inspection game and sharecropping

a paper by Rimawan Pradiptyo about incentive mechanism of punishment. can be found here is an interesting piece by this UGM prof.

this inspired me to work with farmers-casual workers in farming using an inspection game.

Health Seeking behavior

I have been thinking about the possibility of doing a health seeking behavior experiment. Mainly inspired by an interesting powerpoint presented by Ben Olken from JPAL MIT Indonesia at TNP2K around august 2011. He was showing the result of a follow up research to health seeking behavior. Basically he showed that the intervention works very well at the intial year follow up, however further exploration after one year indicate a delining magnitude in treatment effects.

I think some behavioral reasonings can be useful to understand such anomalies. The study show that both demand and supply of health service has been declining several years after the interventions. To my view this to some extent is related to the interactions between medical practitioners-patient. Behavioral economics has been showing the emergence of cooperative behavior, non-self interest inequality aversions, people hate to be treated unequally. Fairness notion then is considerably important to the decision making of homo economicus.

Thaler (1981) shows that Fairness put a constraint/limitation to self interest maximizing utility behavior, as a doctor/ practitioners hold their holy oath to help the poor in need and the sick and vulnerable. This put them into higher pressure if they are in an environment with constant poverty and bad environment condition. Market based intervention then will be useless as people are still contrained to independent factor that is not related to the market.

Doctor-patient relationship could also offer interesting explanation, people go to the doctor often resort for advice and suggestions, for uncertain risk and unknown desease patient will rely to the Doctor to make the final decision, Due to Regret aversions (responsibility cost of decision making), in order to decide a cancer treatements, or surgery people will avoid to make their own choices . The doctor on the other hand had to have an instant list of choices of a practical advice in the middle of complex situation and stressful time pressures. If the Doctor is sufficiently risk averse they will prefer to avoid risky patients

Declining demand of Patient on health service also probably due to the fact that seeking information about doctor advice is costly, information affect beliefs and in turn mental states that diminishes utility. The gesture of seeking information will lead to either (1) bad news, creating anxiety and loss of utility (we used to have the same feeling when we are about to go the lab to take a medical check up). Or (2) more information to enable doctor and patient for an optimal treatment strategy. If patient is risk averse and discounted their health benefit in the future very high, Sick people will not come to hospital/medical practitioner on time.

Bridge to Cross : mental accounting?

I talked to my wife today and I realized that I have not seen any kind of crossing bridge ( a bridge used to cross the street) like we had in Indonesia any where in Europe or Africa. Despite my lack of coverage on all countries it is intriguing for me. My wife pointed out that most of the crossing facilities in Europe are more likely to be underground.

Despite of better quality crossing bridge is now well spread across Jakarta, people seems to be (still) reluctant to climb up the ladder just to cross. While underground tunnels and metro station are always packed with people when they cross the street or railways in Developed countries.

I am aware that the culture difference and income difference can well explain the level of dicipline in crossing the street, but May be just because people prefer to steping down than climbing up.

if you have to choose to step down first and then climbing up later VS climbing up now, and stepping down later, which one will you prefer? Laibson et.al says that people discount the future very high, you prefer to eat chocolate today, but choose apple for tomorrow. You prefer to watch TV now, and postpone exercise for next week.

it is interesting to see that even crossing the street involves risk preference and heavy discounting time preference

trust and law economics

A new Post by Ben Ho at ESA mailing list. need to check this out

Oh, and for those who are interested, my interest in all this comes from Ho (Mgmt Sci 2012) and Ho and Liu (JRU 2011) which looks at how Apologies restore Trust and how apologies interact with medical liability payments.

Falk and Kosfeld (AER 2006) - Trust Game, where Trustor can choose to trust or control (choose a minimum effort level) suggests monitoring or minimum performance requirements are counter productive – justifies incomplete contracts. Ala Bernheim Whinston (1998) – contracts should have stratgegic ambiguity Allen and Gale (1992) and Spier (1992) - complete contracts lead to bad inferences on the first party’s type

Trust and the Law as Substitutes

Negative Effects of Monitoring Theory: Frey (1993) Sliwka (2003), Ichino and Muhlheusser (2004)

Reciprocity and Trust (Ernst Fehr and Simon GÄachter, 2002;Fehr and Bettina Rockenbach, 2003; Fehr and John A. List, 2004; Fehr et al., 2004)

Effect of Incentives: new information regarding the importance or the cost of the task (Gneezy and Rustichini, 2000a; Roland J.M. Benabou and Jean Tirole, 2003), because they insult the agent (Gneezy and Rustichini, 2000b), or because they are in con°ict with social norms of fairness and cooperation (Fehr and GÄachter, 2002; Fehr and Rockenbach, 2003; Fehr and List, 2004; see also the discussion in Gneezy, 2004).

