Renewable energy investments with storage: a risk-return analysis

Abstract

Purpose The purpose of this paper is to study investments in renewable energy projects which are jointly operated with an energy storage system, with particular focus on risk-return characteristics from the perspective of private and institutional investors, taking into account resource risk, energy price risk, inflation risk and policy risk. Design/methodology/approach To this end, this paper presents a stochastic discounted cash flow model which is then applied to a wind farm with a pumped hydro storage system. Findings The results show that energy storage systems have the potential to increase the expected present value of future investment cash flows and to hedge (downside) risk. However, to realize this potential, storage systems have to be cost-effective in terms of fixed operation, maintenance, staffing and insurance costs. Also, several key factors are identified which have a considerable influence on the performance of the operation strategy. Originality/value The paper contributes to the literature by conducting an analysis of (downside) risk and return of renewable energy investments with a storage system taking into account stochastic policy, resource, inflation and energy price risk.

System of Comprehensive Assessment of Project Risks in the Energy Industry

“…Researchers study such aspects as risk-return analysis in application to investments in renewable energy sources [1], the importance of project finance in low-risk projects [2], the advantages of the Monte Carlo modeling method in evaluating public-private partnership projects [3], and conducting feasibility studies based on the Monte Carlo method using multi-energy balance financial equations that take into account the uncertainties and risks associated with different variables in the design and construction of solar thermal power stations [4].…”

Model risk regarding monthly wind energy production for the valuation of a wind farm investment

“…Consequently, stochastically modeling the wind energy produced is important for the indepth valuation and risk assessment of wind farm projects. For the valuation of direct longterm investments in a wind farm project, monthly models can be used (Abadie and Chamorro, 2014;Gatzert and Vogl, 2016;Krömer and Gatzert, 2018). In the case of long-term simulations, such as, for example, in Krömer and Gatzert (2018), with an investment horizon of 25 years, monthly models provide computational advantages.…”

“…For the valuation of direct longterm investments in a wind farm project, monthly models can be used (Abadie and Chamorro, 2014;Gatzert and Vogl, 2016;Krömer and Gatzert, 2018). In the case of long-term simulations, such as, for example, in Krömer and Gatzert (2018), with an investment horizon of 25 years, monthly models provide computational advantages. Despite this, the literature mainly provides approaches to model hourly wind speeds and derives actual wind power output by means of non-linear wind power curves.…”

“…seasonality or autoregression, as well as different distribution assumptions. Thus, the aim of this paper is to assess model risk with respect to these different characteristics and to identify important factors which influence the valuation and risk assessment of a wind farm investment, with the additional consideration of a joint operation with a pumped hydro storage system which can reduce risk and increase profits (Krömer and Gatzert, 2018).…”

“…With respect to the valuation model, this paper follows Gatzert and Vogl (2016) who consider several stochastic risk sources besides resource risk, namely, energy price risk, policy risk and inflation. Accordingly, this article additionally considers the possibility of the joint operation of the wind farm with an energy storage system to increase profits and reduce (downside) risk, which has previously been conducted by Krömer and Gatzert (2018). Both Gatzert and Vogl (2016) and Krömer and Gatzert (2018) use the model for the wind energy produced by Abadie and Chamorro (2014), who propose a normal distribution of the load factor with a long-term mean and a monthly seasonality term.…”