x2)

Where x1 is equal to the number of good 1 Where x2 is equal to the number of good 2

Suppose we have the following budget constraint: M = p1*x1 + p2*x2

Where p1 is equal to the price of good 1 Where p1 is equal to the price of good 2

50

Utility Maximization with Two Goods Cont.

2211

21,

:subject to

),(21

xpxpm

xxuMaxxx

+=

51

Utility Maximization with Two Goods Cont.

2

2

222

1

1

111

22112121

2

2

1

1

0

0

)(),(),,(

pMU

pMU

pxu

x

pMU

pMU

pxu

x

xpxpmxxuxx

x

x

x

x

=⇒

=⇒

=−∂∂

=∂Γ∂

=⇒

=⇒

=−∂∂

=∂Γ∂

−−+=Γ

λ

λ

λ

λ

λ

λ

λλ

52

Utility Maximization with Two Goods Cont. (pj/pi) = (MUj/MUi) can be rewritten as:

(MUj / pj) = (MUi / pi) This says that you are normalizing the change in

utility by the price of the good and then equating it to the normalized marginal utility of the other good.

Another way to look at this is to say that the marginal utility derived from the last dollar spent for each good is equal.

What happens if one side is greater than the other?

53

Utility Maximization with Two Goods Cont. Intuitively what we have done in the

graph is equate the tradeoff from prices to the tradeoff in utility. I.e., (pj/pi) = (MUj/MUi)

Where pj is the price of good j and pi is the price of good i

Where MUj is the marginal utility of consuming good j and MUi is the marginal utility of consuming good i

54

Consumer Equilibrium Example for Two Goods Suppose we have the following utility

function: U = u(x1,x2)= x1 * x2

Where x1 is equal to the number of hotdogs consumed Where x2 is equal to the number of sodas consumed

Suppose we have the following budget constraint: M = p1*x1 + p2*x2

Where p1 is equal to the price of hotdogs consumed Where p2 is equal to the price of sodas consumed

55

Consumer Equilibrium Example Cont. Now consider that you have a price of

hotdogs (x1) equal to $2 and a price of soda (x2) is a $1.

Also suppose that our income is $16. Solution will be done in class.

56

Revealed Preference Revealed Preference is about making

observations of people’s consumption values to understand their preferences.

Suppose there are two bundles of goods (x1,x2) and (y1,y2).

An assumption of revealed preference is that the consumer always chooses the optimal bundle given her budget constraint.

If prices are p1 and p2, then we can represent these two bundles on a graph.

57

Revealed Preference Cont.

Good 1

Good 2

●(x1,x2)

●(y1,y2)

58

Revealed Preference Cont. Define m = p1x1+p2x2 We know from the above graph that

p1y1+p2y2 ≤ m. Why?

We also know that p1x1+p2x2 ≥ p1y1+p2y2. Why?

59

The Principle of Revealed Preference If (x1,x2) is the chosen bundle of goods at

prices (p1,p2), and if (y1,y2) is affordable, i.e., p1x1+p2x2 ≥ p1y1+p2y2, then the bundle (x1,x2) is preferred to (y1,y2), i.e., (x1,x2) > (y1,y2). In this case, we say (x1,x2) is directly revealed

preferred to (y1,y2). Note: If the consumer is choosing the optimal

bundle, then revealed preference will imply preference.

60

Indirectly Revealed Prefer Suppose that there is a set of prices

(q1,q2) where (y1,y2) is directly revealed prefer to (z1,z2), i.e., the consumer chooses (y1,y2) while q1y1+q2y2 ≥ q1z1+q2z2.

Also assume that at price (p1,p2), (x1,x2) is directly revealed preferred to (y1,y2).

61

Indirectly Revealed Prefer Then we know that (x1,x2) is indirectly

revealed preferred to (z1,z2), even if (z1,z2) was not affordable at (p1,p2).

The graph on the next slide gives a visual representation of this.

