How capacity mechanisms drive technology choice in power generation: The case of Colombia

Renewable and Sustainable Energy Reviews 56 (2016) 563–571

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser

How capacity mechanisms drive technology choice in power generation: The case of Colombia Yris Olaya a, Santiago Arango-Aramburo a, Erik R. Larsen b,n a b

Decision Science Group, Universidad Nacional de Colombia, Colombia Università della Svizzera italiana, Switzerland

art ic l e i nf o

a b s t r a c t

Article history: Received 12 May 2015 Received in revised form 30 October 2015 Accepted 22 November 2015 Available online 17 December 2015

Colombia enacted its first legal framework for promoting alternative energies in 2001 and a second framework in 2014. Since the generation technology mix has not changed since 2000, there is a need to understand how regulation and market structure affect the adoption of technologies. In this paper we address the question of what has been the impact of the capacity mechanisms adopted during the 2000s on technology choices for power generation in Colombia. Our approach is to analyze the evolution of market structure and regulation. We found that regulatory uncertainty and low prices drove a surge of small hydro plants during the 2000s. During the 2010s, the new regulatory focus on reliability of supply has resulted in increased coal-fueled generation and large hydro. This increased reliance on hydro power can further delay the entry of renewable technologies and the diversification of the Colombian portfolio. & 2015 Elsevier Ltd. All rights reserved.

Keywords: Power generation mix Capacity mechanisms Renewable energy Colombia

Contents 1. 2.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overview of the Colombian power system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. From central planning to deregulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Effect of capacity mechanisms in the Colombian power sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Before deregulation (Before 1994) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Market system and capacity charge (1994–2006) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Reliability charge or forward firm energy market (After 2006) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. From big to small hydro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Role of (other) renewables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Discussion and conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Technology choices for power generation determine to a large extent the structure of energy supply and have lasting impacts on the environment. During the last three decades, we have witnessed a shift away from coal to gas-fueled power generation, the introduction of renewable energy technologies, and recently in many n

Correspondence to: Via Buffi 13, CH-6904 Lugano, Switzerland. E-mail addresses: [email protected] (Y. Olaya), [email protected] (S. Arango-Aramburo), [email protected] (E.R. Larsen). http://dx.doi.org/10.1016/j.rser.2015.11.065 1364-0321/& 2015 Elsevier Ltd. All rights reserved.

563 564 565 565 565 566 567 568 568 569 570

countries a desire to move away from nuclear generation [1–3]. Some of these changes are motivated by environmental concerns, by technological uncertainty, economic changes, and perceived risk of nuclear power as well as ideological views, all of which play a role in establishing policy priorities. The creation of power markets worldwide is the result of a broader energy policy aimed at increasing efficiency and attracting private investments – with the goal of satisfying a growing demand for electricity, particularly in the developing world [4]. In the early years of market reforms, increasing competition and efficiency in generation were the main issues by regulators and

