This thesis consists of an introductory chapter, three main chapters, and a concluding chapter. In Chapter 2, my co-author and I provide new empirical evidencethat the distribution of liquidity has a strong in-sample and out-of-sample predictive power on intraday market volatility. To this end, we introduce a novel way of summarizing the relative depth provision in the whole limit order book. Our measure, global depth, considers the entire quoted depth and assigns weights decreasing with distance from the best quotes. We document that global depth outperforms alternative predictors of volatility, such as the bid-ask spread, standard depth variables, and measures of trading activity, in explaining the variations in market volatility. The third chapter, forthcoming in the Journal of Banking and Finance, investigates the effects of competition and signaling in a pure order driven market and examines the trading patterns of agents when walking through the book is not allowed. We show that the variables capturing the cost of a large market order are not informative for an impatient trader under this market mechanism. We also document that the competition effect is not present only at the top of the book but persistent beyond the best quotes. Moreover, we show that institutional investors’ order submission strategies are characterized by only a few pieces of the limit order book information. The fourth chapter provides evidence that implied correlation is a significant indicator of market-wide risk. From an aggregate perspective, I document that implied correlation explains an important fraction of the variation in market excess returns, with high implied correlation followed by an increase in subsequent market returns. The predictive power is stronger at a forecast of bimonthly, quarterly and semiannually return horizons and robust to the inclusion of standard predictors. Moreover, I show that the information content of the correlation risk premium on market returns is fully driven by the implied correlation. My findings indicate that periods of high market-wide correlation produce a deterioration of the investment opportunity set and, as a consequence, an increase in the equilibrium expected return.

This thesis investigates the relative merits of econometric modeling, statistical and judgmental techniques for predicting debt crises and assessing the risk of credit migration. The increased reliance on econometric or statistical approaches and credit rating systems in risk management has intensified the need for more rigorous analysis of their finite sample properties. A better understanding of the available tools has implications for credit risk management, regulation and policy decision-making. The thesis contributes to the extant sovereign risk literature in three areas. First, it addresses the question of whether controlling for unobserved heterogeneity is important for predicting debt crises and explores a pervasive inference problem in Early Warning Systems (EWSs). Second, it addresses the development of an `optimal' EWS for sovereign debt crises that accommodates the decision maker's preferences. Third, it considers the measurement of sovereign credit migration matrices using different estimators and explores non Markov effects in the rating dynamics. Chapter 2 confronts competing models of sovereign default that differ in how country-, region- and time-specific effects are treated. Statistical tests and information criteria overwhelmingly favour more complex models with country heterogeneity that possibly changes over time. However, simplicity beats complexity in terms of forecasting. Simple pooled logit parameterization, that control either for regional heterogeneity or for time effects produce the most accurate forecasts and outperform several naive predictors. Chapter 3 investigates the severity of the autocorrelation problem in EWS of sovereign default. This stems from seeking to provide crisis warnings over a horizon that is longer than the frequency at which the forecasts are updated and from the sluggishness of the typical exogenous indicators. Neglecting residual serial autocorrelation in such models is shown to be far from innocuous. Inferences are overturned when using a correction. This phenomenon is generally clearer for the macroeconomic ratios that are more persistent. Chapter 4 combines three fundamentally different classification techniques - econometric, statistical and judgmental- to produce an EWS for sovereign default. The optimal choice of crucial EWS elements is shown to depend on the decision-makers' preferences. The forecast ranking of classifiers is found to be unstable and overall the classifiers appear to have different strengths. Payoffs from forecast combination are documented and the combining scheme is shown to depend on the decision-makers' loss function. Chapter 5 turns to the estimation of sovereign transition probability matrices and evaluates the popular discrete multinomial estimator against two continuous hazard rate methods that differ in their treatment of time-heterogeneity. Bootstrap simulations of the rating generating process reveal interesting insights. Hazard rate estimators yield more reliable default probabilities. Efficiency is further enhanced upon relaxing homogeneity. Downgrade momentum and duration effects are found to be present in the rating process.

