Thesis advisor: Pierluigi Balduzzi / Essay one of this dissertation investigates the relation between fundamental idiosyncratic volatility and stock returns idiosyncratic volatility using data from 56 countries over 1980-2014. I find a strong positive relation between fundamental idiosyncratic volatility and idiosyncratic volatility of returns. This association, however, seems to be entirely driven by the developed economies and I find no effect in the emerging markets. Specifically, fundamental idiosyncratic volatility does not lead to more idiosyncratic return volatility in countries with poor legal institutions and weak shareholder protection laws. The second essay examines the effect of accounting standards on return predictability by using a variance decomposition approach, and is joint work with Pierluigi Balduzzi, Gil Sadka, and Ronnie Sadka. We decompose returns into a cash-flow news component and a discount-rate news component, and investigate cross-sectional and time-series changes in the contribution of each component to return variations. We also decompose returns for 20 industries in three subsample periods to examine the effect of accounting standards on different industries over time. Our results contribute to our understanding of the effect of accounting practices on accounting variables and return predictability. The third essay studies the effect of short-selling on stock price informativeness. Morck, Yeung, and Yu (2000), in their pioneering study of international differences in stock price synchronicity, emphasize the effect of market development on the ability of investors to incorporate firm-specific information into prices. I use a unique institutional feature in the Hong Kong market to investigate one of the important tools investors use to incorporate information into prices and hence, reduce stock price synchronicity; short-selling. Examining the cross-sectional and time-series variation in short-sale constraints in the Hong Kong market, I find that following the removal of short-sale constraints, stock prices become more informative and move less in tandem with the market. My findings contribute to our understanding of the impact of short-sales constraints on stock price informativeness. / Thesis (PhD) — Boston College, 2016. / Submitted to: Boston College. Carroll School of Management. / Discipline: Finance.

none

The objective of this thesis is to explain the behaviour of stock returns around the disclosure of different types of information release by UK companies. Previous literature has documented the existence of both market anticipation and the lagged impounding of value relevant information. The main objective of this research is, therefore, to identify the conditions under which investors choose to be informed in anticipation of and in response to, a corporate disclosure. More specifically, we explain the behaviour of stock returns in terms of the costs and benefits which investors must consider when deciding whether to acquire and interpret information. The results indicate that market anticipation is an increasing function of firm size, the number of years a firm has been trading and the volatility of prior stock returns. However, increased voluntary disclosure by firms would appear to reduce the ability of investors to and anticipate and interpret information. The volatility of stock returns, prior to the disclosure, is nevertheless the main driving force behind the explanation of post-announcement drift. There are also indications that investors' initial reactions to both earnings and non-earnings news are not based on informed judgements, and that bad news is generally associated with greater uncertainty than good news. Bad news would appear to be more difficult to anticipate and interpret, relative to good news. On further examination, however, investor anticipation is shown to be largely based on information as opposed to uninformed trading.

The primary aim of this study is to investigate the stock return volatility surrounding management earnings forecasts. Disclosure by managers of expected earnings are particularly important communications, and as such, it is important to understand the capital market implications surrounding them. In doing so, the research questions are essentially aimed at examining the stock return volatility, first, at the release of a management earnings forecast, and second, at the eventual announcement of the realised earnings for that period. The first test investigates whether there is an increase in volatility surrounding a management earnings forecast for those firms who release them compared to a matched-firm sample of firms without a management earnings forecast at that date, and then further examines that result based on different forecast antecedents and forecast characteristics. Next, this study tests, for firms who do release a management earnings forecast during the year, whether stock volatility is lower than firms who do not release a management earnings forecast at the eventual earnings announcement date. In brief, the evidence using the Garman and Klass [1980] ???best analytic scale-invariant estimator??? of volatility in an Australian context, between 1993 and 2003, finds that stock return volatility is greater for bad news forecasts, forecasts of low specificity, and forecasts issued by firms perceived ex ante as being of lower credibility using both permutation analysis and modelling daily volatility. At the earnings announcement date, however, there is no evidence that stock return volatility is lower for firms that issue management earnings forecasts during the year. Overall, this result challenges the information asymmetry argument in the literature that disclosure will reduce volatility in the long-run.

This dissertation consists of three stand-alone essays on fundamental uncertainty, stock return volatility and earnings management. The first study investigates the role of information about firms?? fundamentals contained in analysts?? forecasts (which I label ??non-accounting information??) in understanding stock return volatility. When combined with Ohlson??s (1995) linear information dynamics, the accounting version of the Campbell-Shiller model (Campbell and Shiller 1988a, 1988b; Vuolteenaho 2002) implies that if current non-accounting information is more uncertain, then future stock returns are expected to be more volatile. The empirical evidence supports the theoretical predictions, and the results are valid for measures of both systematic and idiosyncratic volatility. Additional analysis yields some evidence that both favourable and unfavourable news from non-accounting information increases future stock return volatility. Overall, the results highlight the value relevance of information in analysts?? forecasts beyond what is contained in the current financial statements. The second essay extends the theoretical framework of Callen and Segal (2004) and Vuolteenaho (2002) to investigate the association between the uncertainty of accrual information and stock return volatility. The empirical evidence supports the theoretical prediction that the extent of uncertainty in accounting accruals is increasing with the volatility of future stock returns, and the results are valid for measures of both systematic and idiosyncratic volatility. However, when accrual variability is decomposed into fundamental and unexpected portions, I find that the positive relationship between accrual variability and future stock return volatility is dominated by the fundamental component of accrual variability. The findings therefore suggest that the market places little weight on information conveyed by that component of accounting accruals that is most likely to reflect accounting choices, implementation decisions and managerial opportunism. The final essay argues that the presumed articulation among accruals, cash flows and revenues does not capture decisions on expected accruals when large external financing activities are present. The analysis provides evidence that managers?? ??normal?? operating decisions associated with net external financing activities are likely to lead to measurement errors in unexpected accruals that are part of expected accruals, and erroneous conclusions that significant earnings management exists when in fact there is none. This is especially pertinent in cases where the partitioning variable used to identify instances of earnings management is supposed to be uncorrelated with external financing, when in fact it is correlated. The results underscore the importance of additional specification tests being conducted to control for estimation biases in unexpected accruals associated with external financing. I suggest the use of matched-firm approach using industry and external financing matches in order that reliable and warranted inferences are made.

This dissertation consists of three stand-alone essays on fundamental uncertainty, stock return volatility and earnings management. The first study investigates the role of information about firms?? fundamentals contained in analysts?? forecasts (which I label ??non-accounting information??) in understanding stock return volatility. When combined with Ohlson??s (1995) linear information dynamics, the accounting version of the Campbell-Shiller model (Campbell and Shiller 1988a, 1988b; Vuolteenaho 2002) implies that if current non-accounting information is more uncertain, then future stock returns are expected to be more volatile. The empirical evidence supports the theoretical predictions, and the results are valid for measures of both systematic and idiosyncratic volatility. Additional analysis yields some evidence that both favourable and unfavourable news from non-accounting information increases future stock return volatility. Overall, the results highlight the value relevance of information in analysts?? forecasts beyond what is contained in the current financial statements. The second essay extends the theoretical framework of Callen and Segal (2004) and Vuolteenaho (2002) to investigate the association between the uncertainty of accrual information and stock return volatility. The empirical evidence supports the theoretical prediction that the extent of uncertainty in accounting accruals is increasing with the volatility of future stock returns, and the results are valid for measures of both systematic and idiosyncratic volatility. However, when accrual variability is decomposed into fundamental and unexpected portions, I find that the positive relationship between accrual variability and future stock return volatility is dominated by the fundamental component of accrual variability. The findings therefore suggest that the market places little weight on information conveyed by that component of accounting accruals that is most likely to reflect accounting choices, implementation decisions and managerial opportunism. The final essay argues that the presumed articulation among accruals, cash flows and revenues does not capture decisions on expected accruals when large external financing activities are present. The analysis provides evidence that managers?? ??normal?? operating decisions associated with net external financing activities are likely to lead to measurement errors in unexpected accruals that are part of expected accruals, and erroneous conclusions that significant earnings management exists when in fact there is none. This is especially pertinent in cases where the partitioning variable used to identify instances of earnings management is supposed to be uncorrelated with external financing, when in fact it is correlated. The results underscore the importance of additional specification tests being conducted to control for estimation biases in unexpected accruals associated with external financing. I suggest the use of matched-firm approach using industry and external financing matches in order that reliable and warranted inferences are made.

A key prediction of the Capital Asset Pricing Model (CAPM) is that idiosyncratic risk is not priced by investors because in the absence of frictions it can be fully diversified away. In the presence of constraints on diversification, refinements of the CAPM conclude that the part of idiosyncratic risk that is not diversified should be priced. Recent empirical studies yielded mixed evidence with some studies finding positive correlation between idiosyncratic risk and stock returns, while other studies reported none or even negative correlation. In this thesis we revisit the problem whether idiosyncratic risk is priced by the stock market and what the probable causes for the mixed evidence produced by other studies, using monthly data for the US market covering the period from 1980 until 2013. We find that one-period volatility forecasts are not significantly correlated with stock returns. On the other hand, the mean-reverting unconditional volatility is a robust predictor of returns. Consistent with economic theory, the size of the premium depends on the degree of 'knowledge' of the security among market participants. In particular, the premium for Nasdaq-traded stocks is higher than that for NYSE and Amex stocks. We also find stronger correlation between idiosyncratic risk and returns during recessions, which may suggest interaction of risk premium with decreased risk tolerance or other investment considerations like flight to safety or liquidity requirements. The difference between the correlations between the idiosyncratic volatility estimators used by other studies and the true risk metric - the mean-reverting volatility - is the likely cause for the mixed evidence produced by other studies. Our results are robust with respect to liquidity, momentum, return reversals, unadjusted price, liquidity, credit quality, omitted factors, and hold at daily frequency.

Taiwanese stock market. The results suggest that there exist two heterogeneous agents in Taiwanese stock market, £\-investors behaving as long-term contrarian and £]-investor behaving as short-term momentum traders. To depict in detail the practical financial market, this research empirically tests HAM with different fundamental values (measured by the moving average price in different rolling windows) across different investment frequencies (daily, weekly and monthly). The result suggests that £\-investors (fundamentalists) expect prices to deviate from the short-term moving average but mean revert to long-term moving average. Beta investors (chartists) act as momentum traders in daily and monthly frequency, but short-term contrarian in weekly frequency. In addition, this study tests whether the parameters in HAM can explain some characteristics of crashes and bubbles. The result suggests that there are different investor behaviors in Asian, Dotcom, and Subprime crashes. By comparing the parameters (£\, £], and £^) of each individual stock, the study finds that stocks with contrarian £\-investors and short-term momentum £]-investors acting as short-term momentum traders have more volatile price pattern. As to crashes and individual stock volatility, the result suggests that sudden crashes (abrupt price decline) tend to occur in the stocks with short-term momentum traders, and while general crash (longterm economic cycle) tend to occur in the stocks with long-term contrarian investors. Stocks with larger Gamma, proxy for uncertainty, tends to have general crash only when £\-investors acting as contrarian and £]-investors acting as momentum traders.

In this study, the effects of opening trading mechanism and information flow on return volatility are examined in the Istanbul Stock Exchange. The change in the morning opening mechanism from a continuous auction to a call auction on February 2, 2007 and the extension in afternoon trading hours on September 7, 2007 provide unique opportunities in this respect. First, it is found that the call auction trading mechanism has a decreasing effect on the morning open-to-open interday volatility and morning intraday volatility for low-volume stocks but it does not have an obvious effect on the same type of volatilities for high volume stocks. Second, the study provides evidence that the increased information flow towards the end of the trading day increases the afternoon close-to-close interday volatility for high volume stocks while it does not have such an effect on low-volume stocks. Third, the overnight return volatility is decreased slightly with the extension of trading hours.

Agency theory suggests that information asymmetry between mutual fund managers and mutual fund investors can be mitigated if managers are compensated for the private information that influences mutual fund risk and performance. This study investigates the role of active management in influencing returns and return volatility of mutual funds. Chapter 1 investigates whether real estate mutual funds (REMFs) outperform Carhart’s (1997) four-factor and index benchmarks using daily return data from the CRSP survivorship bias-free mutual fund database from September 1998 to December 2013. We employ generalized autoregressive conditionally heteroscedastic (GARCH) volatility models to estimate more precise alphas than those generated in the extant studies. We document that risk-adjusted alphas of actively managed REMFs are statistically and economically significant, reflecting the informational advantage and skills of active managers. We also show that actively managed REMFs outperform the real estate index benchmark (Ziman Real Estate Index) and generate a yearly buy-and-hold abnormal return of 3.64%. Active management, therefore, provides value beyond the diversification benefits that can be generated by investing into the real estate index. While active managers of REMFs generate abnormal returns (gross of expenses), they capture the entire amount themselves, sharing none with investors (net of expenses). Accordingly, the average abnormal return to investors is close to zero due to expenses associated with REMFs, such as management fees, 12b-1 fees, waivers, and reimbursements. Finally, we find that passively managed REMFs do not generate abnormal risk-adjusted alphas in Carhart’s (1997) four-factor model. Chapter 2 examines managed volatility mutual funds (MVMFs) that utilize a range of investment strategies focused on portfolio volatility. These funds have increased in popularity in the wake of the financial crisis (December 2007 to June 2009) which introduced considerable volatility into the markets. We test whether MVMFs provide better performance during periods of recessions and expansions as compared to conventional mutual funds (MFs). We obtain several interesting results. First, MVMFs underperform compared to conventional MFs by more than 2% during the entire sample period. Second, MVMFs outperform conventional MFs in recessions by over 4% annually. Third, MVMFs underperform conventional MFs by more than 2.5% during expansions. Our results suggest that MVMFs can benefit investors during periods of recessions at the cost of performing worse during expansions. Chapter 3 studies MF return volatility patterns by testing a host of hypotheses for MFs with various style objectives. To conduct the tests, we use daily returns data from the CRSP survivorship bias-free mutual fund database from September 1998 to December 2013. We examine volatility patterns across the following nine styles: Passively Managed, Actively Managed, Sector, Capitalization, Growth and Income, Income, Growth, Hedged, and Dedicated Short Bias. We employ the exponential generalized autoregressive conditionally heteroscedastic (EGARCH) volatility model. Several results are obtained. First, we show that the financial crisis of 2007-2009 had a positive or a negative impact on volatility, depending on the investment style. Second, MF volatility behavior exhibits significant cluster effects in all styles, indicating that larger return shocks lead to greater increases in return volatility. Third, shock-persistence patterns differ across various MF styles with shocks to Dedicated Short Bias MFs being the least persistent and Capitalization and Growth and Income being the most persistent. Lastly, there is considerable negative asymmetry in MF return volatility changes in response to good and bad news in the sense that negative shocks to MF returns increase volatility more than positive shocks of the same magnitude for many Actively Managed MF styles. Significant negative asymmetry of this type makes the industry vulnerable to market downturns and should be addressed by regulators, MF managers, and investors. / Business Administration/Finance