City Research Online

Abstract

The functioning of financial markets is influenced by information and investor learning. In recent years, the increasing availability of media, big data, and regulatory requirements for enhanced disclosure have provided investors with access to more information than ever before. While this greater supply of information has the potential to improve the quality of financial markets by increasing the price informativeness, it can also create challenges and friction in the learning process for investors. Moreover, the strategic manipulation of information disclosure by agencies due to agency issues presents a further concern. Therefore, it is critical to gain a thorough understanding of the mechanisms by which information affects financial markets. Such comprehension is vital for both investors and policymakers to make informed decisions.

This thesis is to investigate the role of information learning in financial markets, with a specific emphasis on the disclosure practices of mutual funds. Chapter 1 investigates the effects of “correlation neglect” in financial markets, where naive traders neglect the correlation between signal errors. Using a model with both naive and rational traders, the study finds that the impact of naive traders on market quality, as measured by liquidity and mispricing risk, depends on the cost of obtaining information. When information is free and the correlation between signal errors is low, the presence of naive traders can reduce mispricing risk. However, when the correlation is high, mispricing risk becomes U-shaped. When information is costly, market liquidity deteriorates and mispricing risk increases with an increase of naive traders. However, market quality can improve when informed rational traders are driven out of the market by the large mass of naive traders.

In Chapter 2, we argue that highly complex funds’ prospectuses limit the ability of investors to effectively use available information and make informed investment decisions. Measuring textual complexity with the Fog Index, our evidence suggests that low-quality funds manipulate their prospectuses, making them more complex, possibly targeting less sophisticated investors. These investors, in turn, use a less sophisticated asset pricing model to evaluate fund performance, react more aggressively to past winners, and are more likely to be attracted by funds with high marketing costs. Our results suggest that funds with low-complexity prospectuses are more trustworthy, and that funds with high-complexity prospectuses are possibly subject to more severe agency issues.

In Chapter 3, I investigate ESG risk disclosures by mutual funds when investors learn from their disclosures in addition to past performance. Using a novel natural language processing method to identify ESG-risk disclosure in mutual fund prospectuses, I find that funds with higher ESG risk are more likely to disclose ESG risk than equivalent funds with lower ESG risk. To understand this, I develop a theoretical model which illustrates how ESG risk disclosure reduces investor reliance on past returns, thereby moderating flow performance sensitivity and smoothing fund fee income. I also show that the key predictions of the model hold in practice when I empirically test the model using U.S. mutual fund data. My results suggest that ESG risk disclosure can be used for risk management purposes to mitigate the adverse effects of high ESG risk exposure.

Publication Type:	Thesis (Doctoral)
Subjects:	H Social Sciences > HG Finance
Departments:	Bayes Business School > Bayes Business School Doctoral Theses Bayes Business School > Finance Doctoral Theses

Export

Downloads