Economic Principles: How the Market Works

Chapter 4: The Choice of the Consumer

• Needs (do we have true and false needs? The role of advertising)
• Economics as a �value-free� science
• Preferences: perception, utility function, indifference curves
• Constraints: budget
• Choice
• Introducing time in the picture

  •  time modifies the budget constraint: savings and loans
  •  time as a mean to produce satisfaction (Lancaster)
  •  time creates uncertainty (weakens the link good/satisfaction)
  

To be remembered:

•  subjectivity of value

•  resources are not given

•  the roles of advertising

•  degrees of substitution

•  the law of decreasing marginal utility

•  time horizon

•  one good satisfies many needs

•  to choose is to give up something for something else

Bibliography:

. Menger on the definition of goods

. J.B. Say on value, wealth and utility

About the next six chapters

This chapter and the five chapters that follow present the key concepts and theories of what is usually called microeconomics . Microeconomics follows the research programme defined by Lionel Robbins (see chapter 2): we try to understand social phenomena as outcomes of many rational interactions. Hence we study what is rational to do for an economic agent (a producer or a consumer) and we see what can result from that, what could be the social outcome. As you will see, the key concept used in those chapters to guess what the outcome will be is the concept of equilibrium . The economy is in a state of equilibrium when all the forces exercised by each individual action �equilibrate�.

Introduce here a supply and demand graph to explain that the next chapters will be explaining the construction of that graph. Where does the demand come from? Why is the slope of the supply curve positive?

This emphasis put on equilibrium is, according to a growing part of the economic profession, including myself, at best a mixed blessing. Actually we can say that it forbids us to understand the dynamics of progress. For that reason we will be criticizing it as we introduce it and we will attempt to offer an alternative analysis in later chapters. If we have chosen to present the equilibrium even though we believe it is limited it is first, because it might be useful, and second, because it remains the common language of modern economists, the reference point from which you argue.

Why it is important to start with a look at consumers' behaviour: the notion of good

Adam Smith, who has been praised in the previous chapters for his sharpness, started his famous book, as you remember, with a story that illustrates how, through the division of labour, productivity can be greatly improved, and the Nations can get wealthier. So why don't we follow the master and start with a study of production processes? Why should we start by the consumer? The answer is simple: we produce to satisfy need and the consumer is the one who carries those needs.

Some time the good disappear as I consume it. This is what happens with the glass of water. In that case I satisfy my need by �destroying� the good; we say that those goods are �perishable�. But some other times, as you know, the good that helps me to satisfy my needs is not immediately destroyed. It just gets older. Think about the fridge that gives you fresh water and helps you to preserve the food, think about your car, or your bike. We call those goods, investment goods.

The fridge is an interesting example. In fact, what you want, what you need is fresh water. For this you need water, of course, but also a device to keep it fresh. This devise can be called a good because it is part of that process leading to the satisfaction of a need. So the fridge is a good. But in the same way, to make a fridge requires many things. Those things can also be called �goods� and for the same reason. Menger called them goods of �higher order� to signify that those goods, for instance the gas that is used in the refreshing process, are not directly consumed by the final user, but are used to produce some other goods that will be consumed.

Hence, all economic activities are directed toward the production of some good, and something deserves to be called �a good�, if and only if, soon or later, it allows satisfying some needs. This is why it is important to understand how the consumer makes his or her selection among the goods presented to him. It is the consumer, as we will see, who through his or her choices gives value to the production and to any economic activity. This is why we start with the study of the consumer's choice.

But does the consumer really �choose�? Is he sovereign in his choice?

Hold on! Will say some. Could not we argue that the causality goes exactly in the opposite direction? Could not we argue that it is not the consumer who tells the producer what to produce but the reverse: the producer is imposing some �goods� to the consumer. So the consumer is not really choosing. In that sense, he is not sovereign. He buys what producers suggest he should buy.

This objection to consumer sovereignty is often met. People would insist for instance that advertising makes you buy things you don't really need, or that, most of the time, you don't know exactly what you buy. Such a view has been supported among economists by John Kenneth Galbraith, a famous professor from Harvard.

But there exist counter objections to that argument. First, as I will explain in a minute, that argument is based on some confusion. Secondly, if that argument was true then we would be in serious trouble.

