Demand For Insurance Products Flashcards
Karlan added to his experiment to understand the demand for insurance.
How did he do this?
Cash grant group offered insurance between 1/8 of fair price and market price.
This creates a demand curve at different prices.
Benefits of this research design (4)
(1 cash, 2 insurance, 1 adoption)
Cash grants vs control allow us to see if capital constraints important (we found small effects tho, so not a huge issue!)
Free insurance allows to observe full population of farmers, not just those who are willing/able to buy.
The then random pricing allows us to create a demand curve.
Adoption takes time: first show benefits, then charge market price
Results of the demand curve (2)
Demand at the actuarially fair price is low.
and when actually purchased, the fraction of farm covered by insurance is even lower.
So demand is low, and inversely related to price, as expected.
What concept do farmers understand that explains this?
Basis risk of index insurance -
Aggregate loss, and the individual loss is imperfectly correlated, worsening the WORST CASE OUTCOME.
Basically the farmer with the most weather prone land knows that they may not enough insurance. (A extremely risky adverse person would not buy since not worth it)
So farmers accounting for basis risk is one reason for low demand:
Other difficulties in the market for INDIVIDUAL insurance products (4)
Adverse selection
Moral hazard
Auditing costs - for individual persons is costly.
Lack of trust - farmers don’t fully believe they will receive payouts when rainfall events occur, hence demand low.
Karlan’s irony of results
Demand increases when accident occurs and receive payouts, but demand is lower if insurance is bought but no accident so no payout.
Irony: insurance offers its largest benefit for low-probability high loss events, and since are low probability, people become less likely to demand the insurance overtime
Cole et al
(Hint: trust)
Found when insurance is endorsed by a well known NGO, it increases demand.
Banerjee (same guy as Spandena) - what did they find?
Assistance in online registration for health insurance increases demand for insurance (3.5% - same amount as nguyen on teacher-parent meetings on attendance)
EVAL: also let to many people registering unsuccessfully (tech challenges) proving state capacity is poor in developing countries.
Other challenges with insurance demand
Social networks - tell friends and increase demand
Financial literacy - if can’t understand then wont demand.
When does informal insurance work well (2), not well
Work well for idiosyncratic risk
Complements insurance products that present basis risk. (I.e have the rainfell index insurance but also ROSCAs correct the imperfect correl between aggregate loss and individual loss)
Not well for aggregate risk (ROSCAs pot only goes to one person, so can’t help in a problem that affects everyone like a flood).