62

Indirectly Revealed Prefer Cont.

Good 1

Good 2

●(x1,x2)

●(y1,y2)●(z1,z2)

63

Weak Axiom of Revealed Preference (WARP) If the bundle (x1,x2) is directly revealed

preferred to bundle (y1,y2) and the two bundles are not the same, then it is not possible that (y1,y2) is directly revealed preferred to (x1,x2) when (x1,x2) is affordable.

64

Graphical Example of WARP Being Violated

Good 1

Good 2

●(x1,x2)

●(y1,y2)

65

Example of WARP Being ViolatedObservation p1 p2 x1 x2

1 2 6 3 5

2 3 3 4 4

66

Example of WARP Being Violated Cont.

Observations (x1,x2)

1(3,5)

2 (5,4)

Prices (p1,p2)

1 (2,6)

36 34*

2 (3,3)

24* 27

67

Strong Axiom of Revealed Preference If the bundle (x1,x2) is directly or

indirectly revealed preferred to bundle (y1,y2) and the two bundles are not the same, then it is not possible that (y1,y2) is directly or indirectly revealed preferred to (x1,x2).

68

Example of SARP Being ViolatedObservation p1 p2 x1 x2

1 1 1 9 4

2 2 1 3 7

3 1 2 1 10

69

Example of SARP Being Violated Cont.

Observations (x1,x2)

(9,4) (3,7) (1,10)

Prices (p1,p2)

(1,1) 13 10 11

(2,1) 23 13 12(1,2) 17 17 21

Consumer Choice, Utility, and Revealed PreferenceAgendaConsumer TheoryUtility FunctionMathematically Representing UtilityExample of Representing Utility in Mathematical TermsExample of Representing Utility in Mathematical TermsExample of Representing Utility Cont.Example of Representing Utility Cont.Some Notes on UtilitySome Notes on Utility Cont.Utility from Consuming One GoodUtility from Consuming One Good Cont.Marginal Utility (MU)Calculating Marginal UtilityTotal Utility and marginal Utility of U(x)=10x-x2Deriving Total and Marginal Utility from a FunctionTotal Utility and marginal Utility of U(x)=10x½Lessons Learned from ExampleLaw of Diminishing Marginal UtilityUtility from Consuming Two GoodsUtility from Consuming Two Goods Cont.IsoutilityIsoutilityDemonstrating an Indifference CurveDemonstrating an Indifference Curve Cont.Slide Number 27ObservationsNotes on Indifference CurvesExample of a Thick Indifference CurveMarginal Rate of Substitution (MRS)Mathematical Representation of MRSExample of Calculating MRSImplication of Indifference CurveBudget ConstraintBudget Constraint Cont.Budget Constraint Cont.Budget Constraint ExampleBudget Constraint ExampleBudget Constraint Cont.Consumer EquilibriumConsumer Equilibrium Cont.Consumer Equilibrium Cont.General Utility Maximization ModelUtility Maximization with One GoodUtility Maximization with One Good Cont.Consumer Equilibrium Example for One GoodConsumer Equilibrium Example for One Good Cont.Utility Maximization for Two GoodsUtility Maximization with Two Goods Cont.Utility Maximization with Two Goods Cont.Utility Maximization with Two Goods Cont.Utility Maximization with Two Goods Cont.Consumer Equilibrium Example for Two GoodsConsumer Equilibrium Example Cont.Revealed PreferenceRevealed Preference Cont.Revealed Preference Cont.The Principle of Revealed PreferenceIndirectly Revealed PreferIndirectly Revealed PreferIndirectly Revealed Prefer Cont.Weak Axiom of Revealed Preference (WARP)Graphical Example of WARP Being ViolatedExample of WARP Being ViolatedExample of WARP Being Violated Cont.Strong Axiom of Revealed PreferenceExample of SARP Being ViolatedExample of SARP Being Violated Cont.