The aim of this thesis is to contribute to the understanding of corporate decision making regarding foreign direct investments (FDI) by applying two behavioural finance concepts, home bias and herding, to the analysis of FDI flows. I contribute to the literature by empirically testing for home bias and herding in an FDI context using a very broad panel dataset. I also contribute by examining the country policy implications of home bias on FDI flows between two countries by estimating the probability of an FDI relationship between two countries. In addition, I contribute by providing a generality of the results at a global, regional and country levels. The analysis in this thesis is conducted on a large panel dataset of the FDI inflows and outflows of 30 OECD member countries with their FDI partners, across 25 years in a bilateral country pair format which is a novel application of this dataset for the purpose of studying home bias and herding in FDI. The findings in this thesis confirm that there is an overall home country bias that is demonstrated through the preference for direct investments in places with greater physical, institutional and cultural proximity to the investor country. These general findings of home bias are observed and confirmed across different data segments: regional and country levels, across time and across different income country groups. I do not find that the effects of home bias have disappeared or diminished across time or at different geographic locations. Herding is another behavioural finance concept which is considered in the context of FDI outflows. Direct investors tend to herd around a perceived world or a regional leader when considering investments in faraway places and when they do not have the familiarity factors in common with an FDI partner country. Finally, by increasing the institutional and cultural familiarity, countries can significantly increase the probability that they will get a direct investment from a country with which they might not otherwise be having an FDI relationship.

Foreign exchange carry trades involve buying high yielding currencies while selling low yielding currencies. Contrary to the implications of the uncovered interest parity condition, carry trades have generated consistent profits in the past decades. As foreign exchange has gained increased relevance as an asset class in its own, the carry trade emerged as a major driver of foreign exchange market turnover. Given the widespread use and ease of implementation of carry strategies, active currency managers should be evaluated relative to a benchmark which incorporates a proxy for carry trade returns. Within this thesis we study the profitability of various carry portfolio strategies on a very recent data set ranging from the 1st of January 1999 to the 5th of March 2010. Within three distinct empirical chapters we analyse whether different asset allocation, market-timing and money management methodologies have the potential to improve the performance of a simple carry portfolio, such as the one implemented by the Currency Harvest exchange traded fund by Deutsche Bank. Three main findings emerge from our investigation on carry trade portfolios. First, we find that a simple carry trade proxy is difficult to outperform with asset allocation and market-timing techniques. Nevertheless, we would not conclude that professional currency managers should cease to implement carry strategies, since they can add value to the investment process by successfully addressing the issue of optimal leveraging for carry trades. Second, we find that the portfolio flows of carry traders do uncover pockets of predictability in the FX market. Strategies which aim at front-running the trades of carry strategies, do generate positive returns with low correlations to traditional carry trade strategies and therefore offer good diversification vehicles for carry portfolios. Lastly, we find that while profitable market-timing seems feasible on historical backtests, the results are strongly dependent on the correctly timing the credit crisis. Thus, we note that our results are affected by a lookback bias. We posit that before the credit crisis, portfolio managers would not have had the foresight to select the correct market-timing indicators. We thus advocate a broad diversification of risk indicators for carry trade timing.