Falk and Kosfeld (2006) considers two mechanisms: Strategy Space or Strategic Response. Finds evidence that it is a strategic response. Also finds evidence that beliefs about principle expectations and emotional response matters.

Charness et al. (forthcoming AER) The Hidden Advantage of Delegation: Pareto-improvements in a Gift-exchange Game. Allow workers to set their own wages. The focus is instead of just encouraging incomplete contracts, we should encourage delegation of the decision of contract terms.

Leider and Kessler (Man Sci 2012) – similar results… theory and experiments on the value of unenforceable “handshake” contracts over mandatory minimum rules

Fischer and Huddart (AER 2008) Optimal Contracting and Endogenous Social Norms – Theory - multi action P-A model where a Norm cost is added to the cost of effort.

Akai and Netzer (JSE 2012) - trust game between Japanese and Austrian subject pools

Bohnet Frey huck (APSR 2001) – finds in trust game that monitoring crowds out intrinsic incentives so moderate levels of monitoring/enforcement optimal. Monitoring is an exogenous probability that untrustworthy behavior is punished

Bohnet et Al (QJE 2010) – gulf states use relation based trust, whereas the west uses rule based trust using reference points and prospect theory. (a decrease in the likelihood of betrayal vs a decrease in the cost of betrayal). Discusses the role of different institutions, but operationalized entirely through differences in reference point (Kozcegi Rabin style)

Gachter et al (2011) “Incentives and Voluntary Cooperation” http://www.nottingham.ac.uk/cedex/documents/papers/2011-06.pdfimplicit incentives only work under non-incentive contracts in the gift exchange game. Essentially compares salary (trust game) to piece rate (gift exchange with effort contingent payment framed as either a fine or a bonus). And sees how Order of play matters. Primarily finds crowding out.

Villeval "Does Monitoring Decrease Work Effort? The Complementarity between Agency and Crowding-Out Theories" Journal of Games and Economic Behavior 2008 and "Respect as an Incentive", IZA 2010: modified trust/gift-exchnage game where workers can burn money as costly signal

Galbiati-Schlag-van der Weele “Sanctions that Signal” - sees the effect of choice to sanction using the coordination / minimal effort game (goeree and holt, 2001, 2005). Sanctions are a full substitute.

Most of the above focuses on Trust and Law as Substitutes, but also there were other studies that focused on Trust and the Law as complements

"Achieving Compliance when Legal Sanctions are Non-Deterrent", by Tyran, J.-R and Feld, L. Scandinavian Journal of Economics 2006, 108(1): 135-56. Showing that non-deterrent legal sanctions have little effect on cooperation when imposed but induce more cooperation if accepted in a vote. Deterrent legal sanctions work fine either way.

"Self-Organization for Collective Action: An Experimental Study of Voting on Formal, Informal, and No Sanction Regimes", by Markussen, T. Putterman, L., and Tyran, J.-R., Vienna Economics Papers, Working paper 1103 (paper) (supplementary online materials) (revise and resubmit Review of Economic Studies) Showing that deterrent legal sanctions are popular when they come at a low fixed cost, but not more popular than informal (peer-to-peer) sanctions when they come at a high cost.

"Public Goods and Voting on Formal Sanction Schemes", by Putterman, L., Tyran, J.-R. and Kamei, K. Journal of Public Economics 2011, 95(10): 1213-22. Showing that when voters can design legal sanction sanctions they overwhelmingly make socially optimal choices (i.e. they choose they choose deterrent rather than non-deterrent legal sanctions for free riding. As in the paper by Tyran and Feld mentioned above, we find that deterrent sanctions are effective, and the effect of non-deterrent legal sanctions depends on whether they are accepted in a vote)

"Microfoundations of Social Capital", by Thöni, C., Tyran, J.-R. and Wengström, E. (supplementary online materials) (revised version) (revise and resubmit Journal of Public Economics) Showing that the commonly used measure of “survey trust” (in the WVS) correlates well with experimental cooperation choices in a heterogeneous population.

