a theory in which they assume that people are rational, utility maximizers. The two

elements that I don't like...one, is the idea that they are rational and the other thing

is, what they're rational about, namely, maximizing utility. And I don't know whether you in your

writing use this phrase, but lots of economists do. Let's start with utility. I don't understand

what it even means, because you can't put a little measure into a person's brain and

read off utility and Irving Fisher had the idea of this unit...until...we know there

isn't such a unit. Why can you talk about maximizing something that you can't measure?

BECKER: Alright, this is a good subject to start with, because I think we do disagree

on this one. And I think you hit the nail right on the head with (regards to this being)

a serious disagreement. I do believe that it's useful. And that's all I would say, it's

a useful way of looking at consumer behavior...to assume that they're maximizing utility. Whether

we want to call that rational or not...to me (that) is a secondary issue, but I do believe

they are maximizing utility. It's not that I believe there is such a unit of until that

we could ever measure. To me, maximizing utility simply means the following: That consumers

can order all the opportunities they have available to them...possible choices. They

can order them so they prefer some more then to others. They can rank them. And they have

limited resources. They have limited income...wealth, time, whatever it may be. And that they attempt

to choose that possibility...that combination of goods and so on that is ranked the highest

consistent with their limited resources. Now I think this is a testable theory. That we

can test it by various implications about how they would respond to prices, income,

changes in wages, whole host of other patterns. It can become a very rich theory. I think

it is becoming a very rich theory. It's not an empty theory. But it does not require the

assumption that everybody has some kind of utils in their head that they're calibrating

and they convey this gives me ten utils and five...Anyone can do away with utils entirely,

but just ask where do they prefer...they want to get more rather than less. They want to

do that in goods that they have a higher preferences for, rather than others. And my claim would

be, this isn't the best theory possible, but it's the best theory that we have. I don't

know of any alternative that gives us the insight that this does.

COASE: You just said that one can get rid of it, why don't we get rid of it and just

study choices? People choose and we can study what choices they make and we can then discuss

whether their consistent or not...which is in some ways necessary for some of the tests.

My guess would be that you find a lot of inconsistencies in consumer behavior.

BECKER: Well, I think consistency is the task that, in fact, one can show that the usual

theory in terms of ranking and ordering, complete ordering of opportunities, is basically equivalent

to inconsistency tests that people behave consistently. Now maybe there had be a lot

of inconsistencies. But let me sort of point out that there had been a few attempts to

study the problem of consistency. Not so much with individual household data, but with more

group data, because one hasn't had the rich household data to do that, which is a limitation.

Let's say if people looked at long term English consumption patterns, American patterns, patterns

of different groups comparing different countries...and it had been very difficult to find dramatic,

very common examples of inconsistent behavior in that type of data. I'm sure they'll show

up more commonly in individual... COASE: You know a lot of work is being done

in showing that there are anomalies in individual behavior. I don't know about this well, but

it wouldn't surprise me if they're right. In fact...

BECKER: But its been only based on experimental evidence not on actual choices. And I think

you and I would agree that we don't necessarily have this same theory to predict how people

behave in experimental situations as they would behave in market situations.