The work undertaken in this study empirically explores the determinants of regulatory bank capital buffers, and how they influence bank decisions. Focusing on bank capital management under the Basel I framework, this thesis serves to address some of the concerns that have been voiced regarding the implementation of the new regulation (Basel II) and the broader economic effects that could result. In particular, the research chapters of this thesis examine the cyclical behavior of European bank capital buffers, the long run relationship between bank capital buffers and charter values, and the simultaneous adjustments of capital and risk. In each of the research chapters, we acknowledge the endogenous nature of the capital decision of a bank, and assume that banks will define an internally optimal probability of default (a function of risk and capital) to be managed over the long term. Adjustment costs, illiquid markets, together with the costs associated with a regulatory breach contribute as factors in a banks internal decision when setting a target capital ratio. Treating capital in this way, we note that it is the amount of capital held above the requirement that determines a banks attitude towards risk. Importantly, this work has shown that excessive risk taking is rarely a consequence of insufficient capital.

The present thesis comprises three essays in international finance, with a focus on the foreign exchange market. The first chapter assesses the predictive ability of a comprehensive set of empirical models of exchange rates, in addition to a standard technical trading strategy, on monthly exchange-rate returns for four developed and four emerging countries across different horizons. I implement a rolling window approach to the estimation and forecasting of the models, and construct an encompassing forecast. I also assess the economic value of the out-of-sample forecasting power of the empirical models using a simple dynamic allocation strategy, and find three key results: (1) the Taylor rule model consistently outperforms, economically and statistically, all the other models at the 1-month horizon. (2) The technical rule has superior predictive power over the random walk benchmark, across horizons, particularly for developed markets. (3) There are statistical gains from an unrestricted combined forecasting model at the 1-month horizon. The second chapter constitutes a survey that focuses on internationally tradable goods and services. Our motivation is that while excellent surveys exist in the literature on this topic, they focus largely on broad baskets of prices and, most commonly, on the consumer price index. We instead focus on the specific subset of the relevant literature that analyses deviations from the LOP applied to individual goods and services and specific sectors. The emphasis is hence on tradable items rather than broad baskets that also include a substantial nontradable component. Specifically, the objective is to distil the literature on the properties of deviations from the LOP applied to internationally tradable goods or sectors. We conclude that a careful reading of the literature suggests that this notion of PPP holds in the long run for a broad range of tradable goods and services and for a broad set of currencies. In the third chapter, I build a "commodity currency strategy" for exchange rate forecasting that conditions on changes in the global prices of commodity indices. The risk-return profile of this strategy reveals that the predictive ability of commodity prices for the exchange rate appears to be significant, and the returns appear to be uncorrelated to popular exchange rate strategies such as the carry trade and currency momentum. The market factor captures more than 70% of the cross-sectional returns of the proposed strategy and suggests a negative relation between equity returns and currency returns that are driven by commodity price changes. The commodity currency strategy is prone to high transaction costs which can only be circumvented by investing in developed markets with low costs and high liquidity.

The purpose of this thesis is to conduct a detailed study of optimism in financial decision making. I contribute to the literature by clarifying the relationship between financial optimism and individual investors’ portfolio choice. I also investigate whether optimism benefits an investor’s objective and subjective well-being within the same study by using large-scale survey data. I then explore how feedback, framing, and personality, contribute to financial optimism using controlled user experiments. Both survey-based and experimental approaches are applied in this thesis to study various aspects of optimism in a financial decision making domain. In this thesis I propose a theoretical framing work for measuring financial optimism and use these measures to analyse investor profiles. My survey-based studies show that optimistic investors prefer to invest in risky portfolios to risk-free portfolios, and borrow higher debt and larger mortgages. Optimists are significantly younger with lower accumulated financial wealth compared to non-optimists. Financial optimism is found to be beneficial in improving objective well-being by increasing future financial wealth, but this positive effect is very limited in terms of increasing future total wealth. Optimism is associated with current happiness and satisfaction which means optimism might help to improve current subjective well-being, but the long-term effect of optimism on happiness might be less desirable if the investor’s realised financial situation is lower than expected. By conducting experiments on subjects given investment tasks in a controlled environment, I find that positive feedback on previous portfolio returns decreases optimism when forecasts on future portfolio returns are made in absolute values, while positive feedback increases optimism when participants forecast in relative terms. I also show that framing influences financial optimism - optimism is higher when forecasting in absolute values than in percentages. I discovered that certain personality traits, such as extraversion and modesty, correlate with financial optimism. Optimism is also strongly positively associated with an attitude for risk tolerance. The overall implications of this thesis is that when making a financial decision, individual investors should not neglect the effect of optimism on their choice of portfolio. Optimism is beneficial towards both objective and subjective well-being, however such positive influence of optimism is fairly limited and should not be magnified. Optimism might not be subject to the control of an individual because optimism could derive from environmental factors, such as feedback and framing, as well as from internal factors to the investor, such as personality and innate risk attitude.