Kimbrough and Vostroknutov (working paper 2012) – arbitrary costly rules (Stop Sign) used to screen for conditional cooperators . then they play a trust game.

Farina, O’Higgins, Sbriglia (RIE 2009) – comparing attitudinal trust and trust game behavior, using dictator game as a control. “I am sending you a recent paper to which I contributed. It stresses that in trusting and being trustworthy the context matters, along with types. I think that there is an indirect link with the "design" of relational contracts. My personal opinion is that trust and rules need to be complements, insofar as rules have no chance to be respected if low trustworthiness makes to follow the rule too costly.”

Other

Schneidler and Vadovic “legitimacy of Control” JEMS (2011) – like Falk akd Kosfeld, but varies initial endowment to vary the framing of property rights. Argues that the legitimacy of property rights matter.

Rizzolli, Matteo and Stanca, Luca. 2012. "Judicial errors and deterrence: theory and experimental evidence," and it is forthcoming on the Journal of Law and Economics. In this paper we use a reverse-dictator as baseline game. – Shows theoretically – using standard expected utility and thru experiments that Type I errors more important to deterrence than type II for judicial enforcement of crime based on a dictator game.

The second paper is Marchegiani, Lucia, Reggiani, Tommaso, and Rizzolli, Matteo. 2011. "How Unjust! An Experimental Investigation of Supervisors' Evaluation Errors and Agents' Incentives," IZA Discussion Papers No. 6254 – Real Effort Task with exogenous Type I and Type II punishments. Similar to the Legitimacy of Control (Schneidler and Vadovic, JEMS 2011) that legitimacy matters, here in terms of the justness of punishment rather than on propery rights

Similarly, Al-Ubaydli, houser, Nye, Paganelli, Pan(JSE 2012) market participation on trust. Priming thoughts about markets increases trust partly thru changing beliefs about trustworthiness.

Al-Ubaydli, Gneezy, Lee, List (JDM 2010) –Large impersonal labor pool and unemployment leads to equal rewards and punishments in essentially a trust type labor game. Driven by strategic concerns not affect. reputational concerns leads to more rewards.

Bolston, Greiner, Ockenfels (forthcoming) – studies feedback in ebay, with experiments that engineer to maximize information of feedback given reciprocity concerns. Each simulated ebay auction is like a trust game.

Dickhaut, Lunawat, Xin, WAymire (2011 working paper) communication and cheap talk in a trust game.

Also Corruption: Lambsdorff and Frank (IRLE 2011)- Results from a extension of the Trust Game where Trust is corruption and can be punished. Finds that men are more likely to reward/reciprocate corruption.

Rule Formation and economic history Journal of Law, Economics, and Organization entitled, "The Ecological and Civil Mainsprings of Property: An Experimental Economic History of Whalers’ Rules of Capture" discusses rule formation of property rights and social cohesion in an experimental game.

Rough Takeaways

• Important to give the principal ability to signal / screen their social type

• Important to give the agent ability to signal / screen their social type

• Incentives Generally a Substitute for Trust

• Incentives can be a complement if they establish norms/legitimacy

• Contrast optimal contracts with Relational Contracting Approach – where all is each infinitely repeated game.

• Economics Features

o Cost of mistrust (Bohnet QJE 2011)

o Expectation of trust (Bohnet QJE 2011)

o Depends on risk intolerance, as measured by risky dictator (Bohnet and Zeckhauser)

? "Is trust a risky decision?", Journal of Economic Behavior & Organization 2004. (Eckel and Wilson)

o Reputation matters

? Feedback Ockenfels Greiner and Bolton

? Also Bohnet and Frey on arab vs US

o Sanctions may lead to crowd out (Gachter)

o But not if designed by participants (Tyran)

• Psych features

o Reference points (Bohnet et al QJE 201?)

o Seprates betrayal intolerance (Bohen and Zeckhauser 2004,2008)

o Beliefs about what is expected (guilt aversion, charness dufwenberg)

o Legitimacy of payoffs

? legitimacy of control – property rights

? inJustice of type ii errors

? Voting and the creation of self-organization

o Affect might matter – but Gneezy and List and al say it does not.

o Compliance with social norms?

• Other contracting features

o Delegation

o Control i.e. restrict choice domain

o Monitoring probability

o Justice of monitoring

o Property Rights

o Legitimacy