WT Week 4 Flashcards
Motivation
Why Do People Stay Poor (Balboni et al., 2022)
“To distinguish between two seperate views on poverty:
1) Everyone has equal opportunities but there are certain traits which make people unsuitable for certain jobs
2) Poor face different opportunities and take low-earnings jobs because they are poor (poverty trap)”
Special settings
Why Do People Stay Poor (Balboni et al., 2022)
Both explainations have the same outcome in equilibrium so need to observe the behaviour of those who pass the threshold. If poverty is due to differences in traits, they will return to where they started. If due to opportunities, they will be elevated out of poverty forever. Rarely see anyone cross threshold, or know where the threshold is anyway.
Theory
Why Do People Stay Poor (Balboni et al., 2022)
Half of 6,000 households randomly selected in 2007 to receive large asset transfer (mainly in the form of cows). Track the long-run dynamics of assets, occupations, and poverty across 11 years in order to understand whether the one-time policy does indeed have permanent effects in the form of lifting people from the trap. Very similar occupational structure of villages which are highly correlated with asset ownership. Use this relationship to understand the asset threshold levels which lead to asset-reliant occupations and causes the rise out of poverty.
Empirical design
Why Do People Stay Poor (Balboni et al., 2022)
“Asset threshold determined as required amount in order to
Construct structural model of occupational choice allowing quantification of occupational misallocation, benchmarking general equilibrium effects and simulate policy counterfactuals.
identification exploits differences in asset ownership before the transfer, like the cart in the example above, which are small relative to the size of the transfer.
- Exploit the differences in the asset holdings between beneficiaries at baseline (eg some would have enough assets to buy the equipment to benefit from cows and others would not)
“
Data
Why Do People Stay Poor (Balboni et al., 2022)
Individual-level panel data gathered over 11 years studying effect of large randomised asset transfer programme in rural Bangladesh. Larger survey covering 23,000 households across wealth distribution in 1,309 villages, situated in the poorest districts of Bangladesh. Track 6,000 poor households across 2007, 2009, 2011, 2014 and 2018.
Key findings
Why Do People Stay Poor (Balboni et al., 2022)
“Average poor household is trapped in poverty.
In absence of credit constraints, 2% of households would be best off specialising in wage labour. 98% exclusively reliant on such work at baseline.
Only 1% specialise in livestock rearing whereas 90% would do if they had same access to asset wealth as middle and upper classes. Implies 97% of households misallocate labour at baseline. Suggests nobody”
Interpretation / policy implications
Why Do People Stay Poor (Balboni et al., 2022)
“Suggests nobody is innately unable to take up a better occupation. As such, the cost of misallocation is significantly larger than the one-off policty cost of taking households across poverty threshold.
Poor are excluded from occupations due to lack of assets and labour/talent is wasted on less productive means and more irregular occupations. Low wage and unreliable nature of the jobs prevents saving enough to fund the purchase of assets needed to run these businesses.
Poor remain poor due to lack of access to better jobs. Creates a significant misallocation of talent. Therefore proving existence of poverty traps.”
Motivation
Underinvestment in a Profitable Technology: The Case of Seasonal Migration in Bangladesh (Bryan et al., 2014)
To understand how temporary migration to nearby urban areas during hunger seasons that offer better employment opportnities provides economic returns and why people were not already participating in it.
Special settings
Underinvestment in a Profitable Technology: The Case of Seasonal Migration in Bangladesh (Bryan et al., 2014)
Scarce work opportunities between harvest seasons which leads to hunger period.
Theory
Underinvestment in a Profitable Technology: The Case of Seasonal Migration in Bangladesh (Bryan et al., 2014)
Randomly assign cash or credit incentive of $8.50 which covers round-trip conditional on a household member migrating during 2008 monga season. Exploring why people who were induced to migrate by the programme were not already migrating despite high economic returns, with risk aversion, credit constraints, and savings as reasons.
Empirical design
Underinvestment in a Profitable Technology: The Case of Seasonal Migration in Bangladesh (Bryan et al., 2014)
“Randomly selected 100 villages across 2 districts and conducted village census starting in 2008.
Randomly selected 19 households in each vilage from those that owneed less than 50 decimals of land and where a member of household was forced to miss meals during prior monga season
Randomly allocated the 100 villages into four groups (cash, credit, information, and control)
Treatments applied and 630 households across 37 randomly selected village were offered $8.50 for one person to migrate; other group offered in form of zero-interest loan”
Data
Underinvestment in a Profitable Technology: The Case of Seasonal Migration in Bangladesh (Bryan et al., 2014)
“Census data before trial and follow-up data after the monga season. Gathered data on income, assets, credit, and savings.
