2. Class
Topics
1. Guest
Speaker:
Marney
Morris
2. Review
of
Price
ElasEcity
and
ProducEon
FuncEons
3. Market
Power
and
Structure
Monopoly,
Oligopoly,
Duopoly
–
Game
Theory
MonopolisEc
CompeEEon
–
Brand
Power
4. Market
Failure
ExternaliEes,
RegulaEon,
Environmental
Policy
5. Guest
Speaker:
John
Bodt
4.
Interac2ve
Design
is
Self
Explanatory
If
a
design
is
good,
a
user
will
immediately
know
how
to
use
it.
To
make
a
design
self
explanatory
there
are
three
simple
rules:
01.
Present
a
minimal
amount
of
informaEon.
02.
Arrange
the
informaEon
so
there
is
an
obvious
starEng
point.
03.
Build
each
"next
step"
on
the
logic
of
the
step
before.
When
designing
for
screens,
here
a
couple
of
addiEonal
Eps:
Reduce
your
variables
so
the
choice
on
the
next
screen
is
obvious.
For
example,
when
you
use
the
same
font,
same
size
font,
same
color
font,
and
line
all
the
text
up,
any
text
that
is
different
(bold,
different
color,
or
different
locaEon)
is
easily
disEnguished.
Register
everything.
When
things
shiY
from
screen
to
screen
they
are
distracEng.
29. How
much
does
it
cost
to
make
a
donut?
Premium
donut
-‐
$2
-‐
$3.50
price
Basic
donut
-‐
$0.80
-‐
$1
Inputs:
Flour,
Sugar,
Oil,
Labor,
Overhead
38. Monopolis2c
Compe22on
CompeEEon
on
brand
or
other
forms
of
market
segmenta2on
You
Are
The
determinaEon
has
to
do
with
Here.
whether
there
are
subsEtutes
–
perfect,
imperfect,
improvised
or
otherwise
39. Strategy
of
MonopolisEc
CompeEEon
• Non-‐price
compeEEon
• Influence
over
demand
via
adverEsing,
branding
• Increase
switching
costs
• Bolster
number
of
complements
• Extended
service
agreements,
frequent
upgrades
• Network
externaliEes
• Technical
standards
(Blue
Ray
v.
HD
DVD)
49. Example:
Maple
Bacon
Donuts
Say
Dynamo
increases
the
price
from
$3
to
$4.50
for
maple
bacon
donut.
They
were
selling
100
a
day.
Now
they
sell
75.
What
is
the
PED?
Revenue?
If
$2
per
donut
and
$75
overhead,
what
is
their
profit?
55. Perfect
vs.
Imperfect
CompeEEon
• Efficiency
vs.
Power
• In
perfect
compeEEon,
firms
are
powerless
–
They
do
not
set
price.
They
can
only
respond.
• Economists
believe
perfect
compeEEon
creates
efficiency
which
allocates
resources
effecEvely
–
“allocaEve
efficiency.
56.
57. Perfect
CompeEEon
• Infinite
Buyers/Infinite
Sellers
–
Infinite
consumers
with
the
willingness
and
ability
to
buy
the
product
at
a
certain
price,
Infinite
producers
with
the
willingness
and
ability
to
supply
the
product
at
a
certain
price.
• Zero
Entry/Exit
Barriers
–
It
is
relaEvely
easy
for
a
business
to
enter
or
exit
in
a
perfectly
compeEEve
market.
• Perfect
Factor
Mobility
-‐
In
the
long
run
factors
of
producEon
are
perfectly
mobile
allowing
free
long
term
adjustments
to
changing
market
condiEons.
• Perfect
Informa2on
-‐
Prices
and
quality
of
products
are
assumed
to
be
known
to
all
consumers
and
producers.
• Zero
Transac2on
Costs
-‐
Buyers
and
sellers
incur
no
costs
in
making
an
exchange
[Perfect
mobility].
• Profit
Maximiza2on
-‐
Firms
aim
to
sell
where
marginal
costs
meet
marginal
revenue,
where
they
generate
the
most
profit.
• Homogeneous
Products
–
The
characterisEcs
of
any
given
market
good
or
service
do
not
vary
across
suppliers.
• Constant
Returns
to
Scale
-‐
Constant
returns
to
scale
insure
that
there
are
sufficient
firms
in
the
industry.
58. monopoly
• Single
seller
• High
barriers
to
entry
• Price-‐maker
not
price-‐taker
• No
subsEtutes
(pure
monopoly),
or
power
in
proporEon
to
availability
of
subsEtutes
• If
more
subsEtute
products
are
available,
demand
elasEcity
will
be
greater
and
mark-‐up
will
be
lower
59. Monopsony
vs.