The confusion lies at two levels. First, those who put forward the argument according to which the consumer does not really choose confuse needs with the way we choose to satisfy needs. For instance, when a kid buys a play station, it is not because a new need was �created� by the producer of the play station and forced upon the kid. The producer of the play station has created a new way to satisfy a well known need , here, the need to relax, to have some good time. When the kid buys the play station it is because she thinks she has found a better way to satisfy an old need. Would you say that every new song creates a new need: the need to have a recording of that song? This surely sounds stupid and whoever tells you that we already have millions of songs to listen to and that one does not need new ones, you will disregard him as a stupid, backwarded, and unimaginative person. And boring on top of that!

The second level of confusion is that one does not understand that the same good, let us say, a car, often serve to satisfy more than one need. Indeed a car is not useful just because it is a mean of transportation. It is also useful to satisfy the need to relax (to go on vacation, for instance; or more simply to satisfy an old dream), or the need for identification. The image others have of me matters to me. I need to be part of a group, I need to relate to people; and the car is sometimes a way to that end. Producers know that, and of course they will use it, including in their advertising campaign.

Now, let us assume for a moment that my counterarguments are not valid; let us assume that consumers are not sovereign, that they don't choose, and that producers can �force them� to buy things they don't need. Then we have two solutions. One is to say: well that's life! Too bad for them. But the other solution, the one usually suggested by those who raised the problem of �false needs� in the first place, is that something should be done. May be advertising should be regulated (in some countries, comparative advertising is forbidden, or advertising of drinks with alcohol, or advertising tobacco); or high taxes imposed on some goods which, according to them, are obviously unnecessary, unessential, and may be even harmful to the consumer. But to implement those remedies, one must first be able to identify the goods that must be prohibited or penalized. And this raises the terrible question: who will do that? It is one thing to say: �I, Pierre, believe that she does not need another pair of shoes�, and quite another thing to say: �This type of shoes should be heavily taxed because it does not help to satisfy essential needs�. The first judgment is a personal judgment that does not advocate any coercion. The second one is extremely dangerous and may lead to the most oppressive political regimes.

Economists conclude from that debate that tastes are subjective . By that they mean that tastes are personal, related to the individual. It is not the business of the economist to judge whether those tastes are good or bad, to judge whether the consumer �really needs� this or that. The consumer's tastes are not open to discussion as far as the economist is concerned. They have the status of exogenous data for economics. This is why one sometimes says that economics is a value-free science. And when we say that, it does not mean that the economist, as a person, has no value. It does not mean either that economists believe that values are unimportant to explain individual behaviour and beyond that to explain social phenomena. 1 It just means that the study of values is out of the realm of economic analysis. What economics study is the following: given a set of values, given the tastes of consumer X and Y, what are the likely consequences? And in particular, what are the �best� ways to satisfy them?

From needs to preferences

So, we take it from granted, from now on, that the individual has some subjective needs that he/she wishes to satisfy. To satisfy those needs, action is usually necessary (not always, think about the need to breathe, it does not really require an action, at least not in most circumstances). This chapter focus on one kind of action: the purchase of some goods and services, and in that section we wish to analyze the link between, on one hand the needs to satisfy, and on the other hand the preferences for good X over god Y. And the first thing to be noticed is the sense of causality : the consumer, let us call her Mila, would prefer owning good X to owning good Y, because, in her view, good X would satisfy more of her needs than good Y. Hence the causality goes from needs to be satisfied to preferences over goods.

Which goods? This is a simple, but fundamental question. Mila will choose among the goods that she knows about; among the goods that she perceives as being available through exchange. In other words, Mila's realm of choice is delimited by her own perception of what is feasible and what is not. That perception is not necessarily correct. For instance, Mila might ignore that there is a gas station around the corner, and drive an extra ten minutes to fill up her car's tank. Had she known about that gas station around the corner she would have preferred going there instead of driving ten minutes. Below, it will be assumed that Mila is rational, but you should remember that does not mean that she knows everything. Her rationality expresses itself inside the world as she perceives it, not as it really is. In that sense one can say that her rationality is subjective : it applies to the reality as she perceives it.