564

Y. Olaya et al. / Renewable and Sustainable Energy Reviews 56 (2016) 563–571

researchers. Bringing competition to the power market meant often unbundling utilities, breaking monopolies up, and creating commodity-like markets for electricity. Two decades later, future reliability is one of the main concerns in liberalized markets [5–9]. In fact, the International Energy Agency (IEA) and European Union (EU) estimate that EU countries need to invest Euro 1 trillion from 2012 to 2020 and a further Euro 3 trillion to 2050 to ensure adequate electricity capacity [8], despite some disagreements with these estimates [10]. Ensuring that such new investments are made in a timely manner is a challenge for power markets and it might be an opportunity for introducing alternative generation technologies. In theory, a competitive market setting should give the right signals at the right time for investors to choose the best technology for expanding generating capacity. Power markets, however, are not perfectly competitive [11], future returns are highly uncertain [12] and investment decisions are constrained by environmental concerns, availability of appropriate generation technologies and the general future economic outlook [13]. As a result of the high uncertainty and the long lead times for new projects, it is common for power markets to observe cycles of over- and under-capacity margin [14]. Such fluctuations increase supply risk, and thus the objective of regulation is to reduce them. Achieving a long-term security of supply (resource adequacy) at a low cost is particularly challenging for developing countries that face increasing demand and need to expand access to energy [5–9,15]. While markets in different jurisdictions have different pricesetting mechanisms, the most common mechanism is prices based on marginal costs. Marginal cost-pricing, however, does not offer sufficient incentives for adding new capacity and increasing longterm security of supply [16], a phenomenon known as the “missing money problem” [17]. Most markets solve it by using a separate, often complex, set of rules for calculating and allocating payments for capacity availability – known as capacity payments. The idea behind capacity payments is to recover capital costs that are not part of the market price, thus keeping enough capacity to satisfy demand at a given reliability level. Nowadays, there is a debate on whether capacity payments are needed, instead of relying on an energy-only market that rewards generating plants with scarcity rents [15,17,18]. Suppliers in an energy-only market bid prices instead of bidding their marginal cost. Price bids should include the capacity cost and they also increase during peak demand periods, when capacity is scarce. Many electricity markets use capacity payments or a related mechanism to increase the revenues of the generators and to incentivize their investment in new capacity. That is the case of Argentina, Brazil, Chile, Colombia, Perú, Spain, and the UK [15]. Alternatively, Australia, Alberta, ERCOT, and New Zealand have complete commodities markets for electricity and have implemented energy-only markets that have been successful so far [17]. The contribution of new capacity to the system's reliability depends on the size and timing of investments and varies widely across technologies. Wind technologies, for instance, have nameplate capacities that are between 60 and 70 percent higher than their annual output, and they only increase capacity margins when wind is available. In fact, a proper mix of technologies increases the reliability of supply. Therefore, regulators need to understand the logic of investment decisions and tailor the resource-adequacy mechanisms to each particular case. To a great extent, the incentives and rules set by regulators drive technology choices by investors in a particular area, and the pace of the investment. The consequences of investment decisions are complex as new capacity takes from some months (e.g. Photo Voltaic (PV)), several years (e.g. Combined Cycle Gas Turbines (CCGT)) to a decade (e.g. big hydro) to come on-stream. Moreover, capacity is often added in large chunks and has a long life-time

compared to other industries, making it more difficult to ensure that the right incentives are in place to achieve the right mix of technologies. The aim of this paper is to illustrate how different regulatory regimes affect technology choices, using Colombia as a case study. Colombia is an interesting case as the incentives for adding capacity have been modified twice since deregulation, in an effort to adapt the regulation to shifting economic and market conditions. The first incentives (1996–2006) aimed at reducing the electricity system's vulnerability during dry periods. These incentives were modified in 2006 in order to provide signals for expansions of the system. While regulation has succeeded in securing supply during dry periods, the technology mix has remained basically unchanged since 2000. Furthermore, with the exception of small hydro plants, penetration of other renewable technologies is almost zero. In the next sections we examine how capacity mechanisms and the structure of the market have influenced technology choices for power generation in Colombia. This paper is organized as follows: Section 2 makes a brief description of the Colombian power system, including the capacity before deregulation and explains the events that led to deregulation. Section 3 describes and discusses the resource adequacy mechanisms used in Colombia, and their impact on the generation mix, including capacity mechanisms’ implications for renewable energy. Section 4 presents the main insights and conclusions from the previous analysis.

2. Overview of the Colombian power system Seasonal variations of power demand in Colombia are relatively small, and high capacity margins are maintained as a reliability strategy. Because the share of hydro generating capacity is larger than 65%, the Colombian power system is vulnerable to weather changes such as the prolonged and intense droughts caused by the macroclimatic phenomena of “El Niño South Oscillation” (ENSO). The interconnected power system provides electricity to 94.6% of the population [19], with high quality standards. Table 1 summarizes the main indicators of Colombia's macroeconomic condition and power system. Hydro power dominates the technology Table 1 Macro-economic and electricity industry indicators in Colombia. Demographic and economic indicators Population (est. Jan. 2015)b 47,965,803 GDP (ppp) Billions US$ 526.5 (est. 2013)c GDP per capita (ppp) 2013 11,100 est.c

Electricity prices ($US/kW h) 2012a Residential 0.2 Industrial 0.22 Commercial

0.16

Market structure

Bid based

Installed generation capacity (MW), Dec. 31 2013d Hydro 9875 Thermal 4598 Wind 20 Co-generation 66 Total 14559

Electricity demand profile (2012)e Industry 31% Transport 0% Residential 41% Commerce and public 24% Agricultural/Forestry/ 4% fishing

Energy intensitye TPES/pop (toe/capita) TPES/GDP (toe/000 2005 USD)

