Graduation date: 1971

This dissertation contains two essays on the analysis of market imperfections. In the first essay, I empirically test whether in a three-level hierarchy with asymmetries of information, more competition among intermediaries leads to more deception against the principal. In this setting, intermediaries supervise agents by delegation of the principal, and compete among themselves to provide supervision services to the agents. They cannot be perfectly monitored, therefore allowing them to manipulate supervision results in favor of the agents, and potentially leading to less than optimal outcomes for the principal. Using inspection-level data from the vehicular inspection program in Atlanta, I test for the existence of inspection deception (false positives), and whether this incidence is a function of the number of local competitors by station. I estimate the incidence of the most common form of false positives (clean piping) to be 9% of the passing inspections during the sample period. Moreover, the incidence of clean piping -- passing results of a different vehicle fraudulently applied to a failing vehicle -- per station increases by 0.7% with one more competitor within a 0.5 mile radius. These results are consistent with the presence of more competitors exacerbating the perverse incentives introduced by competition under this setting. In the second essay, we test whether electricity consumption by industrial and commercial customers responds to real-time prices after these firms sign-up for prices linked to the electricity wholesale market price. In principle, time-varying prices (TVP) can mitigate market power in wholesale markets and promote the integration of intermittent generation sources such as wind and solar power. However, little is known about the prevalence of TVP, especially in deregulated retail markets where customers can choose whether to adopt TVP, and how these firms change their consumption after signing up for this type of tariff. We study firm-level data on commercial and industrial customers in Texas, and estimate the magnitude of demand responsiveness using demand equations that consider the restrictions imposed by the microeconomic theory. We find a meaningful level of take-up of TVP ? in some sectors more than one-quarter of customers signed up for TVP. Nevertheless, the estimated price responsiveness of consumption is still small. Estimations by size and by type of industry show that own price elasticities are in most cases below 0.01 in absolute value. In the only cases that own price elasticities reach 0.02 in absolute value, the magnitude of demand response compared to the aggregate demand is negligible.

Typescript. / Thesis (Ph. D.)--University of Hawaii at Manoa, 1989. / Includes bibliographical references (leaves 149-156) / Microfiche. / x, 156 leaves, bound ill. 29 cm

Typescript. / Thesis (Ph. D.)--University of Hawaii at Manoa, 1989. / Includes bibliographical references (leaves [127]-130). / Microfiche. / ix, 130 leaves, bound ill. 29 cm

Traditional macroeconometric models of the Australian economy estimate the behaviour of wage and price inflation separately, thereby ignoring the possibility that there is a contemporaneous relationship between these two variables. This thesis follows a recent trend emerging in other small open economies, such as the UK and Norway, which is to estimate the behaviour of wage and price inflation in a simultaneous-equations model. In order to capture the behaviour of the major variables which drive wages and prices, a complete model is constructed which embeds these important transmission channels. The model is developed in three stages. First, underpinned by a theoretical framework of a unionized economy with imperfect competition, the core wage- price system is developed whereby consumer prices and average weekly earnings are jointly estimated in a simultaneous-equations framework. Particular atten- tion is given to estimating two identified cointegrating relationships for wages and prices. These equations are interpreted as the long-run targets of workers and firms respectively and are embedded in a parsimonious system of short-run dynamics which drive wages and prices towards their long-run levels. Second, llie behaviour of llie main feedback variables driving llie wage-price system is modelled, with particular attention given to the unemployment rate. While several of the most recent models of unemployment show that the aggregate unemployment rate in Australia does indeed behave differently during periods of low and high unemployment, none can explain what drives the unemployment rate to increase at such a rapid rate and what contributes to its much slower decrease. Another central issue of this thesis, therefore, is to propose a rationale for this as yet unexplained phenomenon. The remaining behavioural variables in the model, including aggregate labour productivity, domestic output and the real exchange rate, are all estimated in a single-equation framework. Third, these equations are then combined with a number of important identi- ties and an interest-rate reaction function to close the model. Then, the impacts of several simulated economic scenarios on Australia's economic landscape are considered. Special emphasis is given to analysing the impact of a large nomi- nal wage shock. The outcomes from these simulated scenarios are pertinent to understanding the inflation process and have important implications for a small open economy like Australia with an explicit inflation target. Overall, the major result to emerge from this thesis is that there is significant statistical support for the hypothesis that wage and price inflation in Australia are jointly determined. This phenomenon has not yet been fully exploited in current macroeconometric models of the Australian economy. The modelling exercise also reveals that the Australian unemployment rate is linear in demand and labour productivity shocks, with nonlinear behaviour caused by real wage rigidity and generous unemployment benefits. Importantly, this simple model is able to simulate the behaviour of the Australian economy extremely well. The outcome from the policy scenarios is clear: both demand-side and supply-side shocks have real and nominal effects on the economy in the short- to medium- run, ceteris paribus. Moreover, a large nominal wage shock to the economy, which results in a real wage rise, will have no sustained effect on the level of domestic activity in the economy, the inflation rate or the real exchange rate. Unemployment is, however, pushed slightly above equilibrium in the short- to medium-run due to a sustained higher real wage level.

Developing timely and reliable house price indices is of interest worldwide, because these measures influence consumer behaviour, inflation targeting, and spot and futures markets. Several techniques for constructing a constant quality price index are available in the literature, but these methods are difficult to apply in localities where market transaction data is limited. Since house price movements are a local phenomena, improving the timeliness of a quality controlled price index at local housing market levels in small countries like New Zealand is a challenge. This thesis comprises three essays that focused on improving the timeliness of reported house price indices at the local market levels. The timeliness issue examined in this thesis has not previously been rigorously investigated and this makes the results of this thesis both important and unique for the benefit of both academic research and practical application. Essay One reviews the sale price appraisal ratio (SPAR) method, which has been applied since the 1960s for producing local house price indices at a semi-annual and quarterly basis in New Zealand. Utilizing a variety of statistical tests and comparing this index with the repeat sales and median price index result in the study highlighting the potential of, as well as the problems associated with, a price index produced by the SPAR method at a monthly level. In the following two essays, monthly price indices are tested using empirical real estate research methods in order to examine their usefulness in exploring the research questions as well as revealing the statistical differences between them. Essay Two studies the relationship between sale price and trading volume, and the ripple effect of local house price comovements. The results show that the trading volume generally leads the sale price in the long-run and the ripple effect is most likely constrained within regions. In Essay Two, the monthly SPAR index produces similar statistical results to those estimated by the repeat sales index for large cities. Essay Three is a study on the market efficiency of housing markets. It is found the local housing market is neither weak-form nor semi-strong form efficient. Local house price movements are strongly correlated and are mean reverting towards their long-run equilibrium. It is further concluded that monthly price indices for small cities are problematic due to the problem of small sample size. Overall, the findings in this thesis show monthly house price indices can be generated by using the SPAR method at local market levels. However, this potential is limited to large cities. Further research can focus on improving the quality of monthly price indices for large cities.

This thesis investigates implications of interdependence between stock market prices in the context of several financial applications including: portfolio selection, tests of market efficiency and measuring the extent of integration among national stock markets. In Chapter 2, I note that volatility spillovers (transmissions of risk) have been found in numerous empirical studies but that no one, to my knowledge, has evaluated their effects in the general portfolio framework. I dynamically forecast two multivariate GARCH models, one that accounts for volatility spillovers and one that does not, and construct optimal mean-variance portfolios using these two alternative models. I show that accounting for volatility spillovers lowers portfolio risk with statistical significance and that risk-averse investors would prefer realised returns from portfolios based on the volatility spillover model. In Chapter 3, I develop a structural MGARCH model that parsimoniously specifies the conditional covariance matrix and provides an identification framework. Using the model to investigate interdependencies between size-sorted portfolios from the Australian Stock Exchange, I gain new insights into the issue of asymmetric dependence. My findings not only confirm the observation that small stocks partially adjust to market-wide news embedded in the returns to large firms but also present evidence that suggests that small firms in Australia fail to even partially adjust (with statistical significance) to large firms??? shocks contemporaneously. All adjustments in small capitalisation stocks occur with a lag. Chapter 4 uses intra-daily data and develops a new method for measuring the extent of stock market integration that takes into account non-instantaneous adjustments to overnight news. This approach establishes the amounts of time that the New York, Tokyo and London stock markets take to fully adjust to overnight news and then uses this This thesis investigates implications of interdependence between stock market prices in the context of several financial applications including: portfolio selection, tests of market efficiency and measuring the extent of integration among national stock markets. In Chapter 2, I note that volatility spillovers (transmissions of risk) have been found in numerous empirical studies but that no one, to my knowledge, has evaluated their effects in the general portfolio framework. I dynamically forecast two multivariate GARCH models, one that accounts for volatility spillovers and one that does not, and construct optimal mean-variance portfolios using these two alternative models. I show that accounting for volatility spillovers lowers portfolio risk with statistical significance and that risk-averse investors would prefer realised returns from portfolios based on the volatility spillover model. In Chapter 3, I develop a structural MGARCH model that parsimoniously specifies the conditional covariance matrix and provides an identification framework. Using the model to investigate interdependencies between size-sorted portfolios from the Australian Stock Exchange, I gain new insights into the issue of asymmetric dependence. My findings not only confirm the observation that small stocks partially adjust to market-wide news embedded in the returns to large firms but also present evidence that suggests that small firms in Australia fail to even partially adjust (with statistical significance) to large firms??? shocks contemporaneously. All adjustments in small capitalisation stocks occur with a lag. Chapter 4 uses intra-daily data and develops a new method for measuring the extent of stock market integration that takes into account non-instantaneous adjustments to overnight news. This approach establishes the amounts of time that the New York, Tokyo and London stock markets take to fully adjust to overnight news and then uses this

In this thesis I examine, within a behavioural finance framework, the impact on stock prices of order and trade imbalance in three separate but related studies. The first study, chapter two, begins with a question that plagues behavioural finance theories???do the investors most likely to be influenced by the behavioural biases described in the literature, i.e., individual investors, affect stock prices? My data enable me to consider the impact of net individual investor trading for the entire market over several years. I find that net individual investor purchasing Grangercauses stock price changes. The correlation is negative, however, contradicting common sense by demonstrating that individuals investor buying pressure makes prices go down and selling pressure forces them up. More investigation is required. Chapter three references order imbalance results from experimental finance. I use field data to test a robust laboratory model and my modified versions. My findings suggest that, with appropriate modifications, laboratory results can be applied to real financial markets. Chapter four combines the data from the chapters two and three to revisit the question of individual investor impact on stock prices. Other studies have argued that individual investor influence is strongest in smaller capitalization stocks. Moreover, various theories propose that individual investors are the driving force behind the irrational stock prices of a bubble. I focus on the stocks from chapter three, bubble stocks, and ask whether, in the context of the trading of the entire market, individual investor trades are influential. Once again I find Grangercausality, but in the wrong direction. Moreover, the activity and volume of the individual investor category of the holdings data is completely overshadowed by that of the two large investor categories, domestic and foreign institutions. I conclude that individual investor trades are not influential in determining stock prices. This conclusion has important implications for some behavioural finance models of asset pricing. I suggest that emphasis might be better placed on educating individual investors about the errors to which they are prone, rather than on trying to explain market anomalies with those errors.

This thesis presents three related essays on the dynamic relationship between different types of investment flow and prices in the equity market. These studies attempt to provide greater insight into the evolution of prices by investigating not ???what moves prices??? but ???who moves prices??? by utilising a unique database from the Korean Stock Exchange. The first essay investigates the trading behaviour and performance of online equity investors in comparison to other investors on the Korean stock market. Whilst the usage of online resources for trading is becoming more and more prevalent in financial markets, the literature on the role of online investors and their impact on prices is limited. The main finding arising from this essay supports the claim that online investors are noise traders at an aggregate level. Whereas foreigners show distinct trading patterns as a group in terms of consensus on the direction of market movements, online investors do not show such distinct trading patterns. The essay concludes that online investors do not trade on clear information signals and introduce noise into the market. Direct performance and market timing ability measures further show that online investors are the worst performers and market timers whereas foreign investors consistently show outstanding performance and market timing ability. Domestic mutual funds in Korea have not been extensively researched. The second essay analyses mutual fund activity and relations between stock market returns and mutual fund flows in Korea. Although regulatory authorities have been cautious about introducing competing funds, contractual-type mutual funds have not been cannibalized by the US-style corporate mutual funds that started trading in 1998. Negative feedback trading activity is observed between stock market returns and mutual fund flows, measured as net trading volumes using stock purchases and sales volume. It is predominantly returns that drive flows, although stock purchases contain information about returns, partially supporting the price pressure hypothesis. After controlling for declining markets, the results suggest Korean equity fund managers tend to swing indiscriminately between increasing purchases and increasing sales in times of rising market volatility, possibly viewing volatility as an opportunity to profit and defying the mean-variance framework that predicts investors should retract from the market as volatility increases. Mutual funds respond indifferently to wide dispersions in investor beliefs. The third essay focuses on the conflicting issue of home bias by looking at the impact on domestic prices of foreign trades relative to locals using high frequency data from the Korean Stock Exchange (KSE). This essay extends the work of Choe, Kho and Stulz (2004) (CKS) in three ways. First, it analyses the post-Asian financial crisis period, whereas CKS (2004) analyse the crisis (1996-98) period. Second, this essay adopts a modified version of the CKS method to better capture the aggregate behaviour of each investor-type by utilising the participation ratio in comparison to the CKS method. Third, this essay does not limit investigation to intra-day analysis but extends to daily analysis up to 50 days to observe the effect of intensive trading activity in a longer horizon than the CKS study. In contrast to the CKS findings, this paper finds that foreigners have a short-lived private information advantage over locals and trades by foreigners have a larger impact on prices using intra-day data. However, assuming investors buy-hold for up to 50 days, the local individuals provide a greater impact and more profitable returns than foreigners. Superior performance is documented for buys rather than sells.

This thesis examines the empirical performance of four Affine Jump Diffusion models in pricing and hedging S&P 500 Index options: the Black Scholes (BS) model, Heston???s Stochastic Volatility (SV) model, a Stochastic Volatility Price Jump (SVJ) model and a Stochastic Volatility Price-Volatility Jump (SVJJ) model. The SVJJ model structure allows for simultaneous jumps in price and volatility processes, with correlated jump size distributions. To the best of our knowledge this is the first empirical study to test the hedging performance of the SVJJ model. As part of our research we derive the SVJJ model minimum variance hedge ratio. We find the SVJ model displays the best price prediction. The SV model lacks the structural complexity to eliminate Black Scholes pricing biases, whereas our results indicate the SVJJ model suffers from overfitting. Despite significant evidence from in and out-of-sample pricing that the SV and SVJ models were better specified than the BS model, this did not result in an improvement in dynamic hedging performance. Overall the BS delta hedge and SV minimum variance hedge produced the lowest errors, although their performance across moneyness-maturity categories differed greatly. The SVJ model???s results were surprisingly poor given its superior performance in out-of-sample pricing. We attribute the inadequate performance of the jump models to the lower hedging ratios these models provided, which may be a result of the negative expected jump sizes.