Monopoly
(one
buyer,
many
sellers
vs.
one
seller,
many
buyers)
WalMart,
Gates
Founda2on,
Warren
Buffeg.
.
.
60. Demand
Curves
(market
power
means
being
able
to
set
price)
P
P
Q
Q
The
market
overall
Facing
the
individual
firm
62. Monopolists
are
the
whole
market
• Firm
demand
curve
=
industry
demand
curve
• For
compeEEve
firm:
MR
=
MC
=
P
because
P
is
“taken.”
• If
you
tried
to
raise
price,
you
would
lose
all
sales.
• For
monopolist,
MR
≠
P;
rather...
• MR
is
always
less
than
P
so
long
as
demand
curve
is
downward
sloping
64. Marginal
Revenue
for
Monopolists
Example:
Perfect
compeEEon:
price
=
$5
MR
=
$5
no
ma_er
what
Monopoly:
To
sell
more
units,
you
have
to
lower
price:
At
price
of
$5,
you
sell
20.
TR
=
100.
At
price
of
$4,
you
sell
30.
TR
=
120.
MR
=
Change
in
total
revenue
divided
by
change
in
output
=
(120-‐100)/(20-‐30)
=
20/10
=
2
65. Bushels
of
Fish
–
p.
199
Marginal
Revenue
<
Price
for
monopolist
for
every
sale
aYer
the
theoreEcal
first
66. Schiller,
p.
200
monopolist
always
seek
to
produce
so
that
MR
=
MC
67. Takeaways
• The
decision
of
how
much
to
produce
is
as
criEcal
for
monopolists
as
compeEEve
firms
but
in
a
different
way
• Over-‐producEon
raises
costs
above
the
point
where
MR
=
MC,
and
profits
would
not
be
maximized
• Even
monopolists
can’t
charge
the
highest
possible
rate
if
they
want
to
be
profit-‐
maximizing
68. oligopolies
Behave
like
a
monopoly
but
made
up
of
individual
actors,
so
they
face
problems
of
coordinaEon
and
game
theory
Examples:
oil
cartel
(OPEC),
airlines,
eyeglasses
pre-‐
Warby
Parker
71. cargo
surcharge
collusion
21
airlines
arEficially
inflated
passenger
and
cargo
fuel
surcharges
between
2000
and
2006
to
make
up
for
lost
profits.
In
July
2005,
LuYhansa
told
the
JusEce
Department
about
airlines
conspiring
to
set
cargo
surcharges.
By
ValenEne's
Day
2006,
FBI
agents
and
their
counterparts
in
Europe
made
the
invesEgaEon
public
by
raiding
airline
offices.
AYer
those
raids,
BriEsh-‐based
Virgin
AtlanEc
came
forward
about
its
role
in
a
similar
scheme
to
set
fuel
surcharges
for
passengers.
72. Price
Agreement
• 19
execuEves
have
been
charged
with
wrongdoing
–
four
have
gone
to
prison
• 21
airlines
have
paid
more
than
$1.7
billion
in
fines
in
one
of
the
largest
criminal
anEtrust
invesEgaEons
in
U.S.
history.
• The
court
cases
reveal
a
complex
web
of
schemes
between
mostly
internaEonal
carriers
willing
to
fix
fees
in
lockstep
with
compeEtors
for
flights
to
and
from
the
United
States.
• Convicted
airlines
include
BriEsh
Airways,
Korean
Air,
and
Air
France-‐
KLM.
No
major
U.S.
carriers
have
been
charged.
• The
price-‐fixing
unraveled
largely
because
two
airlines
decided
to
come
clean
and
turn
in
their
co-‐
conspirators.
75. Some
basic
characterisEcs
of
games
1.
Simultaneous
v.
Sequen2al
Whether
it
is
played
sequen*ally
(one
player
chooses,
then
the
other
like
Ec-‐
tac-‐toe
or
chess)
or
simultaneously
(both
choose
at
the
same
Eme
like
two
magazines
choosing
newspaper
headlines).
The
Prisoner’s
Dilemma
is
a
simultaneous
game
because
they
are
quesEoned
at
the
same
Eme.
Simultaneous
games
tend
to
be
modelled
in
tables,
sequenEal
games
in
trees
(see
later).
2.
Backward
Induc2on
Look
forward,
reason
back.
In
a
simultaneous
game,
this
means
to
imagine
your
way
into
your
rival’s
shoes,
see
what
their
best
strategy
is
and
assume
they
will
be
doing
the
same
to
you
(ad
infinitum),
and
reason
back
to
what
to
do
accordingly.
In
a
sequenEal
game,
it
means
follow
your
move
through
a
few
steps
and
if
you
would
fail
on
step
three
or
four
of
forty,
then
don’t
undertake
step
one.
