Essays on counter strategies against mistakes in information processing with various applications

Mosenhauer, Moritz (2020) Essays on counter strategies against mistakes in information processing with various applications. PhD thesis, University of Glasgow.

Full text available as:
[img]
Preview
PDF
Download (1MB) | Preview

Abstract

This thesis consists of three chapters, dealing with mistakes in information processing and how smart information design may improve upon it.

Chapter 1 studies a firm's organisational responses if its agents misevaluate information. If a manager overreacts to unusual events, it may be desirable for the firm to adopt the management practice management-by-exception. I develop a theoretical framework to study this technique and derive conditions under parsimonious assumptions for when it should be adopted. Moreover, I show how further assumptions can refine the model's predictions, establishing a direct link between the manager's over-responsiveness and organisational rigidity. The strategy is implemented by controlling the information that the manager receives. In fact, in the absence of information transmission and processing costs, it may be optimal to not send inherently valuable signals concerning the economy's state to the manager.

Chapter 2 investigates tools to counter excessive stock trading and increase profits for private households participating in the stock market. Creating a stylised hold or trade-scenario in a computer laboratory experiment, I find that by solely changing the information the participants receive, trading activity can be reduced by roughly 30%, increasing trading profits by more than 0.55 percentage points on monthly net returns. In particular, I consider two information treatments. First, I provide the participants with additional information by giving detailed feedback on their actions and outcomes at every turn. Second, when considering whether to hold a given stock or trade it for another one, I restrict participants' information on the recent performance of their allocated stock. Both interventions lead to significant changes in behaviour. Additionally, the 2x2 experimental design reveals that the effects stack.

Chapter 3 deals with growth diagnostics. Growth diagnostics is an influential policy framework that, in second-best settings, has been used to identify the priorities of policy reform in different countries. With limited information about the nature of interaction between different second-best distortions, mistakes (situations where realized social welfare losses overwhelm any intended gains) could occur in the implementation of growth diagnostics. Even allowing for the possibility of mistakes, would an adaptive implementation of growth diagnostics converge to a socially optimal outcome? This paper sets out the conditions under which such convergence occurs. A number of different historical examples are discussed to illustrate how such a process could play out in practice.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: H Social Sciences > H Social Sciences (General)
Colleges/Schools: College of Social Sciences > Adam Smith Business School > Economics
Supervisor's Name: Ghosal, Professor Sayantan and Vailakis, Professor Yiannis
Date of Award: 2020
Depositing User: Moritz Mosenhauer
Unique ID: glathesis:2020-81652
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 16 Sep 2020 08:26
Last Modified: 16 Sep 2020 08:26
URI: http://theses.gla.ac.uk/id/eprint/81652

Actions (login required)

View Item View Item