This thesis aims to add empirical evidence to the corporate finance literature by looking at the financing decisions with a specific application to small companies in the context of the UK relatively highly regulated Main market, versus the lightly regulated Alternative Investment Market (AIM). I do this by gathering data on all quoted dead and alive companies in both markets from 1995 to 2008. I then split my sample firms in each market into different size groups and test my hypothesis within and across each group and each market. The thesis consists of six chapters. After an introductory chapter, I review the existing literature on capital structure and debt maturity controversies with an emphasis on recent empirical work. The next three chapters consist of three research papers. The first paper looks at the capital structure decisions of companies quoted in AIM and Main market across different size groups. In the second research paper, the maturity structure of debt is investigated in both markets. The third research paper tests the determinants of the delisting decision, particularly the effect of leverage using a sample of AIM companies. In the last chapter, I provide a summary of the main conclusions of the study and highlight some promising ideas for future research. The first empirical chapter analyses the drivers of leverage across firms' sizes and market of quotation. I find that companies that are listed on the Main market have higher leverage than those listed on AIM. My results show that AIM companies are subject to higher business risk and tend to have lower profitability and tangible assets. In addition, in both markets, small companies are different from large firms in their level of leverage, tangibility of assets, and profitability, suggesting that the drivers of the financing choice are size dependent. Interestingly, the impact of taxation is limited to only large companies in both markets. Similarly, the impact of the agency conflict is also limited to large companies, as for small firms I find a positive relationship between leverage and growth opportunities, in contrast to the predictions of the agency theory. These results suggest that size rather than market of quotation is more likely to explain firms' leverage. However, I find that the market of quotation affects their speed of adjustment toward target leverage ratios. Using the dynamic model of capital structure, I find that in the Main market, small companies adjust more rapidly than large firms, suggesting that they rely more on bank debt and thus result in lower costs of adjustment. In contrast, large firms on the AIM adjust more rapidly than small companies, suggesting that small AIM companies are subject to the highest costs of adjustment as they have the highest business risk and the lowest profitability. The second empirical paper investigates the determinants of the structure of debt maturity across firms' size groups in both markets. I find that firms quoted in the Main market use longer maturity of debt in contrast to their AIM counterparts. However, the structure of debt maturity is different between small and large companies, as small companies use shorter debt maturity. Moreover, I find that the determinants of debt maturity are relatively different across the two sets of markets, suggesting that the market of quotation, are likely to affect the structure of debt maturity. Particularly, the effect of leverage is mixed in those markets. In the Main market, companies with higher leverage use more long-term debt in contrast to those quoted in the AIM. In line with my results in the previous chapter, I find that the speed of adjustment depends on the market of quotation. Using a dynamic framework, I find that companies have a target debt maturity, but, while in the AIM large companies adjust more rapidly than small companies, I find the opposite in the Main market. I also contribute to the literature by assessing the impact of firm's life cycle on its choice of debt maturity. I use a sample of newly listed firms and assess the evolution of the maturity structure of their debt four years after their IPO. I find strong differences across the two markets. In the Main market, my empirical evidence shows that in contrast with small companies, large companies change the structure of their debt maturity significantly as they are more likely to use longer maturity of debt in the post-IPO period. While in the AIM, the structure of debt maturity is not affected by size as neither large companies nor small companies change their debt maturity significantly. In the last empirical chapter, I study the impact of leverage on the delisting decision. I address the following questions: Do firms delist from the stock market because they are unable to raise equity capital and redress their balance sheet? Previous studies state that raising equity capital is one of the main benefits of stock market quotation. I expect firms that are not likely to take advantage of this benefit to have higher listing costs and more likely to delist. I use leverage as a proxy variable and a sample of voluntary delisting from AIM. I find that delisted companies have higher leverage as they did not raise equity capital over their public life. My results suggest that companies with higher leverage are more likely to delist voluntarily. These results hold even after controlling for agency conflicts, liquidity, and asymmetric information. I also investigate how the market reacts to the delisting announcement. I find that on the announcement date, stock prices decrease significantly. However, this reaction is not consistent with previous studies that report positive excess returns for companies that go private through different forms of buyouts. The voluntary delisting does not deliver good news to the market and hence voluntary delisting leads to a decrease in stock prices. I also find that firms that increased their leverage in the year prior to the delisting decision generate significantly lower excess returns than other firms. I compare my results to firms that delisted from the AIM but moved to the Main market. I find that that these firms generate statistically higher and positive returns than the remaining firms that delisted voluntarily. My results highlight the negative impact of leverage and a lack of equity financing on firms' market valuation. My results contribute to the literature and to policy making in several ways. First, I test various controversial and new hypotheses by focussing on differences in institutional settings between the AIM and the Main market. The former is less regulated and it is more likely to attract younger, high growth, and riskier companies. These differences allow me to test various hypotheses developed in previous literature relating to the financing choices of firms. In addition, I provide a deeper analysis of the impact of size on the firms' financing choices. I focus on the differences in leverages across the two, markets, changes in maturity from the IPQ dates, and the drivers of the decision and timing from the IPQ date of companies in the UK. Unlike previous studies, I show that the theoretical determinants of leverage, such as taxation and agency costs, across firms' size groups are not homogeneous, independently of the market quotation. However, I find significant differences across the two markets in terms of dynamic changes in leverage. In addition, my results highlight the impact of leverage on the decision to delist, and imply that policy makers need to facilitate the financing of companies when they list on the market, so that the benefits of listings outweigh the costs, and firms will not rush to voluntary delisting.