Electricity demand growthe Average 2004–2008 3.0% Average 2008–2012 4.5%

a

[24]. [22]. c [21]. d [20]. e [23]. b

0.66 0.16

Y. Olaya et al. / Renewable and Sustainable Energy Reviews 56 (2016) 563–571

Fig. 1. Contribution of public utilities (electricity, gas, and water) to the Colombian external debt between 1970 and 1998, [28].

mix for generation, as it represents 65% of the 8525 MW of installed capacity [20]. Since the reforms of the power sector in 1994, Colombia has had a market-oriented system, including regulatory and oversight bodies [25]. Following deregulation, the power market has supplied energy with virtually no interruptions; power companies are today in a good financial position, as evidenced by their ability to enter new markets in Chile, Brazil, Perú, and Guatemala, among others [26]. Nevertheless, as we discuss next, the process to improve the system's performance was not smooth and it has required major structural and regulatory challenges. The issues before the reforms are illustrated by the fact that public utilities were responsible for more than 35% of the Colombian external debt (see Fig. 1). Some companies had difficulties financing expansions and mismanagement was widespread [27], which led to a series of costly blackouts in the early nineties. These major blackouts made reforms easier to pass because they exposed the weaknesses of the existing monopolistic system. The Colombian power sector has gone through several structural reforms, in which the regulator has adapted the strategy to achieve resource adequacy, as we show in the following sections. These changes provide us with valuable data and insights on the impact of regulation on technology choices, as well as on the effects of the initial technology mix on regulatory decisions. We discuss these issues in further detail in the following section, in which we give a short account of the events that shaped the present power system in Colombia. 2.1. From central planning to deregulation In the early 1960s, the Colombian government began planning for the interconnection of the largest regional, verticallyintegrated public utilities. ISA, the national grid company, was created in 1967 and was responsible for the system operation and expansion planning. Between 1970 and 1980 the public utilities, along with ISA, installed more than 1200 MW of generating capacity,1 to satisfy the growing demand. Because of this expansion, utilities (including ISA) were having trouble financing projects and there were problems with completing projects on schedule; such delays led to power cuts in 1981 [27]. To avoid power shortages, the government focused on financing new generating plants which led to a 28% increase in Colombia's external debt between 1970 and 1980. By 1985, the external debt originating from public utilities was 35% [28]. At this point, nearly 40% 1

ISA informes de operación 1970–1998.

565

of the electricity sector's budget was allocated to pay back the loans, and to cover past deficits [29]. During the 1980s, GDP grew on average less than 4% annually [30]. The low economic growth continued through the 1990s, and between 1985 and 1996 per capita electricity consumption in Colombia was one of the lowest in Latin America [31]. Nevertheless, construction delays in new power plants and a strong El Niño in 1991–992, led to major blackouts from March of 1992 to April of 1993 [32]. The severe financial problems of the utilities, their lack of efficiency and inability to make timely investments explain why, by the end of the 1980s, there was a general agreement on the need for a reform of the power sector, to improve its technical and administrative efficiency. The power industry's reform began with the new constitution in 1991, and was completed in 1994 by laws 142 [33] and 143 [34]. Law 143 (Electricity Law) aims at increasing the efficiency and quality of the electricity supply through market mechanisms, whereas Law 142 establishes the role of the government as a regulator. Enactment of these laws was the first step towards deregulation. Law 142 created regulatory, planning, and oversight bodies for electricity and gas. To foster competition, reforms separated transmission from distribution and generation, and achieved partial unbundling of generation and distribution. An independent system operator in charge of coordinating transmission and dispatch was created, and open access to the grid was guaranteed. Monopolies in transmission and distribution were regulated, while generation and the unregulated retail market were opened for competition. A pool-based market was created to increase competition in power generation. In this market, price is set in a day-ahead market in which all generators must participate, regardless of their contracts. Bids are freely placed in the market. Thermal generators' bids are to be based on variable operating costs (fuel) while hydro generators' bids represent the value of water. One of the main motivations for the reform was the need to attract investment in the generation needed to keep pace with the (forecasted) growth in demand, [4,25,35]. For this reason, from the beginning, there were separate payments for energy and for capacity. Capacity payments have been modified three times since 1994, reflecting policy shifts, and these changes have produced distinct responses in capacity expansion that we examine next.

