Wednesday, 18 November 2009

Endowment Effect

I find it quite reasonable that human beings tend to put value to things that they own and numerous reasons can add to this shift of value, however when it comes in trading , endowment effect creates many inconsistencies which result in under trade due to willingness to accept and willingness to pay a certain item. Entitlements do affect value and this has been portrayed in the study of Daniel Kahneman and his colleagues titled “ Experimental Test of the Endowment Effect and the Coase Theorem” . The aim of this study was to examine an alternative explanation for buying and selling discrepancy. Series of experiments were conducted, involving real exchange of physical goods and tokens. Random allocation of buyers and sellers was assigned.
Endowment effect is a manifestation of loss aversion therefore the discrepancies between WTP and WTA occur. In order to test this effect in situation where no utilization is expected in return, several experiments have been conducted where different controls were put in induced markets ( were there was no money involved, instead tokens were given to participants) and real goods markets. The main aim of these two different markets was to isolate factors such as loss aversion since in token markets it would be irrelevant since transactions would be evaluated simply on the basis of gain or loss. Furthermore, by using these two different markets the experimenter would be able to isolate factors such as transactions costs, bargaining habits, lack of knowledge etc which would affect the discrepancies between buyers and sellers.. Before continuing with the results I must point out that 8 experiments were carried out in order to test for misinterpretations, identify the reluctance to sell or buy as well as bilateral bargaining which is contradictory explained by Coase theorem.
In all experiments similar results were obtained; in induced value markets exchange trades were done according to expectations however in goods markets usually under trading occurred due to endowment effect. Median selling price was usually over twice to that of buying price. Even in experiments when a random trading price was chosen by the experiment (with the intention to eliminate the incentive to mistake true value in order to influence price) similar results were obtained.
Even in markets when there was barter method, ( when two goods were exchanged instead of money and goods)which isolates the influence of endowment effect by ruling out income effects again the occurrence of undertrading was not eliminated.
Its interesting to point out that this study suggests an” instant” endowment occurrence. That is because participants appeared to increase the value of the goods as soon they were given to them, suggesting an instantaneous shift in reference point and value given to objects. Contrary to the assumption that preferences are independent of entitlements, the evidence of this study shows that people’s preferences depend on their reference positions, suggesting that endowment effect and loss aversion are characteristics of preferences and they can’t be eliminated by experience, training or market discipline.

Measuring Utility







































The aim of this exercise was to show how much I am willing to gamble based on the amount of money that I would receive and the probability of actually getting that money.
The first graph shows the Certainty equivalence whereas the second graph shows Probability equivalence. In both graphs horizontal axis represents monetary values and the vertical axis represents utility values.
In both graphs there is a similar pattern, which shows that I am not a big gambler. I would prefer a certain amount of money rather than 50% chance of winning 1000 or 0. Furthermore, I would enter a lottery only when I would perceive a high probability of winning 1000, and only then I would be indifferent of the opportunity to have certain amount of money.

Thursday, 29 October 2009

week 2/3

Due to strikes in our University we didn’t have lecture on week 2... I was actually looking forward for this lecture as I wasn’t aware of which journal I was supposed to read ( as I mentioned in my other post I missed first lecture) ... I managed to download Dhami’s report regarding Psychological Models of Professional Decision Making ( fortunately some of the students who missed first lecture have read the same report and we managed to form a group) . The main aim of the report was to test out whether judges use simple heuristics (matching heuristics /fast and frugal heuristics) or a more complex model (Franklins rule) on their decision making. One would suppose that judges take into consideration as many cues as they can prior to their decision making...or at least that is what its expected from a judicial decision, that the information regarding a case is thoroughly taken into account, weighted and combined so a fair and accurate decision can be drawn. Unfortunately that is not always the case, judges tend to base their decision on one or few cues due to time constrains maybe as well as the reliance on the decisions made by police and previous benches. Coming back to the report, the study conducted by Dhami suggests that judges don’t follow the ideals of due process therefore matching heuristics better predicted judge’s decision making... they were maybe intentionally or unintentionally relying on the decision made by other bodies as well as due to consequent rapidity with which decisions must have been made. Finding of these studies should be considered as some actions should be taken into account regarding the validity and reliability of the decisions made by judges... I was actually quite surprise to know that judges tend to be “ shallow” when weighting out cues regarding the cases that they have to make a decision, and then I recalled different cases of how individuals have been given an unfair decision. It is quite sad that these inconsistencies still occur and that the judicial body are not able to manage their workload and offer a service that is expected from them... it would be essential that e more close control is given to the judicial system so that the number of unfair decisions can be at least reduced and maybe a new model of cue weighting be developed so it saves time and increase accuracy for the final decision making of judges.

week 1

Unfortunately I missed the first lecture as I had to go back home due to personal reasons (my home town in Kosovo). However I have read chapters 1 and 2 of the JDMPP book and it gave me a general insight as to what judgment and decision making constitutes and I particularly enjoyed reading the first chapter that focused on what rationality is .Regarding the second chapter, I was particularly interested in the social judgment theory as it takes into consideration multiple factors and their importance in terms of relevance regarding an outcome of some event. I found it very interesting how math language can be used to predict the outcomes of events when of course some of the cues are known as well as their value. Another point worth mentioning was the reason as to why humans are less accurate in making judgments as opposed to the proposed statistical models. Due to the fact that humans tend to be more subjective when weighting out different factors and information regarding their values and outcomes, many inconsistencies can arise. Situational factors affect the overall outcomes of event as well due to the lack of individuals to control them as well as maybe the inability to realize how situational factors affect their behaviour. This theory in overall doesn’t explain much the cognitive processes as to why and how people reach a judgment, however other model known as the Probabilistic mental model provides an explanation as to why individuals might be inaccurate in their judgments. It suggests that individuals tend to base their judgments taking into consideration one cue (as opposed to the regression analysis approach, where multiple cues are taken into account) meaning that they use shortcuts in order to produce an outcome. The validity of such judgments can be questionable as the basis used for such judgments are not strong enough.
I have to admit that this module is a bit different from others; however that is what makes it more interesting I guess... I find the content of the book very interesting and I already started reading other chapters, I jumped in the 15 chapter as the title sound very captivating.