Labour Economics Flashcards
Explain the neoclassical model of labour supply
U=f(C,L) represented in an indifference curve. When wage increases:
- Substitution effect
- Income effect
Explain the rise in women’s labour force participation
Theory suggests a fall in their reservation wage (difficult to measure), and a rise in the market wage (easier to observe)
Women’s time in the household reduced due to:
- Time saving technologies
- Reduction in fertility
- Improvements in childcare services
- Changes in gender norms
Bailey:
- Isolates the effect of the enovid pill on women’s LFP through variations in state laws
- Finds legal access before 21 reduced likelihood of first birth before 22
- Increase in women LFP
- Raised number of annual hours worked
Explain why female LFP hasn’t been rising in all countries
1) Ngai - structural transformation. A U-shaped relationship between logGDP and FLFP
- Due to decrease in agriculture, increase in services
- Then structural transformation as women are incorporated
2) Fernandez - social norms. Explains the importancce of intergenerational effects. The presence of a man brought up in a family in which the mother worked has been a factor
- Can policy accelerate?
Do welfare programs that provide assistance to certain groups lower work incentives?
Depends on the type and implementation (e.g. if cash grant, depends on if leisure is a normal or inferior good)
Most common form is Earned Income Tax Credit (EITC). It affects labour supply on the extensive margin (draw people into the labour force), and on the intensive margin (unclear effects). It’s tax credit that decreases with income earnt
Eissa uses the Current Population Survey to implement a diff-in-diff for the effect of EITC on unmarried women with and without children:
- The result was a 2.4% increase in labour force participation for unmarried women with children
- Conclude that the increase in LFP of women is due to a number of factors
Empirical evidence is mixed on the effects of EITC on the extensive margin of labour supply
Klevin looks again, at the reforms of EITC. Finds the only significant effect was that there was a booming economy (and unobservable factors), also EITC can be seen as a cash grant instead
Explain the effect of minimum wage on employment (Literature)
Card and Krueger (1994):
- What is the effect of minimum wage on unemployment?
- Uses a difference-in-difference in the case of fast-food jobs in NY and NJ
Why fast food restaurants?
1) Employ low-wage workers
2) Comply with minimum wage regulations
3) Products and job requirements are homogenous
- Treatment of a rise in minimum wage in NJ in a recession
- Baseline characteristics:
1) High compliance in minimum wage
2) Substitute skilled worked for low skilled? Some evidence
3) Increased price charged to consumers? Not significant
4) Reduction in non-wage compensation? Not significant
5) Fewer new restaurant openings? No evidence
- For robustness, checks for external variables by using the outcome as change in employment
- Conclude that ~30 years ago, consensus that minimum wage causes job losses, but recent empirical evidence challenges this
- Problems with external validity, and if labour markets are not actually perfectly competitive
McGuinness (2019):
- What is the impact of the 2016 increase in minimum wage in Ireland?
- No evidence that min. wage led to an increase in unemployment in min. wage employees
- However, a reduction in hours worked of ~0.6hrs/week
Can individual employers have market power?
An individual firm faces a perfectly elastic labour supply curve. In perfect competition, it hires as many workers as it needs at market wage
In the case of monopsony, they must pay a higher wage.
If the government imposes a minimum wage then both wages and employment may increase. This could explain Card and Krueger but only if the fast-food places have market power.
There is evidence of monopsony power in the case of nurses wages in hospitals. They find local hospitals face upward sloping labour supply curves, implying they have considerable market power
Explain migration and why you might see rates lower than you would expect
Individuals weigh costs vs benefits (explain)
Within the US, the probability of migration is negatively associated with age and positively correlated with education
There could be lower migration rates between rural and urban areas:
- Credit constraints
- Risk
- Social networks
- Social norms
Can you incentivize migration?
Bryan implemented an experiment where poor workers were provided monetary incentives to migrate to cities. In Bangladesh, the incentives were cash grants or credit
Migration improves consumption and caloric intake
However, experimenting with migration is risky (…), but once they learn it’s possible and useful to migrate, they continue even after the experiment finished
How do workers sort themselves across opportunites?