3. Effect of capacity mechanisms in the Colombian power sector In this section we discuss the effect of capacity mechanisms on generation-technology choices in Colombia. Fig. 2 shows the additions of generating capacity to the Colombian system from 1960 to 2010 and the approved additions to enter between 2011 and 2019. As discussed above, between 1960 and 2014 the electricity sector went through major structural changes. We divide our analysis into three periods: the period before the sector reform in 1994, the first regulatory period focusing on resource adequacy where the mechanism was capacity payments, and a second regulatory period beginning in 2006, corresponding to reliability payments (see Fig. 2). 3.1. Before deregulation (Before 1994) Before 1994, ISA, the national transmission company, coordinated the expansion of generating capacity. Expansion projects were selected using an optimization program that minimized capital and operating costs while satisfying expected demand and transmission constraints, among other conditions. During this period, hydro power was the preferred technology for expanding

566

Y. Olaya et al. / Renewable and Sustainable Energy Reviews 56 (2016) 563–571

Fig. 2. Annual additions to generating capacity in Colombia from 1960 to 2010 and approved additions from 2011 to 2019. Data from ISA's annual operation reports 1980– 2005, and from UPME, [36].

power generating capacity at low cost. In fact, 86% of the 1240 MW added during the 1960s was hydro. In the 1970s, new generating capacity doubled the previous decade's capacity and the share of thermal generating capacity increased to 40% of the total, as more than 5 GW were added. Despite this surge in capacity, power generation could not keep up with increases in demand as some projects were delayed, which led to a blackout in 1983. By 1990, total generating capacity was 8312 MW with 78% hydro and 22% thermal. Capacity, however, continued to grow at a slower pace than demand, creating the conditions for the pivotal 1992 blackouts. The year of 1991 had the lowest rainfall recorded to date and water reservoir levels dropped from 57% in September of 1991 to 17% in March of 1992 [37]. Furthermore, thermal plants had been poorly maintained, and of the 2100 MW of installed thermal capacity, only 1137 MW (62%) were operating [38,32]. Power shortages began in March of 1992 and lasted over a year. More than 2400 GW h were rationed and it is estimated that these programmed power cuts cost about one percentage point of the GDP over the period [38]. During the blackouts, the government invested in refurbishing generating units and accelerated the construction of backup plants, adding 400 MW of capacity. When the blackouts ended, an investment strategy was devised to accelerate capacity expansion, thereby improving the power system's robustness by adding 1950 MW of generating capacity between 1995 and 1999 [32]. Thermal plants accounted for more than 60% of the new capacity, comprising: 750 MW of gas fired plants, 450 MW of coal-fired plants, and 750 MW of new hydro capacity. In 1993, estimated costs for the 1995–2000 expansion strategy were 1.88 billion US dollars [32]. The hydropower projects had already been financed by multilateral bank credits and the public utilities' own resources. However, there was particular interest in securing financing for thermal projects that would decrease the vulnerability of the system to extremely dry seasons. The scheme proposed to finance the thermal projects consisted of long-term power purchase agreements (PPA), guaranteed by the government. This scheme pays for the energy from, as well as for capacity of, the new plants. The same long-term PPA-with-guarantees scheme was suggested for promoting private investment [32]. 3.2. Market system and capacity charge (1994–2006) The blackouts of 1992 led to the construction of thermal capacity, and the pressure for increasing reliability provided by thermal generation continued after the power sector was reformed in 1994. However, while thermal plants provide reliability, their operating costs are higher than hydro under normal