Giving more substance to the rationality assumption

In chapter 2 we said that the rationality assumption is essential to economic reasoning. We take as granted that individual are rational and that social phenomena can be understood as the outcome of interactions between those rational individuals. It is now time to be more precise about what we mean when we say that Mila is rational.

To start with let us call X the set of goods and services that are perceived by Mila as being available; the choice set as she perceives it. We will denote the goods by small bold letters: x, y, z ,� So X = { x , y , z �}. Mila, who has some needs to satisfy, believes her choice to be limited to that set, and, in the light of the needs she wishes to satisfy, she will express preferences among the elements of that set. If she prefers a quantity x of good x to a quantity y of good y we will write x > y . In such a context, we will say that Mila's preferences are rational if they are transitive . The property of transitivity is very intuitive. It says that whenever x is preferred to y and y is preferred to z then x must be preferred to z . If I prefer apple to bananas and bananas to oranges, and if I am rational, I will prefer apple to oranges. In mathematical form: if x > y and y > z then x > z .

Is it possible that a consumer be irrational in the sense just defined? Yes, but it is very unlikely. Why? First, that consumer will probably die before making his first choice. Just imagine Mila in front of the apple, the orange and the banana. She will spend her life hesitating between the three goods: I prefer apple to bananas and bananas to oranges, but I also prefer oranges to apple!! So what should I pick up? Secondly, note that if Mila has intransitive preferences and if you are smart you can get all the money that Mila has in her wallet, and she will end up bankrupt. How would that go? Assume Mila has an orange in her possession (that assumption is not essential to the demonstration), then you must first buy an apple and a banana (as you will see, those expenses will be largely paid for). You then go to Mila and offer to exchange her orange against your banana, and since Mila prefers Bananas to Oranges, she will even accept to give you some money in order to get her preferred fruit. Let us say she will be ready to trade her orange plus 10 cents against your banana. Now that she has a banana, take your apple and offer Mila to trade that apple against her banana plus 10 cents. Because Mila prefers apples to bananas, she should accept. Now she has the apple and you have the banana, the orange, and twenty cents� If Mila's preferences are transitive, the story must stop here with Mila having paid twenty cents to exchange an orange against her favourite fruit: the apple. But if Mila has intransitive preferences, you can continue the story. Now that Mila is in possession of the apple, you offer her an orange in exchange of that apple and ten cents. And she will accept. Now you have thirty cents and she has the orange. I let you imagine the rest of the story. It will stop when Mila has no money left. For that reason economists say that intransitive preferences lead to money-pump phenomena. We have pumped all the money out of intransitive Mila's pocket!

A mathematical representation of preferences: the utility function

If Mila's preferences are transitive, and more precisely, if she can order all the elements of her choice set according to her preferences (with eventually some elements between which she is indifferent) then it will be possible to represent those preferences with a utility function. A utility function, we will call it U, associates to each possible choice a real number with the following property: if x > y then U ( x ) > U ( y ).

Note that only the order of ranking has some economic meaning. By that we mean that if, let us say, U(x) = 2 and U(y) = 4, we must not conclude that Mila likes y �twice more� than x . The only thing that we can conclude is that y is preferred to x because a higher level of utility is attached to it. To emphasis that only the order given by U has an economic meaning we say that U is an ordinal representation.

This also implies that the theory developed below does not require any measure of the intensity of the satisfaction that the consumer receives when consuming a good or a service. All we need for what follows is to assume that Mila is able to rank levels of satisfaction, not to measure it.

Marginal utility

We can now introduce one of the most essential concepts of modern economic theory: marginal utility. For that let us assume that Mila can consume various quantities x 1 , x 2 , x 3 , x 4 � of the same good x , where x 1 < x 2 < x 3 < x 4 � That good can be money, water, croissants, time, or whatever you like. Let us take money (leva). The first 10 leva that Mila will get will be of great use to her. With those ten leva, if she is rational, she will purchase what helps her to satisfy her most urgent needs. May be to eat, or to drink. Then the next 10 leva she receives will be used to satisfy needs which are most able to increase her level of satisfaction, given that she has already had something to eat and to drink; may be transportation to go to work. We then see that, as a matter of logic , the increase in terms of satisfaction that Mila receives when she gets her first 10 leva is higher than the increased in satisfaction she receives when receiving the next ten leva, which is still higher than the increased of satisfaction received when she gets 10 leva for the third time, and so on. This simply because each time she receives more leva she puts it to its best use, according to her own needs.

