7. field experiments

what are field experiments

tests economic theories and models in real world setting rather then controlled lab settings by manipulating real world conditions and observing how peoples behaviour changes

why are lab experiments criticised

• environment

• context

• subject pool

• time

continuum of experiments from controlled to naturally occuring

LAB AFE FFE NFE NE

characteristics of LAB

high control over variables

reproducable

used to validate theoretical models

known subject participants

characteristics of AFE

lab experiment

known public subjects

ffe

field

known public subjects

nfe

unknown public subjects

real world setting and context

ne

no researcher control over assignment

naturally occuring event

deal or no deal blatavaskky and pogenbena 2010

contestants randomly assigned a sealed box with money from £0.01-250000

focuses on exchanges offers in order to test loss aversion

individuals should be exactly indifferent between keeping their box and changing their box

but individuals derive utilty from changes in wealth rather the absolute wealth levels, relative to a reference point

if she keeps her own box she keeps a utility of v(0)=0 as her asset position remains unchanged

contestant derives a strict negative utility from exchanging her box for one of the remianing sealed boxes due to loss aversion so PT predicts her not to exchange her box

eut prediction is to be indifferent between keeping and changing

first offer is a one shot loss aversion

second offer is loss aversion for multiple exchanges

• first offer results, the contestants are indifferent, in all countries between rejecting and accepting the exchange leading to no loss aversions, perhaps explained by EUT

• in the second offer, more reject the offer to accept and keep their box

contestants dont appear to be loss averse when dealing with lotteries involving large stakes