This thesis comprises three essays on international finance, focusing on international capital flows, foreign exchange market and official foreign exchange intervention. The first chapter assesses the relative contribution of common (push) and country specific (pull) factors to the variation of bond and equity flows from the US to 55 other countries. Using a Bayesian dynamic latent factor model, we find that more than 80% of the variation in bond and equity flows is due to push factors from the US to other countries. Hence global economic forces seem to prevail over domestic economic forces in explaining movements in international portfolio flows. The dynamics of push and pull factors can be partially explained by US and foreign macroeconomic indicators respectively. The second chapter presents new evidence on the microstructure of exchange rates in emerging markets. Using a novel dataset that records all spot US dollar transactions in the Chilean foreign exchange intraday market over 4 weeks in 2008 and 6 weeks in 2009, we investigate the relationship between exchange rates and order flows. We find supporting evidence that the contemporaneous relationship between exchange rates and order flows is time-varying. In this market, interbank order flow only accounts for a small portion of the exchange rate impact of total order flow, and the central bank orders influence private order flow behaviour. Compared to advanced economies, cointegration tests and long run relationship estimations between exchange rate and order flow indicate slow reversion from long run trend deviations. In the third chapter, we examine the intraday effects and success rates of official intervention in the Chilean foreign exchange market. The impact of official intervention on exchange rate daily returns has been widely revised in the literature, confirming in many cases the signalling channel for the transmission of the intervention effects. Our investigation at a higher frequency indicates that microstructure channels also work for the Chilean case. Specifically, the central bank's order flow directly affects the exchange rate returns contemporaneously and within a 2 hours range around the intervention event. In addition, actual interventions affect the price impact of private order flows, and are successful at moderating its trend.

The thesis contains an empirical study designed to reveal why initial public offerings (IPO) of common stocks (A-share) in China are on average under-priced from alternative angles as opposed to more established theories, using a dataset of 880 IPOs from January 1996 to December 2003. A much higher degree of under-pricing compared to developed markets and even other emerging markets is a distinct feature of China’s A-share IPOs. Previous literatures based on classical hypotheses have not been able to fully explain such high level of under-pricing. Hence, alternative explanations have been put forward by academics as well as practitioner. It is said that the Chinese government has big influences in China’s primary market through tightly controlled issuing system and opaque regulatory constraints. People speculate that such influences have been both intentionally and unintentionally exerted, causing IPO under-pricing. A major contribution of the thesis is to test some new hypotheses based on three untested statements in China’s IPO literatures i.e. speculation effect, “Western Region Development” policy effect and government protection effect, which are all associated with government direct or indirect influences. More specifically, speculation effect hypotheses assume that the government constraints and regulator drawbacks have caused high level of speculation, which in turn drives the IPO under-pricing. “Western Region Development” policy effect hypotheses claim that government intentionally uses IPO under-pricing to lure investments into less developed and thus less favourable western region companies. Government protection effect hypotheses conjecture that IPO under-pricing is a compensation for investors’ concern of potential government interference in the government-protected firms. The thesis finds that speculation effect hypothesis is largely supported by empirical data while the other two are not. The thesis has also re-tested hypotheses advanced in previous literatures including classical information asymmetry hypotheses, ex ante uncertainty hypotheses, investors’ behaviour hypotheses as well as existing China-specific hypothesis such as listing time lag hypothesis. Proxies such as government retention rate that have emerged in privatisation IPO literatures are also borrowed by Chinese researchers looking into China-specific institutional settings such as dominant government control. The thesis finds that information asymmetry hypotheses in general show more strength in explaining China’s IPO under-pricing while empirical evidence for other hypotheses are either mixed or weak. In the end, the thesis finds that alternative angles emerging from the tests of the three statements combined with some classical hypotheses supported by empirical data have a greater power to explain China’s high IPO under-pricing. The so-called “floatation game” hypothesis has been put forward by researchers such as Tian (2003) who claims that the Chinese government makes use of IPOs with different lengths of IPO listing time lag i.e. the time lag between IPO announcement date and actual listing date to adjust the equity market cycle. In other words, the government let go public the IPOs with longer listing time lag thus more likely higher initial returns in the bear market and vice versa. The thesis does not find support to the “floatation game” hypothesis, although the thesis does find that the market cycle is closely related to the IPO under-pricing and the number of IPOs issued.