Roy Model says earnings depends on skills, which are transferrable. If the returns to skills are higher in the destination, migrants will be positively selected
Workers with skills > s* are better off moving
The Roy model implies the returns to skills are all relative (if you worsen conditions in the destination, it shifts the line)
This introduces measurement problems as migrants are not randomly selected, so measuring returns to migration will be biased
Abramitzky compared the earnings of Norweigan migrants to the US during 19th century, with their brothers who remained in Norway:
- Considered the unobserved ability shared between brothers
- Migrants earn ~70% higher than their brothers who stayed
- Also revealed evidence of negative occupational selection for migrants
- Little selection on average but:
- Negative selection of urban migrants
- Positive selection of rural migrants
What is the effects of migration on the labour market? (Theory)
Answer depends on the substitutability between migrants and natives. If migrants and natives are substitutes, more migrants imply high labour supply, lower wages:
(See graph)
But if migrants and natives are complements, more migrants means natives are more productive, demand for native labour increases, higher wages, more employment:
(See graph)
The long-run effects shifts both D and S right
What is the effects of migration on the labour market? (Evidence)
Boustan analyses the example of the New Deal during the Great Depression and extreme weather events to instument for migration between US regions. Finds that there is little effects of hourly earnings, and prompted some residents to move away.
The Mariel Boatlift (Card 1990):
- Fidel Castro declared Cuban nationals could move freely to US
- By September 1980, 125000 Cuban nationals had moved
- Compared trends in wages in Miami with four placebo cities
- Findings:
1) No evidence of a differential change in wages in Miami relative to the average of 4 cities
2) No effect on wage-stability of non-Cubans
3) No effect for unemployment rates of non-Cubans
4) No effect for wages/unemployment of Cuban immigrants
5) Miami labour market structure and slow population growth shows rapid absorbtion of Mariel immigrants with minimal disruption
Borjas offers a different methodology. Notes 60% of Mariel migrants were without high school diploma, so the supply shock affects high-school dropouts
- Compares the wages of high-school dropouts vs rest of US, which took a sharp drop in 1980 but recovered.
Compare the trends of migration
Tumen analysed the effect of the Syrian war mid-2011. By 2017, more than 5 million migrants moved to neighbouring countries
- Conducts a diff-in-diff in Turkey, with a treatment of 5 regions, and control of 4
- Assumes variation is driven by supply of accomodation and subsidies provided by Turkish gov. in treatment regions, and not by other diffferences
- Native informal employment down 2.3ppt
- Native formal employment up 0.5ppt
- No significant effect on native wages
- Why? Syrian refugees not granted official work permit
Explain the effects of integration of forced migration
Economic models highlight two factors that affect migrant integration:
- Selection
- Determinants of accumulation of human, social capital
Integration of forced migrants differ from normal migration:
- Economic migrants move based on relative opportunities
- Refugee migration caused by different factors
- Refugees migrate with worse HC so are more likely to start on significantly lower wages
Incentives make it more difficult to integrate:
- Lower HC, lower costs of investment, rate of return on investment is higher
- However, uncertain futures! May reduce incentives to invest in host HC
- Traumatic experiences means health issues
- Journey involves long stays in intermediate locations
Define and measure discrimination empirically
Definition:
- Taste based discrimination. Cost of hiring worker is w(1+d)
- Statistical discrimination. Due to imperfect information. Employers use worker characteristics as a signal of their expected productivity
Measuring:
- Blinder-Oaxaca decomposition
Problems with interpretation of Blinder-Oaxaca:
- Omitted variable bias
- Adding more and more variables could obscure interpretation
Explain the experimental measures of discrimination, and their advantages/disadvantages
1) Audit studies
- Individuals matched for all characteristics but the expected discriminatory variable
- Neumark shows the role of sex discrimination in restaurant hiring
- Results shows discrimination against women in high-price restaurants, and discrimination against men in low-price restaurants
- Weaknesses is that the key assumption is that the individuals are identical in all dimensions.
- Also experimenter demand effects, auditors know the purpose of the study
2) Correspondence studies
- Relies on ficticious applicants, researchers send resumes themselves
- The most common way is through names
Key advantages:
- The difference is solely observed differences
- Use of paper applications insulates from experimenter demand effects
- Low marginal cost
Key disadvantages:
- Crude outcome - call backs for interviews
- Results not representative of average minority
- Ads only represent one channel for job search
3) Implicit Association Test
- Computer-based test designed to identify implicit discrimination
Example in UCT (Corno)
- 500 freshmen living in double rooms
- Measure prejudice held against members of different racial groups
- Complete 7 blocks, score at the end
- Also refelects difference in reaction times
- Final score captured by a variable which takes higher values the lower the negative stereotype vs blocks relative to whites
