Three Essays on the Political Economy of Public Sector Governance
Thesis or dissertation
University of Exeter
Reason for embargo
I need time to arrange publication of my research.
This Ph.D thesis is made up of six chapters: together with Introduction and Concluding Remarks, there are one extensive literature review and three main essays. The theme of this thesis is "The Political Economy of Public Sector Governance" and I explore it by analysing the two main actors in the interaction between citizens and politicians: Mass Media and Bureaucracy. The World Bank in several publications since early 2000 has brought to the attention of politicians, public servants, social scientists and, as far as an institution like the World Bank can do, the general public that what really is important and does make a difference in the economic growth and social development of nations are not policies but (political and social) institutional quality. In order to make institutions work well, so they are able to promote the greatest welfare for all the citizens, it is necessary to have good governance. One of the ingredients of an optimal governance arrangement is the possibility for the citizens to make their government accountable for what it does (not) and responsive to their needs. Therefore, in order to have good political institutions citizens have, on one hand, to control their government and, on the other hand, to voice their needs, preferences and ideas, also when the ballot box is not ready at hand. Mass Media has at least these two functions in the relationship between the citizens and the (incumbent) politicians. In the first essay I analyse citizens' voting decisions and collusion between media and politicians and how this phenomenon affects the behaviour of citizens towards disciplining and selecting the incumbent politician, when citizens have at their hands two sources of information about the quality of the incumbents and their performance: the quantity of a good publicly supplied by the government and a signal coming from the mass media on politician honesty. The setting comprises a two period game, where voters, in the first period, have to decide, observing the information available through media and good publicly produced, whether to vote off or reelect the incumbent politician to the second period electorate mandate. By employing both two signals, citizens manage to sort out honest politicians from dishonest ones more often than if they were relying on media information only. Moreover the existence of both signals makes collusion harder to achieve than in the case of one signal only. Furthermore, the welfare analysis reveals that, contrary to previous findings, the presence of media is not always welfare improving. The usefulness of media for citizens depends critically on the time discount factor between the two periods: when the time discount factor is larger than a certain threshold, it is optimal for the citizens to receive information from media; when the time discount factor is lower than the threshold, their optimal decision is not to get any information. Finally, I argue that when rules at the constitutional level are not possible and citizens cannot commit to have less information, then collusion between media and politician can be welfare improving for citizens, contrary to previous results in the literature. In the second essay I investigate the role of Mass Media as a bottom-up way of communicating dispersed information from citizens to incumbent. Citizens transmit useful information thanks to the newspapers they buy and read. However, these newspapers are produced by a third party (a Media Tycoon) that has his own incentives. In particular the Media Tycoon has to decide whether to produce a newspaper that allows the citizens to participate in the public debate (Broadsheet) or does not (Tabloid). Given the fact that this instrument can be bought but not directly produced by the citizens, there exists a tension between the benefit of using a newspaper to express citizens'views and the possibility that this newspaper can be actually produced. Results show that producing a Broadsheet always improves the quality of policy decision making on part of the incumbent. A notable result is that in order to enhance the quality of the public decision making it is better to have any Broadsheet than not having one, whatever is the public stance the newspaper takes about the issue at stake. In this essay I first assume that there is only one group of citizens which is interested in having the optimal policy adopted, i.e. the Middle Class and I assume the Middle Class citizens are the only one who read newspapers. Subsequently I analyse how the results change when citizens from the other classes read newspapers as well. I show how the "partisan readers", committed to buy the Broadsheet supporting the policy they prefer, can ease the production of the Broadsheet. In this case the existence of partisanship and of ideological readers make the implementation of optimal policy easier, not harder, contrary to conventional wisdom. In the work of the World Bank, and in all the scientific production about how to establish and foster the development of good governance, corruption is one of the main diseases that can affect the correct relationship between citizens and public officials. So it is important to study how good institutional quality can fight corruption in several different fields of the political and economic environment. The third essay evaluates the effect of corruption on the regulation of business entry. A theoretical agency model of bribes is introduced, with strategic interaction between the firm, the corruptible public sector employee and the government. This model allows the evaluation of reforms targeting business startup procedures with regards to the incentives of the various actors involved in this process. Findings show that corruption in equilibrium between entrant firms and public servants could be self-sustained in the absence of government intervention. When deriving the equilibrium outcomes of some reforms like performance wages, privatisation and full liberalisation of entry, results show that transaction costs related to bribes are central in determining the optimal reform strategy. Although liberalisation is the preferred reform option for firms, government fiscal revenues and overall social welfare, firms surprisingly would prefer performance wages implemented in public registry service rather than the privatisation of this service. This holds despite the additional tax burden on firms necessary to finance higher civil servants'wages.
Myles, Gareth D.
PhD in Economics