Lesson 4/5: Labor Markets Flashcards
Why do we study labor markets?
- labor is the main endowment of the world’s poorest people
- being able to exploit it fully is the only way out of poverty
- to engange in an occupation that is the best mtach for one’s talent, one must:
1) acquire skills that leverage talent
2) credibly signal these skills (to employers)
3) find the job that has the highest return to these skills
Describe the stats on youth employment and the challenges around it in the developing world.
- young populations (in SSA) imply large cohorts of labor market entrants
- youth unemployment is high and there is a clear gender gap
- gender gaps are not correlated with development
- many young workers are in unstable employment
- firms are small in lower-income countries -> there is a positive correlation between GDP pc and average firm size
What prevents firms from hiring young workers?
**Supply side: **
* Competition: large number of young people entering the market at the same time
* Credit constraints: hinder investments in skills
* Information frictions: limit ways to certify soft skills
**Demand side: **
* Credit constraints: hinder investment in workers’ human capital
* Poor contract enforcement: makes firms unwilling to risk hiring someone they do not already know
Give a short definition of (work) misallocation.
- misallocation arises when inputs of production are distributed across producers in an inefficient way
- an efficient allocation of inputs maximized aggregate output (marginal products of inputs are equalized)
- Possible sources of misallocation are factors tha tinterfere with the equalization fo marginal products: e. g. credit constraints and migration costs
Explain the set-up of the paper by Alfonsi et al. (2020) on tackling youth unemployment.
- Evidence from a labor market experiment in Uganda
- Two policies seeking to make young workers more appealing hires:
1) Vocational training: demand for skilled workers is strong and the lack of skills impedes young workers to fill vacancies
2) Apprenticeships and subsidies to firm training: externalities in human cpaital investment put a cap on firms’ willingness to train workers -> once the skills are embodied, other firms can benefit (subsidies compensate for that)
Model:
* Firm’s perspective: expected payoff of hiring a worker of skill is p(Y(s) - w(s)) + (1-p) (-w(s)) where p is the probability that the worker was dishonest (and runs away with the salary), assume Y’ - w’ >0 (the firm benefits from higher skills even if it pays a higher salary)
How do the two treatments in Alfonsi et al. (2020) affect the production curve of firms?
The firm hires if: pY(s) - w(s) >0
* for low p or low Y they do not hire
* low skill workers less likely to be hired
1) Vocational training (VT): increase slope & wage
2) Appreticeship + wage subsidies (FT): firm’s production curve shifts upwards, slope stays the same -> firms do not pay wages, so they cannot make a loss
How do VT and wage subsidies reduce employer screening costs?
- VT increases payoff from an honest worker and thus makes the firm more willing to bear the risk of hiring dishonest ones
- Wage subsidy eliminates (a large part of the risk)
Explain the two treatments in Alfonsi et al. (2020) who study the training of youth and their work sector opportunities. Where do they test their setting and on whom?
Treatment 1: VT
* 6 month, sector specific training
* Cost: 470 $ per trainee
* 30% theory and 70% practice content
Treatment 2: Firm Apprenticeship (FT)
* firm hire & train a worker on the job and receive 50$ for it ( 12$ for firm + 38$ salary)
* constractual insurance
- Workers: Uganda, disadvantaged youth (60% unemployed)
- Employers: urban labor market, small firms, 8 sectos, constraints to access of skilled labor or with inability to screen workers
What is the impact of VT and FT in Alfonsi et al. (2020)?
- 2-3 yeaers post intervention, workers have accumulated sector-specific skills
- substantially lower take-up in Ft (24%) than VT(68%) -> firm interest as key limiting factor
- both improve a labor market index that combines employment, total labor supply and earnings
->. but differences in dynamic treatment effect
VT: actually improved number of months worker and earnings long term, employment rates increase over time
FT: workers converge back to the control working hours and earnings, no long-term effect (offers a job immediately initially but not long-term)
-> reversal of fortune
Discuss the external validity of Alfonsi et al. (2020) on training and labor markets.