What we see in that example is a general principle. We call the law of decreasing marginal utility. Defining marginal utility as the increase in utility that the consumer gets when consuming an extra unit of a good, that law states that the marginal utility derived from the consumption of a good decreases as I consume more of that good . Hence the first leva I get brings more utility than the second one, which in turn brings more utility than the third one, and so on.

Note that if marginal utility is decreasing, total utility is increasing. Mila prefers having 30 leva than having 10 leva! This is illustrated in the little table below (the number have no importance except their order).

A graph can help us to visualize that law of decreasing marginal utility. As you can see on that graph, even if the good is desired by the consumer, it is possible that after consuming a large quantity of it, no use can be found to an extra unit. Still, in most cases that good can be useful because I can exchange it for something I need. But if this cannot be done and if I cannot get rid off that extra unit without cost, then that extra unit decreases my total level of utility. In economics we say that the consumer has reached a threshold of satiety . The marginal utility becomes negative.

Indifference curves, substitutability and complementarities between goods

We will now introduce some equally important concepts. To do that, assume that Mila has to rank by order of preference, not various quantity of the same good, as was the case above, but various �baskets of good�, each made of a given quantity x of good x , and y of good y . To each basket ( x , y ) she will, implicitly, associate a level of utility U( x , y ). And if that function U correctly represents Mila's preferences then the following must hold:

if ( x1 , y1 ) is preferred to ( x2 , y2 ) then U( x1 , y1 ) > U( x2 , y2 )

An indifference curve is defined as the curve that joins all the baskets that procure to the consumer the same level of utility, that is all the baskets (x,y) which are such that U(x,y) = k, where k is the level of utility common to all the baskets on that curve. If Mila, for instance, is indifferent between (a high quality car and a small apartment) or (a low quality car and a large apartment), those two baskets will be on the same indifference curve. The basket made of (one high quality car and one large apartment) will probably be on another indifference curve corresponding to a higher level of utility.

A map of indifference curves. The arrow points in the direction of higher levels of utility

On the above graph, we have chosen to give a specific form to the indifference curves: they are decreasing, concave and they never cross. Will it be always the case? If I choose randomly a consumer and ask him to rank by order of preferences baskets composed of various quantities of two goods, will I always obtain indifference curves that look more or less like those ones? The answer is: Yes and No. Yes, indifference curves cannot intersect; at least not as long as Mila is rational in the sense defined above. No, indifference curves will not always have that shape. To study more precisely the shape of an indifference curve, one mare concept is required, the concept of marginal rate of substitution.

The two goods present in the basket can be related in many ways. Take the tyre and the rim. A tyre is of very little use if you don't have a rim to put in on, and reversely, a rim is of very little use without a tire. The same thing can be said of the software and the hardware, each one is useless if you don't possess the other. When goods have such relationship between them, we say that they are complementary . At the opposite, you may have in the basket goods that, in the eyes of the consumer, fulfil the same needs. Mila for example may like pasta just as much as rice, so that she can substitute pasta to rice without any problem. We will say that such goods are substitutes . If you take two goods randomly, they will usually have some degree of substitution between themselves, but not be perfect substitutes. Take the car and the apartment. It is clear that they are not perfect substitute, but there is nonetheless some degree of substitution between them because they can both satisfy some common need: in particular the need to relax, to enjoy life, to send an image of you to your friends, and so on.

The shape of the indifference curves will depend on the degree of substitution prevailing between the two goods. This is illustrated below.

Complementary goods ------------substitutable goods --------------some degree of substitution

The last concept we need to introduce in order to complete the description of preferences is the concept of the marginal rate of substitution (MRS). Assuming Mila has a basket made of (2 DVDs and 50 Books). That basket gives her a certain level of utility. Now, we can ask Mila the following question: Assume I take 30 books out of your basket so that you are left with only 20 books, how many DVDs must I add to the basket so that you feel just as well as before, when you had 2 DVDs and 50 books? If she answers 3, it means that she is indifferent between the two baskets: (2 DVDs, 50 books) and (5 DVDs, 20 books). Mila has informed us that, when she possesses (2 DVDs and 50 books), her rate of substitution of DVDs for books is 3 to 30, she is ready to give up 30 books for 3 DVDs. That rate of substitution gives us the slope of the indifference curve (and therefore its shape).

