The cost of equity capital for unlisted firms: evidence from USA data

This dissertation comprises three empirical papers that examine various methods ofestimating the systematic risk of unlisted firms in order to identify more efficient waysof calculating their cost of equity capital. Paper 1: 'The application of proxy methods for computing the cost of private equity:evidence from listed firms'.The two-beta model decomposes the systematic risk in the sensibility of cash flowand discount rate change. We propose a modified version of this model (MTBM) tocompute the cost of capital for private equities (PEs). This model includes not only theaccounting return reaction to long-term changes in consumption, but also linksfundamental reactions to temporal changes in risk aversion. We test this model alongwith three traditional alternatives that are potentially useful in computing the cost ofcapital for PEs: accounting betas (BACC), unlevered betas (PLB), and operational betas(BOP). Using a two different tests, we gauge their capacity to explain cross-sectionalstock returns and their forecasting abilities. We find that PLB, BACC, and MTBM areable to explain (with some limitations) the cross-sectional variations of stock returns.The forecasting experiment indicates that the MTBM produces the best output. Paper 2: 'Unlevered betas and the cost of equity capital: an empirical approach'.This paper calculates systematic risk based on the capital asset pricing model(CAPM) in order to determine the significance of financial leverage. Instead of testingthe unlevered beta directly, we develop a multinomial model with theoretical targets inthe unleveraged/leveraged process. We find that it is statistically more robust to includetax shields as a part of the unleveraged/leveraged process than to omit them. Our resultsalso suggest that the use of the proxy levered beta to address the lack of marketinformation for both non-traded firms and individual business units is not misleading. Paper 3: 'Estimating the cost of equity capital for private firms using accountingfundamentals'.Financial literature suggests the use of BACC as a proxy for CAPM market beta(BMKT) when estimating the cost of equity capital in the absence of stock prices.Previous researchers have made this estimation by determining the correlation between the accounting variables and BMKT. However, the magnitude of the resultingcorrelation error remains unknown. This study attempts to test the accuracy of BACC asa proxy measure for market risk and to examine the magnitude of error in correlationbetween these two measures. Our findings indicate that BACC over-estimates theBMKT by 20%-50%. This error may narrow to 22%-25% by applying correctivemeasures such as scaling operational earnings by equity; however, the error is noteliminated. Our output also suggests that BACC may be biased when assessing the riskof small firms. This study concludes that MTBM seems to be a more efficient method forcomputing the cost of equity capital for unlisted firms than traditional methods. Theresults suggest that BACC and PLB, while less efficient, can still explain the behaviourof stock returns, albeit with limitations. In addition, these two methods are stronglyrelated to market beta. However, both may exacerbate CAPM issues when computingthe cost of equity capital for small firms.