The main aims and objectives of my thesis are to test the various conflicting hypotheses developed in the previous literature to explain firms’ dividend policy, focusing specifically on IPOs and cross-country analysis. In particular, I explore the theoretical links in the context of the important dividend theories including signalling, agency costs, lifecycle and catering and then empirically test the hypotheses by using a very large dataset of UK IPOs from 1990 to 2010, which is extracted from offering prospectuses. The first empirical study focuses on two aspects of post-IPO decision-making: the decision to initiate dividends and the timing of dividend initiation. I develop the testable hypotheses by linking the dividend decisions of IPOs with a number of firm characteristics and IPO-specific factors in the context of the theories relating to dividends and IPO. I find a strong negative relation between underpricing and the propensity of dividend initiation. This finding is in line with the implications of Dividend Discount Model and Rock’s (1986) “winner’s curse”. My results show that the likelihood of initiating dividends is positively associated with managerial ownership, underwriter reputation, firm size, profitability and long-term debt ratio. In addition, the results show that the initiation propensity is negatively influenced by a serial of factors including the length of lockup period, VC backing, managerial stock option, growth opportunities of IPOs, technology intensity, and selection of growth stock exchange (i.e. AIM). Finally, I find that the IPOs issued in the years when the market put a price premium on dividend paying payers are more likely to pay dividend after IPO and initiate dividends earlier. Overall, my results show that IPO characteristics relate to dividend decisions of IPOs through miscellaneous mechanisms of dividends. The most homogeneous results are associated with the life cycle and catering theories. There is also some empirical evidence in support of signaling and agency theory. The second empirical study examines the determinants on the dividend policies stated in IPO prospectuses. At the stage of preparing for IPO, pre-IPO financial status is very likely to influence the initial dividend policies. My results provide strong evidence that IPOs that experienced superior performance in profitability and cash inflow from operating activities during pre-IPO period tend to make active dividend policies relatively, consistent with the implication of Lintner (1956) and Benartzi, Michaely and Thaler (1997). My results also show that IPOs with higher turnover ratio and lower capital expenditures tend to choose more active dividend policies when going public, consistent with residual theory and free cash flow hypothesis. In addition, the possibility of choosing relatively active dividend strategies at IPO stage is negatively associated with VC backing, length of full lock-up restriction period, stock option, technology focus, and institutional ownership. In contrast, IPOs with more reputable underwriters tend to declare relatively active dividend policy in prospectuses. The evidence relating to long-term debt ratio and managerial ownership is weak. Moreover, IPOs issued in the ‘internet bubble’ period or in 2000s opt for relatively conservative dividend strategies. The overall results in this empirical chapter support lifecycle theory, substitution assumption-based agency theory and free cash flow hypothesis, while the evidence on signaling and catering theories is mixed. Furthermore, my results support the conjecture that IPOs with active dividend policies release sufficient information through dividend policies declared in offering prospectuses and therefore their formal dividend initiations fail to shock the market. I find that dividend- paying companies outperform non-dividend paying counterparts during three post-IPO years, indicating that non-dividend initiating IPOs rather than dividend-initiating ones account for the decline in long-run underperformance. Additionally, I find evidence in support of the conjecture that the dividend policies stated in prospectuses communicate the information, and thus reduce the possibilities that outside investors are overoptimistic over the prospect of the invested companies and that managers overstate the pre-IPO financial data at IPO stage. The third empirical study examines the trends in dividend policies across seven western countries: U.S., Candada, U.K., Germany, France, Japan and Hong Kong. In general, the proportion of dividend paying firms fell significantly from 1989 through to the early 2000s, with the exception of Japanese firms. Thereafter, the percentage reverted slightly in the US, Canada, Japan and in Hong Kong, but continued to decrease in UK, France, and Germany. In contrast, the aggregate amount of dividends increased continuously across countries and firms retained stable dividend payout ratios, and total payout ratios relatively. Share repurchases took over from dividends as the dominant payout method in the US and the increasing importance of repurchases is observed in Canada and in the UK as well. A declining propensity to pay dividends is seen in all the sample countries apart from in Japan, controlling for key firm characteristics. I find that the likelihood that firms payout dividends or repurchase shares positively correlates with firm’s size, profitability and the ratio of earned/contributed capital, and negatively related to long-term debt ratio. The impact of growth opportunities on payout decisions is not uniform across countries, in line with Denis and Osobov (2008). There is some evidence that cash holdings have a negative relation with the probability of paying dividends and a positive relation with the probability of buying back shares. There is also some evidence that R&D expenditure and technology intensity have a negative influence on a firm’s tendency to pay dividends, but such influence is country-dependent. The effect of M&A on the incidence of payouts is highly country-dependant. For example, US acquirers are reluctant to pay dividends while UK acquirers are more likely to pay dividends. I also examine the determinants of the amounts of corporate payouts. Profitability, growth rate of total assets, and retained earnings are important positive factors in determining dividend amounts. Market to book ratio have a significantly positive effect on both dividend amounts and the repurchase amounts, consistent with Lee and Suh (2011), Alzahrani and Lasfer (2012). Finally, the empirical tests using Lintner model indicate that the link between cash dividends and earnings has weakened, in support of Choe (1990) and Brav, Graham, Harvey, and Michaely (2005). In line with Eije and Megginson (2008), the data demonstrates that dividends are still responsive to earnings. Overall, the evidence in this empirical chapter supports agency cost-based lifecycle theory.

This study looks at some factors influencing mutual fund performance. Fund management location, family status and asset allocation and timing ability are examined. Using monthly returns on 4545 funds from Morningstar from January 1970 to June 2010, the study examines whether location influences the return a fund generates. It is found that U.S. managed funds outperform European managed funds, regardless of market invested in. This can be seen in terms of higher mean alpha, and statistically significant outperformance. A comparison is also carried out between the performance of family funds and non-family funds. Using the recursive portfolio technique and Rhodes utility based measure of persistence, the persistence of funds that are in a family are compared to those that do not belong to a family. A second hypothesis is also examined here, analyzing whether fund managers make their risk decision to influence performance for the second part of the year based on their performance in the first part of the year. It can be concluded that family status, family size or market does not affect persistence in performance. The study found that family rank has an impact on the risk adjustment behaviour of fund managers. The fact that the coefficient is negative suggests that managers are not behaving strategically. When markets are examined individually, fund managers within families compete in the U.S. and behave strategically in Europe. Finally, using asset allocation data on balanced funds, the study examines the skill of balanced fund managers to time particular asset classes. It is found that there is little timing ability present, across all markets and models.

This thesis proposes theoretical models to study bank risk-taking and financial stability. Three issues are explored: (1) the moral-hazard incentive for securitisation, (2) the socially optimal banking structure for the economy, and (3) the relationship between bank competition and financial stability, based on bank funding structures and fire-sale risks. Chapter 2 proposes a model to study how bank securitisation affects the value of bank equity, and hence what leads a bank to securitise its assets. The proposed model shows that moral hazard (which is induced by the deposit insurance scheme), can be one essential motive for the securitisation of deposit-taking commercial banks. This chapter also discusses some factors that can restrain the moral-hazard and risk-taking behaviour in bank securitisation. Chapter 3 investigates the social value of different banking structures. The proposed model finds that total separation is not the optimal banking structure for an economy, because it forbids the liquidity transfer between subsidiary banks, which is socially valuable. The comparison between ring-fencing and universal banking is more complicated; Chapter 3 shows that whether ring-fencing or universal banking is the best banking structure for an economy depends on the returns to the different subsidiary banking sectors. Chapter 4 studies how asset fire-sales risks and bank funding structures can affect the relationship between bank competition and financial stability. The proposed model finds that the funding-structure risks of the banks can create an incentive for excess risk-taking in a multi-bank economy. Moreover, the model shows that the excessive risk taking increases with the number of banks in the economy. This result is similar in spirit to the Cournot equilibrium in standard microeconomic theory.

The thesis focuses on risk measures used to calculate solvency capital requirements. It consists of three independent papers. The first paper (Chapter 2) investigates time-consistency, the relation that should hold across risk measurements of the same financial position at different time points. Sufficient conditions are provided for coherent risk measures, in order to satisfy the requirements of acceptance-, rejection- and sequential consistency. It is shown that risk measures used in practice usually do not satisfy these requirements. Hence a method is provided to systematically construct sequentially consistent risk measures. It is also emphasized that current approaches to dynamic risk measurement do not consider that risk measures at different time points have different arguments. Here we briefly discuss this new setting highlighting that the notions of time consistency presented in the literature need to be reinterpreted. The second and third papers (Chapters 3 and 4) consider respectively the risk arising from parameter and model mis-specification due to estimation from a limited amount of available data. This risk may have a substantial impact on risk measures used to quantify solvency capital requirements. We introduce a new method to quantify this impact measured as the additional capital needed to allow for randomness in the data sample used for the estimation procedure. This level of capital we call residual estimation risk. In the second paper, for parameter uncertainty we prove the effectiveness of three approaches for reducing residual estimation risk in the case of location-scale families. These are based on (a) raising the capital requirement by adjusting the risk measure, (b) Bayesian predictive distributions under probability-matching priors and (c) residual risk estimation via parametric bootstrap. Risk measures satisfying standard properties are used, for example the popular TVaR. For more general distributions only (a) and (b) are investigated and a truncated version of TVaR is used. Numerical results obtained via Monte-Carlo simulation demonstrate that the proposed methods perform well. In the third paper (Chapter 4), we compare the effectiveness of four different approaches to estimate capital requirements in the presence of model uncertainty. For a given set of candidate models the model posterior weights can be obtained via a Bayesian approach. Then we consider approaches based on: (a) worst case scenario, (b) highest model posterior, (c) averaging the capital under each model according to the model posterior weights and (d) determining the predictive distribution of the financial loss and using it to calculate the capital. It is shown that all these methods are very sensitive to the set of candidate models specified. If this has been carefully selected (for instance via expert judgement) the approach based on the highest posterior performs slightly better than the others. Alternatively, if there is poor prior information on the model set the effectiveness of all these approaches decreases substantially. In particular, the worst case approach has a very low performance. It also emerges that mis-specifiying the model by using distributions that are more heavy-tailed than the one generating the data, may reduce the capital and thus it is not a conservative approach.

