Week 6 - History, expectations, traps and multiple equilibria 2

Question 1

What are the 2 economic mechanisms that can lead to multiple equilibria?

Answer 1

Complimentarity and IRS

Question 2

True or false: IRS makes it easier for firms to enter the market

Answer 2

False - makes it harder

Question 3

What is the difference between intermediate inputs in developed and developing countries?

Answer 3

Construction in developing countries is labour intensive e.g. remove rubble by hand, mix cement by hand etc
Construction in developed countries is more automated e.g. cranes, diggers, precast concrete

Question 4

Why is producing an intermediate input only worth it if demand is large enough?

Answer 4

Since this production of these intermediate goods is subject to IRS e.g. each intermediate good is produced by a single firm with a high fixed cost - producing 1 crane is v expensive

Question 5

Assume a final good is produced using only labour and intermediate goods. What kind of production function would we see?

Answer 5

Constant returns to scale PF. Doubling these doubles output

Question 6

How can this production function lead to increasing returns to scale instead?

Answer 6

increase the variety/range of intermediate goods as well as the amount of labour and intermediate goods

Question 7

Using this mechanism, how can a poor economy be stuck in a poverty trap

Answer 7

Poor economy, low demand for final product, low demand for intermediate inputs, high prices for these inputs, use few varieties of intermediate inputs; switch towards labour, lower productivity, low income per person, low demand

Question 8

How does IRS fit in this mechanism to switch to an equilibria with better demand?

Answer 8

IRS allows intermediate inputs industry to thrive when demand is high, leading to a large variety of intermediate inputs being produced at low cost, and further expanding production of the final good.

Question 9

These theories of multiple equilbria have assumed a limited market size. What does this mean?

Answer 9

Sectors cannot expand past their limited demand. If production does expand, the price will fall in order to find buyers. Similarly, expanding production will push up the price of intermediate inputs.

Question 10

Why do the models for multiple equilibria not work if we have competitive markets?

Answer 10

since in these markets agents are price takers so could buy and sell unlimited quantities at a fixed price. e.g if steel expanded production, this would not bid up prices for coal (the intermediate output) so the steel sector can expand without being constrained by backward and forward linkages.

Question 11

What kind of sectors are the theories we’ve discussed relevant for?

Answer 11

Sectors with goods that are not traded internationally e.g, infrastructure.
Intermediate goods as well often come with a package of producer services which make them less tradable.
Transport costs often high in developing regions e.g. SS Africa

Question 12

What are two other ways history matters in the convergence to bad equilibria?

Answer 12

Social norms
Status quo bias

Question 13

How do social norms result in multiple equilibria?

Answer 13

Peoples behaviour is constrained by social norms: costly to deviate. Some norms can have a negative effect.
It’s basically another example of complementarity: the more people depart from the norm, the lower the cost of deviating. Therefore can get potentially stuck in potentially bad equilibrium.

Question 14

What is an example of a bad social norm?

Answer 14

Foot binding in 10th century China

Question 15

How does the status quo bias lead to convergence to a bad equilibria?

Answer 15

If a new policy is proposed where there are winners and losers, but the gains far outweight the costs, it might not be implemented.

Question 16

What factors drive the status quo bias?

Answer 16

It’s very difficult to value the different implications of a project/policy
Difficult to quantify gains and loses, and people who gain and lose

Question 17

What role does uncertainty play in the status quo bias?

Answer 17

Uncertainty about who gains and loses might cause a policy or reform that everyone agrees is beneficial for society to be voted down.

Question 18

Theoretically, the winners of the policy could compensate the losers to push through the policy. Why might this not work?

Answer 18

Because winners might renege on their promise to compensate the losers once the policy has been implemented.

Question 19

Why might policies which are bad for an economy are kept in place?

Answer 19

Gains for an existing policy that is bad for overall welfare might accrue to a small number of people that can easily organise themselves and lobby for no change.

Question 20

Briefly explain Munshi and Myaux (2006)

Answer 20

They use a social norms-based theory to
explain some features of the transition in rural Bangladesh from high to low fertility, and test it.

Question 21

Why is the fact that fertility is regulated by social norms relevant in this case?

Answer 21

This can explain the slow and differential response to an external intervention, as each community gradually converges to a new reproductive equilibrium.

Question 22

Why is there a long delay in the change of this social norm?

Answer 22

Information on what others do is severely restricted, and women are mostly restricted to their house, except to other houses on social visits.

Question 23

Describe the theoretical model by Munshi and Myaux (2006) (6)

Answer 23

• A women can take 1 of two actions: no family planning (t) or family planning (m)
• Each women gets intrinsic positive net utility from fertility control.
• 2 types of women: reformists get higher net utility from m then conformists
• deviating from social norm bears a social cost: each period you are randomly matched to another women and you suffer a cost if she chooses t and you choose m
• family planning is an external intervention that increases the net utility from m
• each period community health workers contact a fraction of the women in the village and convince any reformist to take up m for a single period

Question 24

What is a problem with this project for individuals?

Answer 24

On the outset, individuals do not know whether they live in a community with a lot of reformists or not.

Question 25

How does this problem solve itself?

Answer 25

They find out gradually, through social interaction since in every period you are randomly matched to another women in your community and observe their decision to see if you live with loads of reformists or conformists.

Question 26

Explain the dynamics of the model with a lot of reformists compared to a lot of conformists

Answer 26

CHWs persuade a fraction of the people they meet (the reformists) to switch to family planning (m) for one period ⇒
since there are a lot of reformists, when a woman is randomly matched to someone from her society, this is more likely to be someone who chose m ⇒
this strengthens our individual’s belief that she lives in a society with a lot of reformists ⇒
her expected utility from choosing m increases ⇒
she is more likely to choose m ⇒
more women choose m in the village ⇒
you are more likely to be matched to someone who chooses m ⇒…

Question 27

What would happen in a society with few reformists?

Answer 27

Will remain in equilibrium without family planning and with high fertility.

Question 28

Describe the empirical model created from this study

Answer 28

OLS - Using a contraceptive depends on: control variables, contraceptive use in previous period, lagged average contraceptive use in the village

Question 29

What is a problem with this empirical model

Answer 29

error term includes some village level observables which may effect the use of contraceptives between villages. e.g. quality of villages health workers.

Question 30

What do Munshi and Myaux add to the regression to decrease ednoegneity

Answer 30

they create 2 variables: Lagged average contraceptive use within your own religious group in your village and lagged average contraceptive use for the other religious group(s).

Question 31

What does the social norms theory predict about the coefficients for these 2 variables?

Answer 31

Own religion = positive
other religion = 0