(In
chess,
if
you
see
a
checkmate
five
moves
away,
you
don’t
move
your
pawn
out
on
move
1.)
76. 3.
Dominant
Strategy
Whether
there
is
one
way
to
behave
that
always
works
best,
or
a
dominated
strategy
in
which
case
it
is
always
the
worst
outcome—or
neither.
The
way
to
figure
out
whether
you
have
a
dominant
strategy
is
mentally
to
picture
a
row
of
the
table
and
to
superimpose
it
on
the
other
rows.
If
it
is
always
higher
for
every
box,
it
is
dominant.
4.
Credibility
and
Commitment:
Threats
and
Promises
Many
games
–
eg,
nuclear
disarmament,
poliEcs,
parenthood
–
rest
on
the
credibility
of
threats
and
promises.
An
incenEve
has
to
be
believable,
and
a
threat
has
to
be
plausible
and
in
proporEon
to
the
offense.
Commitment,
as
in
the
game
of
chicken
(cars
driving
head
on)
is
a
form
of
threat
or
promise.
5.
Repeat
or
Single
Game?
Be
more
fierce
in
a
single
game
and
more
cooperaEve
in
a
repeated
game.
6.
Predictability
It
is
always
strategically
be_er
to
be
unpredictable,
as
in
a
sport
like
football.
77. Expected
Value
“a
good
decision
can
have
a
bad
outcome”
If
I
play
poker
with
Bob
and
Anne,
these
three
things
could
happen:
• A
50%
chance
of
making
$20
• A
30%
chance
of
losing
$20
• A
20%
chance
of
breaking
even
Expected
value
=
the
probability
x
the
value
of
the
outcome
What
is
the
expected
value
of
each
outcome?
50%
chance
of
$20
=
.5
x
$20
=
+$10
30%
chance
of
$20
=
.3
x
-‐$20
=
-‐$6
20%
chance
of
zero
=
.2
x
$0
=
0
78. Expected
Value
Example
2:
I
am
going
to
teach
a
class
and
charge
$30.
• If
I
get
great
a_endance,
100
people
will
come.
($30
x
100
people
=
$3,000)
• If
I
get
poor
a_endance,
20
people
will
come.
($30
x
20
people
=
$600)
At
50/50
odds,
that
is
$1500
+
$300
=
$1800
At
20/80
odds,
that
is
$600
+
480
=
$1080
79. Nash
Equilibrium
each
player
is
assumed
to
know
the
equilibrium
strategies
of
the
other
players,
and
no
player
has
anything
to
gain
by
changing
only
his
or
her
own
strategy
unilaterally
80. Cournot
Companies
compete
on
the
amount
of
output
they
will
produce,
which
they
decide
on
independently
of
each
other
and
at
the
same
Eme
Firm
treats
the
output
level
of
its
compe&tors
as
fixed
and
decides
how
much
it
should
produce.
Cournot
Equilibrium:
each
firm's
output
maximizes
its
profits
given
the
output
of
the
other
firms
81. Demand
Curve
in
Oligopoly
• Li_le
consensus
among
economists
on
oligopoly
price
theory
•
Kinked
demand
curve
is
a
fudge:
really
economists’
way
of
saying
“we
don’t
know
how
rivals
are
going
to
respond
to
an
oligopolist’s
change
in
price,
so
here
are
the
major
opEons”
• If
rivals
match
price
reduc*ons
but
not
increases,
kink
results
• Unmatched
price
decrease
leads
to
increased
sales
and
market
share
at
lower
prices:
unlikely
outcome
• Unmatched
price
increase
leads
to
reduced
sales
at
higher
price
• Matched
price
decrease
leads
to
growth
in
total
quan*ty
demanded,
but
revenue
falls
•
Pricing
behavior
cannot
be
precisely
formulated
for
oligopoly,
because
it
is
highly
dependent
upon
assumpEons
used
84. Natural
Monopoly
Industry
like
a
uElity
or
the
post
office
that
is
so
vast
only
one
firm
can
support
the
fixed
cost
structure
85.
86. Government
RegulaEon
of
Natural
Monopolies
1.
Regulate
price
• P
=
MC:
consistent
with
opportunity
cost
• because
MC
is
always
less
than
ATC,
would
cause
a
loss
on
every
unit
produced
2.
Provide
Subsidy:
•
P
=
MC
+
subsidy
3.
Regulate
Profit:
• P=ATC
• Doesn’t
regulate
cost
4.
