|
|
Essays on Endogenous Information Acquisition in Economics
Ilinov, Pavel ; Steiner, Jakub (advisor) ; Kováč, Eugen (referee) ; Michelucci, Fabio (referee)
English Abstracts Pavel Ilinov 1 First Chapter In the first chapter we consider two types of models: (i) a rational inattention problem and (ii) a conformity game, in which fully informed players find it costly to deviate from average behavior. We show that these problems are equivalent to each other, both from the perspective of the participant and the outside observer: each individual faces identical trade-offs in both situations, and an observer would not be able to distinguish the two models from the choice data they generate. We also establish when individual behavior in the conformity game maximizes welfare. 2 Second Chapter The second chapter shows that the principal can strictly benefit from delegating a decision to an agent whose opinion differs from that of the principal. We consider a "delegated expertise" problem in which the agent has an advantage in information acquisition relative to the principal, rather than having preexisting private information. When the principal is ex ante predisposed towards some action, it is optimal for her to hire an agent who is predisposed towards the same action, but to a lesser extent, since such an agent would acquire more information, which outweighs the bias stemming from misalignment. We show that belief misalignment between an agent and a principal is a...
|
|
|
Essays on Information Economics
Matysková, Ludmila ; Steiner, Jakub (advisor) ; Kamenica, Emir (referee) ; Dean, Mark (referee)
In the first chapter, we study the effect on the economy of platforms for online consumer reviews. Consumer reviews may have perverse effects, including delays in the adoption of new products of unknown quality when consumers are boundedly rational. When consumers fail to take into account that past reviewers self-select into purchases, a monopolist may manipulate the posterior beliefs of consumers who observe the reviews, because the product price determines the self-selection bias. The monopolist will charge a relatively high price because the positive selection of the early adopters increases the quality reported in the reviews. In the second chapter, we study a game between a sender and a receiver in a framework of Bayesian persuasion. A sender choosing a signal to be disclosed to a receiver can often influence the receiver's actions. Is persuasion more difficult when the receiver has additional information sources? Does the receiver benefit from having the additional sources? We extend a Bayesian persuasion model to a receiver's acquisition of costly information. The game can be solved as a standard Bayesian persuasion model under an additional constraint: the receiver never gathers her own costly information. The `threat' of learning hurts the sender. However, the resulting outcome can also...
|
|
|
Using Bluetooth technology for COVID-19 contact tracing
Jann, Ole ; Kocourek, Pavel ; Steiner, Jakub
The coronavirus is transmitted indiscriminately through proximity, which makes tracing infections difficult. Bluetooth tracing apps can reliably record transmission possibilities even when the participants do not know each other and do not remember the interaction. This can be done with a high degree of privacy. A well-designed app provides a similar level of privacy to not using an app at all. Decentralized data storage means that the privacy and security of the system is highly resilient against exploitation by any powerful actor (such as a government). A disadvantage of protecting privacy through decentralization is that tracing apps need to be taken up by the population one person at a time. Their use cannot be checked remotely and hence cannot be effectively mandated by governments or health authorities. A very high degree of take-up is necessary to make them an effective weapon against the virus. The eRouška app by COVID19cz follows these principles and offers a very high degree of privacy protection. Specifically, it does not collect any data except the phone numbers of non-infected users, and only collects anonymized meeting data (and no location or other metadata) of infected users – this data is only available to a hygienist after voluntary data transmission by the user. No data is transmitted to the server without explicit user agreement.
|
|
|
An overview of foreign testing practices from an economic and statistical perspective
Steiner, Jakub ; Kulveit, J. ; Matysková, L. ; Jann, Ole ; Kocourek, Pavel ; Novák, Vladimír
Our testing algorithm for Covid-19 infection and how we adapt it in response to the epidemic's progress, contact tracing technology, and knowledge of the virus, will be crucial in our fight against the epidemic. That algorithm must be developed with input from epidemiologists, biochemists and others. As economists, we believe we also play a relevant role, since testing for Covid-19 infections is a classic case of the problem of allocating rare goods – tests. We have compiled the following annotated overview of testing practices abroad as a starting point for an interdisciplinary discussion, fully aware of our lack of sufficient knowledge in epidemiology, biochemistry and medicine.
