In Ely (2017), the author develops a model of dynamic information management in a principal-agent problem setting. In the model, the principal wishes to align one’s interests with myopic agents; in fact, because of the dynamic nature of the problem, the principal’s objective is to maximize the time that this interest alignment lasts for. The agents have certain beliefs or suspicions about the state of the world, where the state is only privately observed by the principal. The author describes that it is the existence of such beliefs that make perfect information (described as a “beep”) and no information dissemination (“no-beep”) sub-optimal strategies for the principal. The author reaches the conclusion that the optimal strategy characterized by the model is one that utilizes delays in information dissemination, using mathematical proofs and dynamic programming. He also showcases examples in which this model can be applied to, including within-firm information control as well as a monopolistic market selling goods that depreciate at a certain rate.
After having analyzed the paper, I suggest the following extensions to Ely (2017). First is a theoretical extension in which I propose to look at agents that are not bound to myopia. The current model discussed in Ely (2017) assumes that the agents, as opposed to the principals, are myopic in their utility maximization. In addition, as long as their threshold value of belief is not reached, the agents are assumed to align their interests with those of the principals. While this allows for the formulation of a more tractable dynamic programming and other forms of simulations, it may be considered somewhat limited in its empirical implications. For instance, in a situation such as a potential for bank run (which was also discussed in the paper as well), it would potentially be an oversimplification to state that the agents (account holders or customers at the banks) have a completely aligned interest with the principals (bankers) even when their belief that a bank run may occur is below a designated threshold. The agents who have higher stakes in the said setting may continuously want to verify the financial and managerial stability of the principals and may instead be willing to pay more for greater amount of information. Extending Ely (2017) dynamic model to encompass a greater range of agents will allow for the analyses of many more empirical questions.
My second suggestion is an empirical extension to managerial economics, which has to do with the within-company principal-agent problem that was mentioned in Ely (2017). Borrowing from the analysis of managerial abilities in Goldfarb and Xiao (2011), I will try to examine whether the managers with higher managerial abilities are capable of designing a more effective information-dissemination mechanism, close to one that is presented in Ely (2017). One significant problem with proposing such an empirical study is, of course, the availability of data, and even if such data were available, it is highly likely to be suffering from issues such as missing data on the information of managers. A potential data source that I am considering is one provided in de Grip and Sieben (2005), which has to do with smaller firms (pharmacies, specifically). If this extension is successfully conducted, I believe that it will serve as a bridge between managerial economics and information economics.
References:
Ely, Jeffrey C. 2017. “Beeps.” American Economic Review. 107 (1): 31-53.
Goldfarb, Avi, and Mo Xiao. 2011. “Who Thinks about the Competition? Managerial Ability and Strategic Entry in US Local Telephone Markets.” American Economic Review. 101: 3130-3161.
de Grip, Andries, and Inge Sieben. 2005. The Effects of Human Resource Management on Small Firms’ Productivity and Employees’ Wages.” Applied Economics. 37 (9): 1047-1054.
In Ely (2017), the author develops a model of dynamic information management in a principal-agent problem setting. In the model, the principal wishes to align one’s interests with myopic agents; in fact, because of the dynamic nature of the problem, the principal’s objective is to maximize the time that this interest alignment lasts for. The agents have certain beliefs or suspicions about the state of the world, where the state is only privately observed by the principal. The author describes that it is the existence of such beliefs that make perfect information (described as a “beep”) and no information dissemination (“no-beep”) sub-optimal strategies for the principal. The author reaches the conclusion that the optimal strategy characterized by the model is one that utilizes delays in information dissemination, using mathematical proofs and dynamic programming. He also showcases examples in which this model can be applied to, including within-firm information control as well as a monopolistic market selling goods that depreciate at a certain rate.
After having analyzed the paper, I suggest the following extensions to Ely (2017). First is a theoretical extension in which I propose to look at agents that are not bound to myopia. The current model discussed in Ely (2017) assumes that the agents, as opposed to the principals, are myopic in their utility maximization. In addition, as long as their threshold value of belief is not reached, the agents are assumed to align their interests with those of the principals. While this allows for the formulation of a more tractable dynamic programming and other forms of simulations, it may be considered somewhat limited in its empirical implications. For instance, in a situation such as a potential for bank run (which was also discussed in the paper as well), it would potentially be an oversimplification to state that the agents (account holders or customers at the banks) have a completely aligned interest with the principals (bankers) even when their belief that a bank run may occur is below a designated threshold. The agents who have higher stakes in the said setting may continuously want to verify the financial and managerial stability of the principals and may instead be willing to pay more for greater amount of information. Extending Ely (2017) dynamic model to encompass a greater range of agents will allow for the analyses of many more empirical questions.
My second suggestion is an empirical extension to managerial economics, which has to do with the within-company principal-agent problem that was mentioned in Ely (2017). Borrowing from the analysis of managerial abilities in Goldfarb and Xiao (2011), I will try to examine whether the managers with higher managerial abilities are capable of designing a more effective information-dissemination mechanism, close to one that is presented in Ely (2017). One significant problem with proposing such an empirical study is, of course, the availability of data, and even if such data were available, it is highly likely to be suffering from issues such as missing data on the information of managers. A potential data source that I am considering is one provided in de Grip and Sieben (2005), which has to do with smaller firms (pharmacies, specifically). If this extension is successfully conducted, I believe that it will serve as a bridge between managerial economics and information economics.
References: