Financial decision-making using the implied cost of capital

Al Mutairi, Saif (2019) Financial decision-making using the implied cost of capital. Doctoral thesis, University of East Anglia.

[thumbnail of 2019AlMutairiSaifPhDMangt.pdf]
Download (3MB) | Preview


In this thesis, I aim to shed light on the value of the market-implied cost of capital (ICC) in financial-decision making in three distinctive contexts. ICC is a forward-looking proxy for the expected return of a firm’s stock which is implied from the current stock price and a choice of analyst forecasts or accounting forecasting models. As there has been a large variety of ICC models proposed in the literature, I first aim to identify the models with superior forecasting performance. To this end, I show through a comprehensive comparison that simple ICC models work better than more complicated widely used formulations in terms of forecasting future realised returns, and as a statistical quantity in terms of out-of-sample bias and measurement error variance. Specifically, a dividend discount model with a terminal value based on analysts’ price target, or a price-over-earning ratio based estimate outperform more complicated ICC and risk factor models. These simple models coincide with market beliefs about expected returns more than more complex models. I find that ICC derived from models based on the residual income framework have better forecasting power than models that assume abnormal growth in earnings, in contrast to theory. Using mechanical earnings forecasts to replace analysts forecasts in ICC models does not consistently improve the forecasting ability of the models except for dividend discount models. Furthermore, adjusting the ICC estimates for firm characteristics and popular risk factors lead to better forecasts, and result in lower out-of-sample mean error and error variances, especially with models based on analysts earnings predictions and dividend discount models. I also develop a new estimator based on free cash flow to equity, and show that it predicts future returns and exhibit errors comparable to the best performing models, and I argue that it is a more economically sound construct than dividends.

Second, I capitalise on the ICC literature to derive forward-looking estimates of expected returns to improve the out-of-sample performance of portfolio selection strategies. I find that using ex-ante ICC estimates instead of the ex-post first moment as a proxy of expected returns in a tangency portfolio yields a higher out-of-sample risk adjusted returns and lower turnover. Moreover, I demonstrate that ICC-based market timing portfolios beat the conventional market-timing portfolio and naive 1/N strategy in terms of out-of-sample Sharpe ratio and turnover. The evidence presented contributes to the research on how accounting information and models can be used to enhance investment decision making.

Third, I study the effect of risk similarity between acquirers and targets as captured by market implied cost of capital on mergers decisions and outcomes. I propose a new measure of risk similarity between two firms. This employs forward-looking market-implied cost of capital estimates to proxy for systematic risk. I use the new measure to study how risk similarity affects merger formation and outcomes. The empirical analysis provides evidence that firms with similar risk profiles are more likely to merge. The level of risk similarity is positively associated with the probability that an announced acquisition deal will be completed and negatively associated with the length of the period between deal announcement and completion. Mergers resulting from firms with high pre-merger risk similarity tend to lead to higher combined abnormal returns in the short-term and higher operating performance and lower risk in the long-term. The results indicate that risk similarity in mergers is in line with shareholder preferences, leads to less suboptimal investment in the target and facilitates improved management of the acquired assets. The evidence presented contributes to the research on determinants of M&A success, provide a new perspective on the impact of how the risk-profile of a company as understood by the market affect investment decisions, and offers a new methodology for defining risk similarity between firms.
JEL classification: G11, G12, M41.

Item Type: Thesis (Doctoral)
Faculty \ School: Faculty of Social Sciences > Norwich Business School
Depositing User: Nicola Veasy
Date Deposited: 04 Jun 2024 10:03
Last Modified: 04 Jun 2024 10:03


Downloads per month over past year

Actions (login required)

View Item View Item