Kerestecioglu, Semih
(2021)
Essays on Economic Uncertainty and Financial Markets.
PhD thesis, University of Liverpool.
|
Text
201304013_Nov2021.pdf - Unspecified Download (2MB) | Preview |
Abstract
This thesis examines the role of economic uncertainty in investors’ decision process and analysts’ forecast bias. The first empirical chapter investigates the effect of firm-level exposure to economic uncertainty (EUE) on cross-sectional returns through differentiating the mispricing from ambiguity-premium effects. Conditional on a common mispricing index, I find that EUE induces disagreement among investors, which amplifies mispricing. The highest EUE quintile produces a significantly higher mispricing alpha than the unconditional mispricing effect. By contrast, the high-minus-low EUE portfolio in the non-mispricing group generates a significant positive premium in the sense of the ambiguity-return trade-off. The EUE-induced mispricing effect is different from existing limits of arbitrage explanations, such as idiosyncratic risk. The ambiguity premium is a new source of the risk premium that is robust to the latest risk models. The second empirical chapter studies the role of market-wide sentiment in relation to the mispricing and the ambiguity effects documented in the first study. Considering the presence of the market-wide sentiment combined with short-sale constraints, I find larger mispricing spread in stocks with high EUE following high-sentiment periods. This mispricing effect is stronger following the periods with both increasing economic uncertainty and high sentiment. It suggests that economic uncertainty indeed leaves more room for the sentiment effect in the market. The ambiguity premium in the non-mispricing group, however, is significant only following low-sentiment periods during which the mispricing effect vanishes. This is consistent with the previous finding that the market pricing is more rational when investor sentiment is relatively low. The final empirical chapter examines whether there is an effect of EUE on analysts’ optimism. Existing literature shows that equity analysts have an optimistic bias. I find that analysts are even more optimistic for stocks with higher EUE. This is especially true following periods with high economic uncertainty. This study confirms that such an increase in optimism is for incentives given that high uncertainty reduces their reputation costs by lowering the chance of them being caught for such bias. The effect is more pronounced when firm-specific information quality (measured by earnings quality and information availability at the market level) is lower, and investors are less sophisticated (measured by the proportion of institutional ownership). Finally, analysts issuing optimistic view for stocks with higher exposure to economic uncertainty impede the price efficiency in the market. The EUE-induced mispricing is significantly apparent among stocks with high optimism. This thesis contributes to the literature in several ways. First, economic uncertainty has two seemingly contradicting mechanisms in asset pricing, including the ambiguity premium and the mispricing effects. This thesis reaches a clear conclusion that both mechanisms are at work through disentangling the two effects, which has not been studied by the existing literature where these two mechanisms are studied in isolation. Second, I identify EUE as a common mispricing component across anomalies, which is different from but complements investor sentiment and arbitrage risk, contributing to existing studies by suggesting that mispricing has common components across stocks. Finally, analyst optimism for incentives exacerbates the mispricing effect of economic uncertainty. This finding complements studies which suggest that analyst bias impedes the price efficiency in the market.
| Item Type: | Thesis (PhD) |
|---|---|
| Divisions: | Faculty of Humanities and Social Sciences > School of Management |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 18 Nov 2021 13:51 |
| Last Modified: | 18 Jan 2023 21:24 |
| DOI: | 10.17638/03143368 |
| Supervisors: |
|
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3143368 |
Dimensions
Dimensions