The slope in general changes as we move along the indifference curve. If we observe points A and B on the right hand graph, we see that the slope in A is much steeper than the slope in B. Now, if you think about it, this makes perfect sense. The point A represents a basket with a little bit of good x and a lot of good y (for instance, 2 DVDs and 60 books). In such a situation the consumer is ready to give up a lot of books (good y) for some more x (DVDs). The point B, on the other hand, is a basket with a lot of good x and little of good y. In point such as B the rate of substitution is naturally different from the one that prevails in A. In B, the consumer is ready to give up a lot of a good x for some more y.

As we see, the rate of substitution depends on how much of each good I have in the basket. If I have a lot of good x, the marginal utility of x is low; if I add or retrieve one unit of x from the basket this will not substantially change my general level of utility. If I have little of good y in my basket, then one more unit will greatly increase my level of satisfaction. Hence one should not be surprise that the marginal rate of substitution at one point (let say A) is equal to the ratio of the marginal utility of y to the marginal utility of x. In A that ratio is high; in B that ratio is low. We hence have obtained the following definition:

Back to scarcity: the budget constraint

The basket of goods that Mila will choose depends on her preferences, but that is not the only parameter of her choice. You don't choose just what you prefer; you also choose what you can afford. To understand her choice we must therefore introduce another constraint, the budget constraint . This is the clearest expression of scarcity in this description of the consumer's choice. 2

The budget constraint is often very complex. This is due to the fact that a household can usually borrow some money and therefore spend more than the revenue possessed at one point in time. Inversely, the household can decide to spend only part of its present revenue, and save the rest. Borrowing and saving are two mechanisms which allow the consumer to spread through time his consumption plan so that the money that get in at time t does not have to be spend at time t, and the money spend at time t does not have to be equal to the money receive at time t. Those two mechanisms therefore increase the well-being of consumers by increasing the set of possible choices.

But for now we will assume those mechanisms to be absent. So that the consumer, Mila, will have to do with her revenue for the period considered. We will denote that revenue R .

Beside her revenue, Mila must also take into account the prices at which the goods are offered. In real life, the consumer sometimes bargains for the price. Here we assume that Mila has no bargaining power . She takes the prices has given. We say that she is a price-taker , by opposition to a price-maker.

We can now give a mathematical form to that constraint. If the goods offered are x, y, z, � and the corresponding prices are p x , p y , p z , �; then Mila can choose a basket ( x , y , z ,..) only if the following holds:

This inequality just says that Mila's expenses must not exceed her revenue.

If Mila has the choice between only two goods, x and y , then the set of all affordable baskets is easy to graph. It is limited by the straight line with equation: R = p x x + p y y . That line is easy to draw. It crosses the x-axis at R/p x and the y-axis at R/p y . Indeed, if Mila spends all her revenue buying good x , she can buy a quantity R/p x . Similarly, if she spends all her budget on good y, she can get R/p y . Let us note finally that the slope of that line is given by the relative price. It is equal to (� p x /p y ).

Mila can choose any basket on or below that line. The baskets above the line will cost too much.

Mila's choice

We finally have all the elements necessary to understand Mila's choice. She will try to reach the highest possible level of utility given her budget constraint. Graphically, she is �superposing� the budget constraint and her indifference map.

The basket chosen will be the one corresponding to the point A which contains quantity x * of goof x and y * of good y . Mila would like to chose a basket located on a higher indifference curve, but she cannot afford it. Also note that it would be a mistake for her to but a basket such as the basket B. At point B she spends all her revenue (since that point is on the budget constraint), but she has a level of utility lower than the one she reaches in A.

Clearly that optimal choice depends on Mila's preferences: change the indifference curves, you will change the optimal choice. It also depends on her revenue, and on the prices of the two goods (which together give the budget constraint). We will see in the next chapter how Mila's choice is modified when some of those elements change.

