L4: trust Flashcards
Why is trust important?
Often key in reducing transaction costs since many day-to-day activities require it (eg. ebay)
Explain Berg et al.’s (1995) investment game?
Participants paired up, one receiver and one sender, both have fixed endowment of £10
S can send R any amount, y, between 0 and 10, and experimenter triples it to 3y
R can then return any amount they want
Explain how Berg et al.’s (1995) investment game measures trust?
Amount that S sends to R signals trust levels, amount returned signals trustworthiness
Stylised findings of Berg et al.’s (1995) investment game?
1) Senders typically send £5 (half the maximum)
2) Receivers typically send back slightly less than y therefore trust doesn’t pay
Explain a treatment imposed in Berg et al.’s (1995) investment game?
Social history vs no history
Those in the social history are given a report summarising the decisions of the ‘no history’ treatment
How is the social history treatment expected to influence the results? Berg et al.’s (1995) investment game
May lead to senders being more aware of non-reciprocating and therefore may decrease investment.
How is social history actually found to affect the results in Berg et al.’s (1995) investment game?
Found participants actually exhibit HIGHER levels of trust and HIGHER levels of reciprocity than ‘no history’ treatment (see notes)
What did Ortmann et al. (2000) investigate?
How robust the findings of Berg et al (1995) are
How did Ortmann et al. (2000) test robsutness?
Used same experiment as Berg et al. (1995) but added 3 treatments:
1) Social history II: in addition to a summary table, also given a graph showing specific values
2) Questionnaire: to prompt strategic reasoning, asked questions like ‘how much would YOU return?’ etc
3) Above two treatments combined
What were the findings of Ortmann et al. (2000)?
Neither the way the info was presented nor the strategic prompts mattered statistically tf Berg et al.’s results are robust!
Therefore: does this behaviour represent willingness to trust strangers or a desire to share?
What did Gneezy (2000) try do?
Tried to understand behaviour of senders by systematically varying the amount the responder could return
See
treatment for Gneezy (in notes) (important!)
What were Gneezy’s (2000) hypotheses?
If senders are motivated by the desire to share, then the amount the receiver CAN repay should have no effect on the amount that can be sent
BUT if senders are motivated by expected reciprocation, would expect the senders to send more when higher repayments are possible
Findings of Gneezy (2000)?
Avg. amount sent when receiver can repay $2 is $2, and when receiver can pay up to $10, avg. amount sent is $6.50, and when receiver can pay up to $10, avg. amount sent is $5.63 (tf not stat. different)
TF expectations affect trusting decisions (see notes don’t get this one yet!)
