Scribd
Upload
8 views3 pages

Proving and Disproving Preference Properties

Uploaded by

wgao1
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views3 pages

Proving and Disproving Preference Properties

Uploaded by

wgao1
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

ECON 357: Proving and Disproving Preference

Properties

Erik Lillethun
Colgate University

1 Hints
• If the preferences are defined as [x ≿ y] ⇔ [f (x) ≥ f (y)] for some function f ,
then this is a utility representation, and f is a utility function. You automatically
have completeness and transitivity, because ≥ on R is complete and transitive. If
f is a continuous function, you also have continuity, because xn ≥ y n for all n
implies limn→∞ xn ≥ limn→∞ y n (i.e., ≥ is a continuous relation). If f is a strictly
increasing function, you also have monotonicity. If f is a quasiconcave function,
you have convexity.

• Some types of violations of these properties will be obvious from the previous point.
If f is discontinuous, you get violations of the continuity assumption, if f is not an
increasing function or you flip the inequality, you get violations of the monotonicity
assumption, etc.

• If [x ≿ y] ⇔ [f (x) ≥ g(y)] for some functions f ̸= g, then there may be some


problems.

– Completeness may fail, because f (x) < g(y) (¬[x ≿ y]) does not imply f (y) ≥
g(x) (y ≿ x). Graphically, the UCS and LCS for a particular bundle are
not exhaustive; some other bundles are in neither. Consider the example of
g(z) = f (z) + 1. Some bundles will be too close together to have any relation
to each other.
– Transitivity may fail. If f (x) ≥ g(y) and f (y) ≥ g(z), this does not allow you
to compare f (x) and g(z), because f (y) ̸= g(y). Graphically, for two bundles
x ≿ y, their upper contour sets are not necessarily nested within one another.
The same goes for their lower contour sets.

0
– Continuity will still hold if both f and g are continuous functions.
– Monotonicity may fail. Often, you have a thick indifference set, because the
UCS and LCS overlap too much. Thick parts of the indifference set means
that you can move up and to the right (more of both goods), and still be
within the indifference set. Consider the example where g(z) = f (z) − 1.

• If ≿ is defined using >, then . . .

– Completeness may fail, because x may not even be weakly preferred to itself!
– Transitivity typically holds, because > satisfies transitivity.
– Continuity typically fails, because > is not continuous. The two sides of the
> can converge to the same thing. See lexicographic preferences. Graphically,
you may get dashed lines or holes in your UCS and/or LCS, which usually
creates problems for some limits converging to a point on the dashed line or
hole.
– Monotonicity typically holds, because increasing a quantity causes it to be >
than it used to be.

• If ≿ is defined using “and”, then . . .

– Completeness often fails, because there exist pairs of bundles where each satis-
fies only one of the two conditions relative to the other bundle. See ≥ relation
on R2 .
– Transitivity typically holds. Each condition satisfies transitivity individually,
and these combined give you all required conditions for x ≿ z.
– Continuity typically holds. Each condition satisfies continuity individually, so
all conditions continue to hold in the limit.
– Monotonicity typically holds. If both f (x) ≥ f (y) and g(x) ≥ g(y) need to
hold, and both f and g are strictly increasing functions, then an increase in
the bundle quantities will increase the values of both f and g.

• If ≿ is defined using “or”, then . . .

– Completeness typically holds. The “or” of two conditions admits even more
possibilities (i.e., is more forgiving) than just a single condition, so it is more
likely for each ≿ comparison to be true.

1
– Transitivity may fail. If Condition 1 but not Condition 2 holds for x ≿ y, and
Condition 2 but not Condition 1 holds for y ≿ z, then both conditions may
fail for x ≿ z. We did an example of this in class.
– Continuity typically holds. It is hard to explain why without relying on pre-
vious real analysis background, so just take this for granted.
– Monotonicity typically holds. Same reason used for “and” applies here as well.

To summarize some of the less obvious problems you may see...


Peculiar Feature Typical Violations
Uses a > or < completeness, continuity
Different functions on each side of inequality completeness, transitivity, monotonicity
“and” completeness
“or” transitivity

You might also like