8- Empirical IO, Identifying conduct Flashcards

1
Q

What was the chain of causal links that Structure-Conduct-Performance paradigm (SCP) presumed?

A

Industry structure (number of firms/market share etc) affected conduct (price setting, advertising etc) which affected performance (profit, cost efficiency etc)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What was Bain’s 1951 cross industry regression?

A

Average industry performance regressed on market concentration and other covariates
yᵢ = α + β₁HHIᵢ + β₂Xᵢ + εᵢ

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the Herfindhal-Hirschman Index (HHI)?

A

Measure of market concentration, sum of squared market shares (0 is perf comp, 1 is monopoly)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What was the main finding of Bain’s 1951 cross industry regression?

A

β₁ > 0 and strongly significant: more concentrated markets are more profitable on average

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What was Bain’s conclusion from his 1951 regression findings?

A

He takes this result as evidence for abuse of market power, so suggests the radical policy of breaking up firms to curb market power and increase welfare.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What was the main flaw of Bain’s reasoning in his 1951 findings?

A

Arguments rely on presumed causal nature of SCP: Estimation method (OLS) assumes, rather than proves, a single causal link running from Structure to Performance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What does SCP theory suggest wrt mark-ups?

A

Market share is related to mark-ups, but high market share is not sufficient for high mark up: if demand is relatively elastic, then even a high market share firm cannot enjoy high mark-up.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What does SCP theory suggest wrt causality between Structure and Conduct?

A

No causal link running from Structure to Conduct, they are determined simultaneously in equilibrium

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are the 2 main Chicago school ideas on Industry?

A

-When monopoly (or market power in general) exists, it is often transitory
-Firms become big for a reason: market share differences can reflect differences in efficiency

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What does empirical SCP suggest on market structure?

A

Cross industry regression could just reflect firm level
efficiencies: profits (mark-ups) can be high, not because of high prices, but because low costs. Therefore, breaking up firms could just lead to moving
production from efficient to less efficient firms

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the market power hypothesis?

A

All firms have high mark-ups

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is the efficiency hypothesis?

A

High mark-ups only for large (efficient) firms

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What are the 3 main reasons there is no single causal link between Structure, conduct & performance?

A

-Concentration and profits determined simultaneously, low cost producers have high market shares and mark-ups.
-Competition reduces profits, which discourages entry: competitive industries can be very concentrated.
-Structure is often not exogenous: entry barriers can be endogenous (e.g. advertising)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is the Bresnahan equation encompassing all market structures?

A

P(Q) = MC - λQP’(Q)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the intuition behind Bresnahan’s equation?

A

When elasticity changes, we get different responses: firms pricing close to 𝑀𝐶 will not react to changes in price sensitivity, while firms with some degree of market power will adjust prices and quantities

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What does conduct parameter λ show?

A

Indicator of the extent firms can increase prices by restricting output

17
Q

What are the 3 ways for estimating conduct (β₁) from the Bresnahan equation?

A
  1. Get cost data: estimate 𝑀𝐶 function and then you have β₁
  2. Demand shifts: not sufficient for identification in general
  3. Demand rotations: sufficient for identification
18
Q

How can you estimate the marginal cost function and β₁?

A

If high quality cost data are available, one could
estimate a cost function although (N)EIO argues that the econometrician should try to avoid
using cost data altogether

19
Q

How can you use demand shifters for identification?

A

When marginal cost is constant, demand shifters can trace out equilibrium outputs

20
Q

How can you use demand rotations for identification?

A

When demand rotates, firms will adjust to changes in
their market power (i.e. demand elasticity)

21
Q

What will the responses to demand rotation be in 3 main market structures?

A

-Perfect competition: no change
-Oligopoly: some quantity and price response
-Monopoly: drastic change

22
Q

What are the 3 steps for estimating change in slope (demand rotation)?

A

1.Estimate demand to get its coefficients
2.Create new demand variable using the supply coefficients: Qᵢ* = Qᵢ/(α₁ + α₃Zᵢ)
3.Use Qᵢ* to estimate modified supply function