Economic growth, politics and institutions
Date
2015
Authors
Okabe, Tomohito
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
After the introductory chapter, this thesis comprises three further chapters that aim to shed light on issues that are relevant to economic growth, politics and institutions. Chapter 2 studies transitions in individual preference for political parties, using a Markov chain. The partisanship preference is referred to as party identification (PID). This study proposes a novel method for estimating the individual-level Markov dynamics from aggregate-level data. This method is particularly useful when panel data that directly convey individual-level PID transitions cannot be accessed. It allows the identification of the most likely underlying dynamics from a set of potential Markov processes. It also enables the researcher to impose extra information or a hypothesis on the Markov chain and test for it. In this study, the proposed method is illustrated with U.S. cross-section data, estimating transitional matrices of PID Markov chains for African Americans and Caucasian Americans. The results have captured distinct dynamics between the two races. The transitional matrices are also estimated, imposing Mebane and Wand's [1997] "gradual transition hypothesis", which asserts that the PID of American people gradually transitions through "Independent status" rather than jumps from one extreme status to another (e.g., from Democrat to Republican and vice versa). Compared to the results gained without the gradual transition hypothesis, it is found that the individual-level persistence of some partisanship is enhanced for both African Americans and Caucasian Americans. Chapter 3 presents a political-economy model of endogenous growth intended to account for regional differences in growth. The model is described by parameters representing three local channels: a market, a political institution and inherent potential. These local channels serve to shape the balanced growth path of an economy via democratic elections. Electoral competition is modelled as probabilistic voting, in which conflicts of interests over allocation policies for a productive public good and a lump-sum transfer arise among heterogeneous agents. The Nash equilibrium is characterised by the joint distribution of agent types and ideology. The agent types are endogenously determined through an ex ante choice process. The voters' ideology is assumed to change over time according to Markov chains. This study has shown that the economy converges to the balanced growth path as the ideological Markov chains approach the steady state. Chapter 4 proposes this research's original growth accounting, using the political-economy model presented in Chapter 3. This growth accounting exercise seeks to identify the most influential local channel by evaluating the relative impact of individual key parameters on growth. For this purpose, the model is calibrated with state-level data from 26 American states. The study finds that the market institution channel causes the largest impacts in most of the selected states and the political institution also has a sizeable impact. Numerical exercises of comparative statics and distortion analysis are also presented. The former numerically investigates the equilibrium behaviour with respect to the key parameters. The latter investigates the distortion due to democratic politics and political bias embedded in the political-economy model.
Description
Keywords
Citation
Collections
Source
Type
Thesis (PhD)
Book Title
Entity type
Access Statement
Open Access
License Rights
Restricted until
Downloads
File
Description