|
|
|
The economics of testing for Covid-19: beware of greater damage than benefit
Kulveit, J. ; Steiner, Jakub
There are limited numbers of tests for Covid-19, especially of the more precise type known as PCR tests. That means they must be used as efficiently as possible, in terms of which people are tested. Efficiency in the use of these tests is not only a matter of revealing how many people are infected, but also has to do with the potential ability of those tested to spread the infection further, which a positive test result can help to prevent. The algorithms by which the short supply of Covid-19 tests is assigned must therefore be grounded in the social, rather than individual benefits of testing. People whose testing brings maximum social benefit should be given priority. The social benefit of testing a particular person is calculated primarily in terms of the a priori likelihood (rough estimation) of that person being infected, based on information about where they live and their lifestyle. The second important criterion is that individual’s epidemiological significance, which is an indication of how much the individual in questions comes into contact with, and is likely to come into further contact with other at-risk people. This can also be established by asking pertinent questions to the individual themselves, complemented if desired by a survey in the place where they live. Although our estimations of the social benefit of testing are based on imprecise and incomplete data, the algorithm for allocating the limited numbers of PCR tests we have that is based on them is more socially effective than blanket use of them for testing anyone suspected of having contracted Covid-19. The greatest social benefit of testing comes from identifying the infection in people whose level of social interaction is high, during the phase of the infection in which they do not yet have any symptoms, that is, in so-called superspreaders. When evaluating the results of any test it is necessary to bear in mind that no test is ever entirely precise (reliable). Although PCR tests are very precise in laboratory conditions, errors can occur when samples are collected in the field, for example through poor sample handling. Our interpretation of the test result is then sensitive to the a priori likelihood that the person in question is infected, which might be low even when the test is positive, because the tests are not entirely reliable. It is not appropriate to use tests in situations in which nothing about the healthcare official’s decision about the next steps to take will change, regardless of the test result. If the healthcare official knows beforehand that they will not change their approach even if the result of the test is surprising, they should not waste one of those rather rare tests on testing the patient in question. Among the general public, testing can create undesirable motives that facilitate the spread of the infection. For example, if only people with a high a priori likelihood of infection are tested, people will exaggerate their own likelihood of being infected in an effort to gain access to testing. That’s why, for example, at the beginning of the Covid-19 epidemic in the Czech Republic, some Czechs who wanted to get themselves tested told healthcare officials they had been in Italy, when in fact they had not. Having been approved for testing, they then unnecessarily exposed themselves to the virus at the testing facilities. Social stigma surrounding infection with Covid-19 also poses complications, since it motivates people to conceal any symptoms they are experiencing and avoid being tested. For that reason, it may be sensible not to publicise details of the algorithm according to which individuals might be selected for testing, and to work systematically and intensively to prevent any stigma associated with Covid-19 infection through media campaigns and raising public awareness.
|
|
|
Essays on Information Economics
Matysková, Ludmila ; Steiner, Jakub (advisor) ; Kamenica, Emir (referee) ; Dean, Mark (referee)
In the first chapter, we study the effect on the economy of platforms for online consumer reviews. Consumer reviews may have perverse effects, including delays in the adoption of new products of unknown quality when consumers are boundedly rational. When consumers fail to take into account that past reviewers self-select into purchases, a monopolist may manipulate the posterior beliefs of consumers who observe the reviews, because the product price determines the self-selection bias. The monopolist will charge a relatively high price because the positive selection of the early adopters increases the quality reported in the reviews. In the second chapter, we study a game between a sender and a receiver in a framework of Bayesian persuasion. A sender choosing a signal to be disclosed to a receiver can often influence the receiver's actions. Is persuasion more difficult when the receiver has additional information sources? Does the receiver benefit from having the additional sources? We extend a Bayesian persuasion model to a receiver's acquisition of costly information. The game can be solved as a standard Bayesian persuasion model under an additional constraint: the receiver never gathers her own costly information. The `threat' of learning hurts the sender. However, the resulting outcome can also...
|
|
|
Selective sampling with information-storage constraints
Jehiel, P. ; Steiner, Jakub
A decision-maker acquires payoff-relevant information until she reaches her storing capacity, at\nwhich point she either terminates the decision-making and chooses an action, or discards some\ninformation. By conditioning the probability of termination on the information collected, she\ncontrols the correlation between the payoff state and her terminal action. We provide an\noptimality condition for the emerging stochastic choice. The condition highlights the benefits of\nselective memory applied to the extracted signals. The constrained-optimal choice rule exhibits (i) confirmation bias, (ii) speed-accuracy complementarity, (iii) overweighting of rare events, and (iv) salience effect.
|
|
|
Essays on Economic Theory
Goryunov, Maxim ; Steiner, Jakub (advisor) ; Fabinger, Michal (referee) ; Kircher, Philipp (referee)
In the first chapter of this work, I study the sorting of workers to firms, when firm size is explicitly taken into account. I develop a method to non-parametrically identify match production function from data on workers' wages and firms' revenues and posted job vacancies. Under the proposed identification procedure, ordering of workers and firms is identified independently, and can therefore be achieved using potentially different data sets. The model sheds light on the question of exporter wage premium: exporters pay higher wages because they are larger, and higher wages are required to support a larger firm size. In the second chapter we elaborate on Anas' (2004) impossibility theorem, which states that monopolistic competition or economies of scale alone are insufficient to explain the growth of cities in response to growing population or decreasing trade costs (under constant urban costs); cities shrink. To enhance the realism of assumptions, instead of Anas' normative approach, we introduce migration and developers' equilibria and another sector. Still, "vanishing" remains robust! Ultimately, we argue that the "vanishing" mechanism looks realistic and can have an explanatory power: industries, free of externalities, should locate in small towns. Moreover, the comparative statics shows how such...
|
|
|
Optimal illusion of control and related perception biases
Gossner, O. ; Steiner, Jakub
We study perception biases arising under second-best perception strategies. An agent correctly observes a parameter that is payoff-relevant in many decision problems that she encounters in her environment but is unable to retain all the information until her decision. A designer of the decision process chooses a perception strategy that determines the distribution of the perception errors. If some information loss is unavoidable due to cognition constraints, then (under additional conditions) the optimal perception strategy exhibits the illusion of control, overconfidence, and optimism.
|
|
| |
guest ::
