Investing in Bridging Fuels: The Unit Commitment Problem of Public vs. Private Ventures.

• Author: Ioannidis, Filippos

1. INTRODUCTION

   In the context of the ongoing "Energy Transition" and decarbonization, natural gas is widely considered as the bridging fuel for the transition from conventional to renewable energy forms (Safari et al. 2019). One of the basic uses of natural gas is for electricity generation, while in parallel it plays a pivotal role towards the achievement of an integrated European electricity market. Over the previous decades individual countries have made considerable progress towards integration, yet the EU's objective of a single European electricity market is still away from completion. One of the countries that still lags behind in that race is Greece, since the dominance of public utilities in the national energy market hampered competitiveness and delayed the liberalization process (Dagoumas 2019).

   Despite this significant delay, Greece aims to gradually be transformed into an important energy hub of Southeast Europe and through the existing interconnections with neighboring countries to cover significant part of the electricity demand in the region. This imminent increase in demand along with the ongoing phase-out of lignite units and the dominance of natural gas into the electricity generation mix, has triggered both public and private sector to invest in new natural gas-fired units. Hence, the rivalry between the two groups raises specific questions for the efficiency of the existing units in terms of competition (Hogan 2002).

   According to prior research, there is no clear dominance for either of the two groups since the outcome varies based on the peculiarities of each market under investigation. To our knowledge, none of the previous studies have analyzed the efficiency of natural gas-fired units under the scope of the well-known Cash Flows at Risk (CFaR) and Risk Weighted Return (RWR) methodologies, that are typically met in the financial literature (Da 2009). Previous studies argue that enhanced operational efficiency under the UC problem is translated into increased social welfare via a reduction of the electricity prices and profit maximization (David and Wen 2000; David 1993).

   Hence, this paper seeks to identify which group is more efficient under the perspective of the Unit Commitment (UC) problem and its impact on risk weighted returns. By utilizing a unique hourly dataset for the period 2015-2019, we report a clear superiority of publicly owned natural gas-fired units in terms of operational efficiency that is robust to time and across methodological factors. Yet, the efficiency of privately owned natural gas-fired units is growing at a faster pace and is expected to surpass the efficiency of public units within 2 or 3 years. Our findings have significant policy implications. On the one hand, both national and European policymakers need to revise the role of Public Power Corporation (PPC) during the decarbonization period. On the other hand, the rapid efficiency pace of private units clearly suggests that market liberalization assists competition in the Greek electricity market.

   The paper is organized as follows: Section 2 provides a brief literature review on the topic under examination, while Section 3 describes the wholesale electricity market in Greece, by paying particular attention on the role of gas-fired power plants and the UC problem. Section 4 describes the applied methodology which is based on balancing mechanism and presents an overview of the various scenarios constructed. Moreover, the same section presents the CFaR and RWR methodologies used in our investigation. Section 5 presents the empirical findings. The paper ends with a summary and conclusions.

2. LITERATURE REVIEW

   2.1 Public versus Private Units

   Along with the liberalization of national electricity markets (Jamasb and Pollitt 2005), the conflict between public and private ownership has been thoroughly examined by prior research. An early examination by De Fraja (1993) argues that public ownership always results in a higher degree of productive efficiency. However, Boyne (2002), claims that public organizations are more bureaucratic, and public managers are less materialistic and have weaker organizational commitment than their private sector counterparts. In this context, privatization combined with the applications of performance based mechanisms has generally yielded significant cost reductions without reducing service quality (Joskow 2008).

   The aforementioned studies highlight the different aspects and incentives in terms of the utility maximization problem that both types of ownership face (Hart, Shleifer, and Vishny 1997). In general, there is no clear dominance for either of the two groups since the outcome varies based on the market under investigation. For instance, Bagdadioglu et al. (1996), provide illustrative information regarding efficiency and ownership on electricity distribution for the case of Turkey and found that private firms outperform compared to state owned. Next, according to Arocena and Price (2002), who analyzed the efficiency of both public and private electricity generators in Spain, concluded that public generators outperformed under cost of service regulation, while private generators responded to incentive regulation by increasing efficiency. Additionally, another case-study exploring the difference between public and private generators for the case of Korea, conclude that, public units were more effective in proving stable supply of electricity at a lower price (Lee and Ahn 2006).