- Alfonsi: Uganda labor market, VT & FT, disadvantaged youth
Follow-Up studies: - Bassi & Nansamaba (2021): Uganda -> results of a verifiable test of noncognitive skills available to both workers and employers encourages hiring
- Carranza et al. (2022): assessing the skills of jobseekers and giving them their assessment results ina certificate they can credibly share with frims also increasing employment in South Africa
Explain how worker’s expectations respond to VT (+ matching)? (Bandiera et al. (2022))
** VT**
-> more optimistic job outlook -> behave accordingly -> search more intensively -> look for more stable jobs
VT + matching:
matching decreased expectation of job offer (revise down their expectations) -> search less intensively -> less picky with work (end up at worse firms & jobs)
* do significantly worse than without matching
Reason: they are imperfectly informed and so missattribute the lack of call back from match offers, causing them to revise down their expectations over their own prospects and search differently as they first transition into the labor market
MATCHING -> DISCOURAGEMENT
What different types of policies have we discussed to overcome labor market imperfections?
- VT (with and without firm matching)
- subsidizes firm training (apprenticeships)
- certifications
- credit constraints
Explain the motivation behind the study on selection of talent by Abebe et al. (2021). How do they go about answering it?
ETHIOPIA
Research question: Do labor market frictions make it hard for firms to make good hires? Can firms overcoe these frictions by offering financial incentives to make job applications?
Strategy:
1) RCT: study how the quality of applicants for a clerical job in urban Ethiopia changes in response to two different types of financial incentives
2) (not discussed) Structural model: they use the experimental variation to quantify application costs and their correlation with worker ability
Describe the hiring problems in the Ethiopian labor market.
Firm Perspective
*Inadequate skills
* Turnover
* Poor conduct
* Low motivation
* Absenteism
Worker Perspective
* job seekers only apply for a fraction of the jobs (job board)
* formal jobs requre an application in person (time consuming, usage of public transport)
What puzzle about application cost and ability do Abebe et al. find? Why?
Application costs also appear positively correlated with ability in the data.
- high ability workers with low application costs, find work faster and leave job search faster
- they become progressively underrepressented in the pool of job seekers
Explain the treatments used in Experiment 1 on labor market inefficiencies on hiring quality in Abebe et al.. Also explain, how workers were recruited and their ability was tested.
1) Application costs are reduced with 100 ETB (4.5$) incentive (half of estimated application cost)
2) The wage doubles from 1600 ETB to 3200 ETB (25th to 75th percentile)
Recruitment:
* position advertisment on job boards & newspapers
* random allocation of callers (based on stratification)
* to apply for the job, people have to visit the application center and take ability tests (cognitive ability, non- cognitive ability and workexperience & GPA)
What are the key findings to Abebe et al. (2022) on workers ability and labor market constraints?
Application incentive: **
* improves the quality of the applicant pool (higher ability & GPA)
* increases applications by 28%
* stronger effect for disadvantaged groups (clear heterogeneity of impact) -> taps into pool of low-income job seekers who stand to benefit most from the job -> higher allocation efficiency
** Wage increase:
* increases applicant quality
* increases application by 45%
* no (significant) heterogeneity of impact
How do Abebe et al. (2022) test the robustness of their findings on workers ability & labor market inefficiencies.
Concerns:
* applicants have different preferences
* changes in test effort
* changes in beliefs about labour market prospects, position or of salience of the application decision
-> none of them is found to be particularly salient in the data
Summarize the findings by Abebe et al.
**The incentive improves the quality of the applicant pool. **
* cognitive ability is higher at mean & top of distribution
* size of the effect is similar to doubling the wages
* the effect is driven by workers with lower incomes
High quality woereks face large application costs.
* costs are 13% of wage, 30% of jobseekers are credit constrained
* costs are positively correalted with ability
* panel evidence sugegsts correlation is the result of selection
**Results point to a group of high-cost, high-talent job seekers that can generate high value for firms **
–> misallocation of talent!