weather conditions, making them less competitive. The electricity market began operating in 1995, and it became clear that thermal output sold at peak prices during dry seasons would not generate enough revenue to justify investments in thermal capacity. The response from regulators was to create a capacity charge [39,40], which rewarded the contribution of each generator to securing supply during dry periods, in particular for an extreme El Niño event. Such charges had previously been applied in the UK, Chile, and Argentina [15]. Capacity payments were proportional to the fixed costs of a thermal generation plant (capacity charge). The reference value for estimating capacity charges was the cost of installing 1 kW of an open-cycle gas generator (5.25 US$/kW). Charges were collected from the energy pool and were a lower cap for the electricity price [4]. Each month, the capacity payments for each plant were calculated as the product of the capacity charge multiplied by the fraction of demand that each plant could supply in a critical hydrology scenario. To calculate the availability of plants in a dry year, the regulator used cost and critical hydrology data to run a model of the market [41]. After some adjustments, capacity payments were applied from 1996 to 2005. As shown in Fig. 3, the average share of capacity payments awarded to thermal and hydro technologies is similar to the thermal/hydro share of the system (approximately 30%/70%). The thermal capacity expansion from 1996 to 2000 can be explained, at least partially, by the capacity payments mechanism and to a less degree by the government's push for a higher share of thermal generation. The investments of this period changed the technology mix from 87% hydro in 1995 to below 70% in 2000, which corresponds to an increase in thermal capacity of 2514 MW. The new thermal capacity included 1230 MW from the government's emergency plan of 1993; the only hydro plant to come on stream in this period (340 MW) was also initiated by the emergency plan. The increase in thermal generating capacity achieved its purpose as it made the system more reliable. In fact, the 97–98 El Niño, which was stronger than the 91–92 El Niño, caused no blackouts, although spot prices were about four times higher than in previous dry seasons. By the late 1990s, Colombia faced an economic crisis with GDP falling by 4.2% in 1999, and growing at rates below 5% until 2003 [30]. The economic recovery was slow, electricity demand growth stalled, and capacity margins increased to above 50%, which kept prices low making the return on investments less attractive. As a result, only 1050 MW were added to generating capacity between 2000 and 2005, of which 95% was hydro. This included about 200 MW of small hydro, and a 400 MW hydro plant commissioned before 1994 (see Fig. 2). Another factor contributing to the delay of

Y. Olaya et al. / Renewable and Sustainable Energy Reviews 56 (2016) 563–571

567

100% 90%

CC Thermo

80%

CC Hydro

70% 60% 50% 40% 30% 20% 10% Jan-04

Jul-03

Oct-03

Apr-03

Jan-03

Jul-02

Oct-02

Apr-02

Jan-02

Jul-01

Oct-01

Apr-01

Jan-01

Jul-00

Oct-00

Apr-00

Jan-00

Jul-99

Oct-99

Apr-99

Jan-99

Jul-98

Oct-98

Apr-98

Jan-98

Jul-97

Oct-97

Apr-97

Jan-97

0%

Fig. 3. Share of capacity payments awarded to thermal and hydro technologies in Colombia between 1997 and 2004. Data from NEON (www.xm.com.co).

investment decisions was general economic uncertainty, as well as the expectation for a new regulation period beginning in 2005 [25]. By the year 2000, capacity payments were contested by both thermal and hydro generators. In fact, the regulatory commission was studying alternatives for increasing capacity adequacy. It was argued that these payments did not send clear signals for expansion, that they favored hydro generators, and that the technology used as a reference (GT) had low capital costs [42]. Most of the critiques from generators were about the model's validity and assumptions. In particular, generators argued that the optimization model used by the regulatory commission was not adequate for the reformed market because its parameters did not represent the system's characteristics. It was also argued that capacity charges were only a source of revenue and that reliability was not guaranteed [43,4]. These problems led the regulator to introduce the reliability payments, which we discuss in the following section. 3.3. Reliability charge or forward firm energy market (After 2006) In 2006, the regulator changed the mechanism for securing generation adequacy in the Colombian system. The previous scheme (capacity charge) had worked well on thermal-based systems, such as the UK, but as discussed above, it was argued that paying for firm energy instead of capacity would be more appropriate for a hydro-based system such as the Colombian system [44]. The new mechanism is called the reliability charge, and it is based on a forward market for firm energy, where firm energy is the capacity to deliver energy in a dry year. The regulator auctions off sufficient obligations to supply firm energy (OEFs) to satisfy the system's forecasted demand, three years ahead. OEFs can be seen as a call option, backed by physical capacity. A generator receiving an OEF must supply a given quantity of energy if the spot price is above a previously defined scarcity price [45]. A generator that supplies more than its share during scarcity periods receives the spot price, while a generator that supplies less is penalized. Therefore, there is an incentive for the generator to be able to deliver the agreed quantity as long as the penalty for not being able to deliver is severe enough [44]. Contract length for new plants is currently 20 years. Simulation analysis indicates that the reliability mechanisms attract investment [46], lower market risk, and improve coordination in investment [47]. This latter feature is desirable, since lack of coordination in investment has been related to undesired long-term cycles of over- and undercapacity [14]. Similar auction schemes are being used in Brazil, Chile, Peru, and Panama [15]. The auctions of long-term supply contracts provide stable revenue and more certainty for the investors in new generation capacity so that adequate resources are provided when they are needed. Moreover, forward markets lower risk for buyers