   2.2 The Unit Commitment Problem

   Considering the role and main characteristics of Combined Cycle Gas Turbines (CCGT) units, prior research has analyzed in detail various aspects of their operation. For instance, Lu (2004), implemented a number of case studies to illustrate the advantages of CCGT units in modern electricity markets. Lobato et al. (2012), presented an optimization model based on the daily operation of a CCGT plant in Spain. Keatley et al. (2013), studied four natural gas-fired power units from Ireland to estimate their start-up costs under various scenarios in the view of increased Renewable Energy Sources (RES) penetration. Ma and Zhao (2015), explored the operational efficiency of CCGT units in China. Within the same scope, Wogrin et al. (2016) presented a methodology able to indicate the optimum operational strategy of a CCGT unit. Furthermore, Krishnany et al. (2016), developed a tailor-made algorithm which aimed to optimize a CCGT plant operation over a long-term time horizon. Woo et al. (2016), analyzed the actual profit effects of multiple fundamental drivers of natural gas-fired power plants in California. Moreover, Hermans and Delarue (2017), showed that flexibility in terms of quick start-up leads to greater profitability despite the increased maintenance cost.

   Furthermore, plethora of papers have analyzed the economics of a CCGT under the scope of maximizing their UC problem (Colpier and Cornland 2002; Padhy 2004). An early paper provided by Johnson et al., (1997) discussed the efficiency of UC problem in competitive electricity markets. In general, companies owning natural gas-fired units seek to maximize the optimal schedule and production level over a given time period. In line with Abdou and Tkiouat (2018), balancing production with demand and at the same time optimizing costs and resources is the ultimate goal behind the UC problem. Rodilla et al. (2014), illustrated a novel approach for modeling the generation schedule of natural gas-fired power plants subject to the UC problem. Cost optimization includes fuel costs, shutdown costs and start-up costs. Hence, it is essential for natural gas-fired units to follow an efficient procedure that would indicate best possible commitment status and at the same time reassure reliability of the system in times of uncertainty.

   Concerning the Greek wholesale electricity market, Biskas et al. (2011), investigated the operational and financial behavior of private CCGTs. Andrianesis et al. (2011), described in detail, the "instructed7 and "uninstructed" deviations of generators from actual system demand. A comprehensive analysis of the electricity market in Greece is provided by Kalantzis et al. (2012), who described the steady growth of natural-gas-fired power units in the system and their impact on Day-Ahead market prices. Biskas et al. (2012) concluded that the operation of the Greek electricity market under the presence of private units mitigate the dependence of Greece on electric energy imports, thus stimulating the Greek balance of trade and the national economy. Moreover, the authors identified that the presence of private units alleviates the economic burden that PPC would incur otherwise and enforce competitiveness in the Greek retail electricity market.

   In addition, Papalexopoulos and Andrianesis (2014), analyzed the deployment of the so-called Reliability UC that is of particular interest for the electricity Transmission System Operator. Sirvent et al. (2017), presented a case-study based on the Greek mandatory pool and modelled the UC problem subject to both unit technical and system operational constraints. Next, by utilizing a pioneer model of artificial neural network, a study presented by Dagoumas et al., (2017), incorporated the UC problem with electricity price forecasting and developed a risk management model.

   The operational scheduling under prespecified combined order scenarios of all CCGTs in Greece have been thoroughly discussed by Koltsaklis and Dagoumas (2018). A recent study by Koltsaklis, Zenginis, and Dagoumas (2020) proposed an optimization-based methodological framework for the optimal economic dispatch problem in power exchanges, through the development of a mixed-integer linear programming (MILP) model. The study utilized an iterative process, by quantifing the impacts of the proposed new options to the optimal energy mix, the wholesale prices, and participants economic performance. Finally...