This thesis considers specific topics related to the dynamic modelling and management of risk, with a particular emphasis on the generation of asymmetric and fat tailed behavior observed in practise. Specifically, extensions to the dynamics of the popular GARCH model, to capture time variation in higher moments, are considered in the univariate and multivariate context, with a special focus on the Generalized Hyperbolic distribution. In Chapter 1, I consider the extension of univariate GARCH processes with higher moment dynamics based on the Autoregressive Conditional Density model of Hansen (1994), with conditional distribution the Generalized Hyperbolic. The value of such dynamics are analyzed in the context of risk management, and the question of ignoring them discussed. In Chapter 2, I review some popular multivariate GARCH models with a particular emphasis on the dynamic correlation model of Engle (2002), and alternative distributions such those from the Generalized Asymmetric Laplace of Kotz, Kozubowski, and Podgorski (2001). In Chapter 3, I propose a multivariate extension to the Autoregressive Conditional Density model via the independence framework of the Generalized Orthogonal GARCH models, providing the first feasible model for large dimensional multivariate modelling of time varying higher moments. A comprehensive out-of- sample risk and portfolio management application provides strong evidence of the improvement over non time varying higher moments. Finally, in Chapter 4, I consider the benefits of active investing when the benchmark index is not optimally weighted. I investigate advances in the definition and use of risk measures in portfolio allocation, and propose certain simple solutions to challenges arising in the optimization of these measures. Combining the models discussed in the previous chapters, within a fractional programming optimization framework and using a range of popular risk measures, a large scale out-of-sample portfolio application on the point in time constituents of the Dow Jones Industrial Average is presented and discussed, with clear implications for active investing and benchmark policy choice.

No description available.

Petroleum markets are undergoing rapid financialization and integration, leading to increased volatility and exposing participants to potentially much greater risks. This thesis addresses the explicit modeling of petroleum price volatility in a multivariate framework and analyzes the relative merits of multivariate models to describe change in the context of petroleum markets risk. The focus of this thesis will be on explaining the dynamic interdependencies in petroleum markets and further demonstrate whether the existence of such interdependencies prompt for the need to assess risk differently, by which this thesis contributes to the existing economic or econometric theories in three aspects. The first empirical part examines the importance of volatility spillovers and asymmetry in petroleum markets and their influence on optimal hedging strategy. To address in a realistic way the dynamic conditional correlation of petroleum spot and futures markets, we develop a new theoretical framework by accounting for the effect of time-varying conditional correlations in the conditional volatility processes of the VARMA-AGARCH model in what is termed the VARMA-AGARCH-DCC model. Results demonstrate that the proposed model is the best for OHR calculation in terms of the variance of portfolio reduction and tail risk analysis. The second empirical part, for the first time in the literature of energy economics, examines the volatility and correlation interdependence between oil market and China stock market at the sector-level. Results indicate that oil price fluctuations constitute a systematic asset price risk at the sector level and information content embedded in oil market volatility is an effective and valuable variable for constructing an optimal oil-stock holding. Finally, the third empirical part, for the first time in the literature of energy economics, investigates the volatility transmission mechanism among three benchmark oil markets and quantifies the size and persistence of these connections through employing the Volatility Impulse Response Function (VIRF) methodology. Results suggest markedly different responsiveness to historical events and volatility/correlation dynamics across crude oil benchmark markets. Overall, the findings of this thesis have important implications for crude oil market trading and risk management, as well as stock market investors, by providing valuable information on the oil price volatility dynamics and will help market participants develop efficient risk measurement schemes and devise sound risk management strategies.