Same questions asked in follow-up but also asked about migration experiences over past 4 months.
Conducted follow-up survey in the year following and after following monga season.”
Key findings
Underinvestment in a Profitable Technology: The Case of Seasonal Migration in Bangladesh (Bryan et al., 2014)
“1) Migration in this setting is very profitable and underutilised
2) Low utilisation explained by high levels of risk
3) Do not fully understand the migration choices of households, lack information to understand
Migration induced by intervention increases food and non-food expenditures of migrants’ family members remaining at the origin by 30-35%.
Improves caloric intake by 550-700 calories per person per day.
Households in treatment areas continue to migrate at a higher rate in subsequent seasons, even after the incentive is removed; migration rate is 10% higher in treatment areas a year later; figure drops to 8% 3 years later.
Rational households choose not to migrate due to fear of risk and uncertainty of prospects at the destination. Fear the cost of moving and failing, which is especially concerning for a family already in hunger.
Insuring the first trip against devistating outcome leads to long-term benefits as they learn of the returns and how to carry out migration.
Found that they cannot learn of the returns on migration from othres, which might explain frictions in what keeps agricultural workers in rural areas rather than urban areas where productivity is persistently higher.
Households that are close to subsistence start with the lower migration rates but are most responsive to the intervention.
Households induced to programme are less likely to have pre-existing network connections at destination and learn about migration opportunities and destinations in future choice as whether to re-migrate and to where.
Found that migration more responsive to incentives (i.e. credit conditional on migration) than to unconditional credit.
Where saving to migrate is possible, they do so rapidly - however, doesn’t change migrations levels, showing how high the risk aversity is.”
Interpretation / policy implications
Underinvestment in a Profitable Technology: The Case of Seasonal Migration in Bangladesh (Bryan et al., 2014)
“Large effects on migration rates, consumption, and re-migration suggest policy of encouraging migration could be beneficial.
However, given that people weren’t already migrating suggests to other issues which may instead require different intervention which would have the same effect.
““Poverty as vulnerability”” - poor cannot take advantage of profitable opportunities because they are vulnerable and afraid of losses.”
Motivation
Bandiera 2017
investigates how enabling the poorest women in village economies (Bangladesh) to engage in the same labor activities as their wealthier counterparts can set them on a sustainable path out of poverty.
Theory
Bandiera 2017
The key question we examine is whether enabling the poorest women to take on the same work activities as the better-off women in their villages can set them on a sustainable path out of poverty. Intuitively, if the poor face barriers to entering high-return work activities and this is what keeps them in poverty, we expect program beneficiaries to change their labor allocation and escape poverty once such barriers are removed.
Special settings
Bandiera 2017
study contributes to the broader academic and policy debate on poverty traps and the mechanisms that can help individuals escape them.
Empirical design
Bandiera 2017
“RCT - Targeting the Ultra-Poor (TUP) program. Income data collected from 21,000 households in 1,309 villages. Surveyed four times over a seven-year period. Eligible women received asset transfers (livestock and skills) valued at around $140. Provides a source of continuous income through livestock rearing, milk production, and eventual offspring.
-DiD strategy to compare changes in outcomes over time between the treatment and control groups.”
Key findings
Bandiera 2017
“ultra-poor women devoted 217% more hours to livestock rearing compared to their counterparts in control villages. 21% rise in total earnings. These gains do not crowd out the livestock businesses of noneligible households
- 17% reduction in hours spent on agricultural labor, leading to higher wages from lower labour supply.
- likelihood of being below the $1.25 per day extreme poverty line was reduced by 14%”
Interpretation / policy implications
Bandiera 2017
“important to combine asset transfers with relevant training and support
- The study highlights the efficacy of specifically targeting the ultra-poor
- The success of the TUP program in Bangladesh indicates that similar programs could be scaled and adapted to other contexts.”
Data
Bandiera 2017
Our survey gathers detailed data on hours worked, days worked, and earnings for each labor activity of each household member. The program, evaluated through a large-scale randomized control trial covering over 21,000 households in 1,309 villages surveyed four times over a seven-year period, involves a one-off transfer of assets and skills to the poorest women, enabling poor women to start engaging in livestock rearing.