Without knowing the various parameters that determine Mila's choice (her preferences, her revenue, and the prices), it is of course impossible to predict what the choice will be. To predict the choice of a consumer is not in the power of the economist. But there is however a property that Mila's choice will verify. A property that holds for any consumption choice. We will end that chapter by the presentation of that fundamental property.

The characteristics of the consumer choice

Two things characterize the optimal choice of the consumer. First, as explained above, the consumer will spend all his revenue. This, as we said before, will be the case whenever there is no possibility to save. The second one can be seen on the graph. 3 We see indeed that the indifference curve on which the optimal choice (point A) is located is tangent to the budget constraint at that point. This means that the slope of the indifference curve at that point is equal to the slope of the budget constraint. Now you remember that the slope of the budget constraint is given by the relative price. It is �p x /p y . We also explained above that the slope of the indifference curve is given by the Marginal rate of substitution at that point. We can therefore conclude that, at the optimum, the rate of marginal substitution equals the relative price.

What does that mean? A lot! As a matter of fact that result is one of the most important in the theory of economic choice. But first let us see through an example what it says. Assume that the ratio of the price of good x to the price of good y is 1 to 1. In other words let us assume that the two goods have the same price, and let us say it is 3 leva. Let us also assume that Mila chooses a basket with the quantity ( x , y ) such that the above equality does not hold. This will be the case for instance if when Mila holds that basket, the marginal utility of x is 2 and the marginal utility of y is 1. Hence we have at that point a MRS equals to 2 and a relative price equals to 1. I will show that this is not a smart choice for Mila. Indeed Mila can do the following. She can sell one unit of good y . She now has a basket made of x units of good x and y -1 units of good y . But she also has 3 leva! With those three leva she can buy one unit of good x . So she ends up with a basket ( x +1, y � 1). Is she better off? The answer clearly is yes because the marginal utility of x , that it is the increase in utility following the consumption of an extra unit of x , was 2; while the marginal utility of y was only 1. There is therefore a net gain in terms of utility when you trade one unit of good y for one unit of good x .

Finally note that the MRS for that new basket containing x + 1 units of good x and y � 1 units of good y is not the same than when she had the basket ( x , y ). Following the law of decreasing marginal utility, because Mila has now more of x , the marginal utility of x , is lower ; and because she has less of good y , the marginal utility of good y is higher . Hence the ratio of the two marginal utilities is lower than before. We have started with a ratio of 2, we have now a ratio below 2. The MRS has moved closer to the ratio of prices (which, if you remember, was equal to 1). It is only when the MRS is equal to the ratio of prices that Mila can no longer increase her satisfaction by changing the composition of her basket.

This example illustrates how a consumer makes his choice. He looks on one hand, to prices. But a price alone does not mean a lot. What has economic significance is the price ratio . Now let us say that this price ratio is �in favour� of good x will he buy more x ? Well, this will depend on the ratio of the marginal utility, which in turn depends on how much of each good he already possesses.

Hence, a good does not have an �intrinsic� value. The value attached to a good depends on the consumer's preferences, that is on what needs he believes can be fulfilled thanks to that good; and on how much of that good and of other goods that consumer already have. The evaluation made by the consumer is therefore very contingent. It depends of time and place. That equality underlines the subjectivity of value .

Exercise:

Can you find examples in real life that apparently violate the law of decreasing marginal utility (think about addiction)? Can you reconcile this observation with the law? (Yes, if you recall that the law apply for preferences at a given point in time. If preferences are not stable, if they change over time, then the law does not apply. What we are attempting to do here is to explain the choice of one individual at one point in time, so t makes sense to take preferences as given. Addiction changes the preferences over time. It even changes the physiology of the individual).

Exercise:

Explain why the indifference curves of a rational consumer cannot intersect each other?

----------------------

1 Later in the course we will elaborate on the role of values.

2 Recall that a first constraint was that Mila has to satisfy her needs with the goods and services that she perceives has being available. This is important. One way to get wealthier, to improve his well-being is to hold more revenues. But there is another way, too often neglected, it is that you get better off when more goods and services are offered to your choice.

3 Those properties can also be demonstrated mathematically. They hold for almost all types of preferences.