Essays on Fiscal Transparency and In°ation Expectations
MetadataShow full item record
This dissertation comprises two related essays on ¯scal transparency and a third essay on in°ation expectations. The ¯rst chapter provides a brief introduction to the next three chapters. The second chapter proposes some indices of ¯scal transparency based on the IMF's Code of Good Practices on Fiscal Transparency. Each country is assigned a category for a number of aspects of ¯scal transparency based on the information in the IMF's \Fiscal Transparency Reports on Observance of Standards and Codes (ROSC)". This classi¯cation is used to construct indices covering four clusters of ¯scal transparency: data assurances, medium-term budgeting, budget execution reporting, and ¯scal risk disclosures. I consider the robustness of these indices to di®erent choices associated with construction of the indices. Lastly, I present some cross-country comparisons of ¯scal transparency and analyze the relationship of other institutional variables to ¯scal transparency. The third chapter examines several hypotheses regarding ¯scal transparency using the indices developed in the second chapter. I discuss the channels through which ¯s- ii cal transparency can a®ect market credibility, ¯scal discipline, and corruption. After controlling for other socio-economic variables, more transparent countries are shown to have better credit ratings, better ¯scal discipline, and less corruption. The ¯nal chapter considers the question whether in°ation expectations are driven by household in°ation experience. Household surveys reveal that in°ation expecta- tions vary considerably across households. Furthermore, studies have found that these expectations vary systematically over demographic variables. This chapter suggests that the variation in individual expectations of in°ation may be based partly on the in°ation experienced by individual households. I calculate a household speci¯c level of in°ation based on the BLS consumer expenditure survey (CEX) data. Then a two- sample two-stage estimation methodology is used to study the correlation between the experienced household in°ation and reported in°ation expectations data in the Michigan Survey data for similar households. I ¯nd that expectations of in°ation indeed vary with the in°ation experience, moreover personal experience seems to be overly in°uential.