Country risk and its impact on Pretoria Portland Cement's (PPC) valuation of its African expansion projects

Abstract

Investing in projects in Africa is generally a challenging venture as a result of the political, economic and social uncertainties and challenges faced by a potential investor. The expected rewards are however, high, due to natural commodities, untapped markets and availability of affordable resources. It is for this reason that in 2010 PPC decided to venture into Africa on the backdrop of a competitive and shrinking South African market. In 2016, following the completion/near completion of its cement project in DRC, Ethiopia, Rwanda and Zimbabwe, PPC realised that the returns received were not in line with the expected returns as modelled at the start of the project. In light of this, PPC contracted an external consulting company to develop a premium for country risk (CRP), which it could apply to its valuation model to cater for additional risks expected in each African country. The CRP developed was modelled on Damodaran’s melded approach model (MAM). This consulting report set out to review the newly developed valuation model and more particularly, the CRP applied. The objectives of the report were to further test the PPC developed CRP against other suitable risk assessment models’ scores to confirm its suitability and to determine if there exists a better model of determining country risk than the current PPC developed one. The study commenced by conducting a literature review on the subject of country risk and identified the most reputable risk assessment models for determining country risk that could potentially be applied to PPC. In this respect, it was established that three categories of risk assessment models exist, namely, risk service agencies, credit rating agencies and sovereign default models. The risk service agency models that were considered as potential risk models for PPC were the WGI, PRS, WEF, EFW and F-HSE. Moody’s rating and Damodaran’s melded approach model was considered as part of the credit rating agency and sovereign default model, respectively. v The research design was based on identifying a suitable proxy for country risk which was to be used to test the PPC CRP and the risk assessment models’ scores. The data collection for the risk models was based on secondary data spanning a 10-year period (2006 to 2015) for the respective African countries. The BC-DSD which is a database that contains the actual sovereign defaults for all countries was identified as a suitable proxy for country risk and this was confirmed by testing its association against the risk assessment models. The PPC CRPs for the respective countries was then tested against the BC-DSD as well as against the risk assessment models. Based on the correlation analysis conducted, it was found that the results obtained were not statistically significant as a result of the high p-values. A determination of whether the PPC CRP was a true reflector of risks in the respective countries could therefore not be made. The risk assessment models were then tested again the BC-DSD and amongst the risk assessment models themselves. Based on the correlation analysis, the WGI was found to be the best reflector of country risk, with PRS and EFW a close second and third respectively. Damodaran’s melded approach model was found to be ineffective in the African countries and it stands to reason as the model relies on accurate financial market data which is non-existent or if it exists, is not very reliable. In view of the findings, it is recommended that the PPC CRP model is reviewed and that Damodaran’s MAM be replaced with a combination of the WGI, PRS and EFW respectively.