Regulate
output:
• Require
minimum
level
of
output
• Doesn’t
specify
quality
88. Market
Power
• Consumer
segmentaEon
and
discriminatory
pricing
• Divide
the
market
according
to
willingness
to
pay,
and
charge
different
prices
89. Barriers
to
Entry
patents;
monopoly
franchises;
exclusive
suppliers
(control
of
key
inputs)
or
outlets;
predatory
liEgaEon
(large
firms
have
resources
to
liEgate);
acquisiEon
of
compeEtors;
scale
economies,
brand
(to
an
extent)
90. Some
Determinants
of
Market
Power
• ReputaEon
(for
toughness
toward
new
entrants)
• Long-‐term
contracts
with
key
suppliers
(that
make
entry
hard)
• Licenses/patents
• Learning
curve
effects
(first
mover
advantage)
• Brand
advantage
(strong
for
“experience
goods”)
• High
exit
costs
(deters
risk
of
entry)
91. Tests
of
Market
Power
• Number
of
producer
firms
• Size
of
each
firm
• Barriers
to
entry
• Availability
of
subsEtute
goods
In
essence,
is
the
market
contestable?
92. Red
Flags
for
AnEtrust
• Patents
• CompeEtors
must
establish
alternate
means
of
producing
or
license
process
from
patent
holder
• Supply
&
DistribuEon
Control
• Control
distribuEon
outlets
• Also:
control
essenEal
supplies,
resources
• Mergers
and
acquisiEons
94. MicrosoY’s
anEtrust
defense:
That
prices
hadn’t
increased,
as
they
would
have
under
expected
monopoly
condiEons
of
restricEng
supply
Also,
that
significant
anEtrust
acEon
would
sEfle
innovaEon
The
reality
is
that
prices
didn’t
need
to
be
super
high,
only
for
MR
=
MC
Court
found
that
MicrosoY
had
behaved
illegally
to
maintain
its
monopoly
posiEon
and
to
take
advantage
of
the
fact
that
suppliers
and
consumers
had
limited
alternaEves
given
network
effects,
etc.
95. Cross
elasEcity
as
measure
Cross
elasEciEes
help
answer
quesEon
of
market
definiEon:
•
%
change
in
QuanEty
demanded
of
X
divided
by
%
change
in
price
of
Y.
Analysis
of
cross-‐elasEcity
shows
the
strength
of
subsEtuEon
• If
cross-‐elasEciEes
are
strongly
posiEve
–
if
increase
in
price
of
product
being
examined
causes
significant
increase
in
quanEty
demanded
of
another
product
=
effecEve
subsEtutes
96. ConcentraEon
Measures
of
Monopoly
1. ConcentraEon
RaEo
–
market
share
of
top
4
firms
>60%
is
considered
oligopoly
2.
Herfindahl-‐Hirschman
Index
–
sum
of
squares
of
all
the
market
concentraEons
– if
merger
creates
an
HHI
>
1800:
DOJ
will
challenge
– if
merger
creates
an
HHI
between
1000
–
1800:
DOJ
will
challenge
if
merger
if
likely
to
increase
HHI
by
100
points
or
more
100. Public
Good
• ConsumpEon
by
one
person
does
not
interfere
with
consumpEon
by
another
– Private
good:
donut
– Public
good:
water,
air
• Free-‐rider
problem
• The
market
tends
to
overproduce
private
goods
and
underproduce
public
ones
101. ExternaliEes
• Costs
or
benefits
of
producEon
that
are
borne
by
a
third
party
–
Costs
that
don’t
get
added
or
subtracted
+
beekeeper
who
lives
next
to
a
flower
grower
-‐
polluEon,
excessive
noise,
environmental
damage
• PosiEve
externaliEes
tend
to
get
subsumed
into
business
models,
negaEve
externaliEes
are
oYen
regulated
102. Examples
of
ExternaliEes
• Costs
of
PolluEon
• EsEmaEng
the
cost
of
environmental
damage
• Tangible
or
specific
costs
can
be
calculated
by:
• Diminished
life
expectancies
• Lost
work
days
• Direct
medical
costs
• ProducEvity
losses
• Cost
of
replacing,
restoring,
remediaEng
environmental
damage
103. Remedies?
• Regulate
(producEon
or
purchase)
• ShiY
to
consumers
(consumer
tax)
• Assign
Property
Rights
(Coase
Theorem)
104. Coase
Theorem
A
resource
alloca*on
can
be
efficient
as
long
as
contrac*ng
costs
are
sufficiently
low
and
property
rights
are
assigned
clearly,
enforced
well,
and
exchanged
readily.
• Ronald
Coase,
1960
• The
free
market
is
more
powerful
in
producing
efficient
outcomes
than
previously
thought
• Pricing
in
externaliEes
lets
people
trade
them.
Even
if
the
price
isn’t
set
right
the
first
Eme,
the
market
will
correct
it.