This thesis draws upon contemporary research in economic geography and the social sciences to explore the spatial relationships that exist between residential mortgage lenders, investment banks and investors and the subsequent geographies that are produced through these intertwined networks. The research is informed through empirical material collected through semi-structured interviews with directors and associates working in the financial sector to see how consumer mortgages are produced and restructured into debt securities. There is a particular focus on how the UK financial sector has undergone restructuring, as a consequence of the politics of financialisation since the 1990s, which aligned the residential mortgage market with the circuits of international capital. The thesis examines three areas of banking and finance to comprehend how retail mortgages have become embedded within international finance. First, the thesis explores how deregulation in the UK initiated a spatial reorganisation of mortgage production networks and funding. Second, the research investigates the migration and adoption of automated decision-making technologies, highlighting how these devices have reshaped the geographies of banking, and are inherently geographical themselves. Third, the thesis focuses on how mortgages are (re)engineered into debt-securities, with a particular focus on how geography is used to mitigate credit and tax risks. It is argued that the restructuring of the UK retail sector and its increased integration with the international circuits of capital exacerbated the exposure of the UK’s economy to the effects of the international credit crunch. Furthermore, the thesis underlines the effect of geography which has shaped the adoption, of new financial technologies and strategies, through local regulations, epistemic cultures and histories.

One of the most important aspects of foreign direct investment (FDI) is that it embodies advanced technologies and business practices which can spill over to domestic firms via various channels, e.g. labour mobility, input-output linkages, export of multinational affiliates, demonstration and competition. This research combines computable general equilibrium (CGE) modelling and econometric techniques to quantify FDI productivity spillovers. The research is conducted in the context of the Chinese economy. A static lOl-sector CGE model is constructed to measure the endogenous productivity spillovers of FDI. Spillover effects are analysed under three different market structure assumptions, namely perfect competition, monopolistic competition with homogeneous firms, and monopolistic competition with heterogeneous firms. The research results show that the presence of FDI productivity spillovers can generally improve the productivity and output level of domestic enterprises in China. Spillovers make foreign firms' total output decrease. But collectively, spillovers exert positive impact on national aggregate variables, i.e. GDP, total output and welfare. The market structure assumptions of monopolistic competition and firm heterogeneity provide more perspectives (e.g. product variety and scale) for this research than the assumption of perfect competition does. A removal of preferential corporate income tax treatment on foreign enterprises can increase the output level of domestic enterprises and promote national welfare. From a dynamic perspective, it could also promote the productivity splllovers from foreign firms.

The aim of this thesis is to analyse formal and informal credit in Ethiopia and Malawi. As credit markets in developing economies are dominated by informal institutions, the analysis of the interaction between formal and informal institutions is crucial to understanding how welfare improvements can be achieved. The thesis begins with an explanation of the motives for demanding credit. It then focuses on analysing the existence, diffusion and persistence of informal nance in developing economies. Much research on this topic remains hamstrung by the quality and availability of data and by the lack of empirical models, constraining the meaningful identification of the characteristics of the localities where informal institutions operate. The central idea of the first essay is to develop an empirical model that explains the determinants of participation in informal credit arrangements. We adopt an endogenous switching regression model of access to informal credit where the availability of a particular type of informal arrangement varies across clusters in rural Ethiopia. This strategy allows for taking into account substitutability between sources as well as household and cluster socioeconomic characteristics. The second essay exploits the idea that banks can crowd out informal borrowing in Malawi by creating microfinance institutions that acquire information in innovative ways. We adopt propensity score matching and find that the creation of a specific microfinance programme reduces informal borrowing. The third essay uses the credit limit variable to test liquidity constraints and the spillover hypotheses in Malawi. A ten percent increase in the informal credit line increases households' demand for informal credit by more than nine percent. We also find that a 10 percent increase in the credit limit of a microfinance programme reduces the informal demand by four percent, partly explaining the coexistence of formal and informal credit institutions.