because less energy is traded at volatile spot prices [47]. The downside for the buyers is that they will have to pay more for the energy during periods where prices otherwise would have been low. To some extent, this mechanism shares some of the characteristics of mothballing, a controversial practice in which buyers also pay for having less volatility [48]. The period 2006–2009 was designated as a transition period by the regulator [45]. During this period, a centralized mechanism was used for assigning and pricing capacity. A total of 138 MW of new generation capacity was added during this period. The expansion consisted of 57 MW of thermal and the rest of small hydro-power plants. The first auction took place in 2008, awarded contracts for 3420.8 MW of new generating capacity that will come on stream between 2013 and 2019, where about 90% of this capacity is hydro. The large share of hydro generation is explained by the mechanism for estimating contributions to firm energy. For hydropower plants, the estimation of firm energy is based on an optimization model that considers critical hydrology. By contrast, firm energy offered by thermal power plants must be backed by a longterm fuel contract [49]. Such long-term fuel contracts are not available at competitive prices, in particular for natural gas contracts. Before markets for natural gas were established around the world, the typical gas contract was a 20-year take-or-pay contract. Although take-or-pay contracts help to pay for dedicated assets, during the 2000s, concerns about security of gas supply, changes in natural gas regulation, and noncompetitive behavior from sellers have resulted in a lower offer of firm gas contracts in Colombia [50]. As a result, gas-fired power generators have been required to have other fuel backups. Since fuel prices and supply are highly uncertain, we argue that the mechanism for estimating firm energy favors hydro power. This is indeed supported by the choice of generation technology observed in the past firm-energy auctions. It also provides another example of how gas and electricity markets are becoming increasingly interdependent and, in this case, what is “bad” for gas would be “good” for electricity. For the last 20 years, power demand in Colombia grew at a yearly average rate above 2%. This, along with the potential for hydro power, made mid- to large-size hydro plants the most efficient option to satisfy a growing demand, both under central planning and market rules (see [51]). Thermal plants are built to run during dry seasons or to supply zones that have transmission constraints. However, given the “bias” in the firm energy mechanism, there might not be enough thermal capacity available for a dry season. Other technologies, in particular renewables, have capital costs higher than hydro and there have been few opportunities to develop them. As we discuss next, small hydro (less than 20 MW) is the only renewable technology that has a noticeable share of capacity and that is profitable in the market.

568

Y. Olaya et al. / Renewable and Sustainable Energy Reviews 56 (2016) 563–571

80 Small Hydro Plants Capacity additions, MW

70 60 50 40 30 20

0

1960 1963 1966 1968 1971 1973 1975 1977 1979 1981 1983 1985 1987 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013

10

Fig. 4. Small-hydro generating plants built between 1960 and 2011 and connected to the power grid. Additions in 2000 include previously built plants that were rehabilitated. Source www.xm.com.co.

4. From big to small hydro Large hydro-power plants are frequently perceived as environmentally unfriendly because they disrupt ecosystems and livelihoods of communities [52–54], even though they have low emissions. Furthermore, mid to large-sized hydro plants have higher capital requirements and longer lead times than thermal plants, which increases uncertainty for the investors. In Colombia, they attracted no private investment during the first years of deregulation; in fact, no private firm was willing to make the necessary financial commitment over the long time horizon in a system where nobody knew how the market might evolve. Instead, a surge in small hydro-power projects was observed during the 2000s, as shown in Fig. 4. The large base of hydro capacity means that electricity prices in Colombia are low for most of the year, i.e. outside the dry seasons, and competitive generation technologies must have low operating costs. In Colombia, hydropower plants with less than 20 MW capacity are classified as small hydro, and under the current regulation, these plants are always scheduled for dispatch at the spot price. Colombia's potential for generating power with small hydro centrals is estimated to be between 8000 and 25000 MW [55]. There are currently 95 small hydro plants in the interconnected system, with a combined capacity of 643 MW [20]. More than 50% of this capacity (387 MW) came on stream between 2001 and 2010 (Fig. 2). The reforms to the power sector have had a positive effect on small hydro deployment; this can be inferred not only for the new capacity being installed but also from the rehabilitation of more than 20 MW of small hydro plants installed before 1960. Since small hydro plants have low operating costs and their capital costs are lower than solar and wind, they are profitable at the Colombian spot market price, while solar and wind plants are not [51,56]. We turn now to discuss renewables, in particular solar and wind. 4.1. Role of (other) renewables Currently, the Colombian generation technology mix relies almost exclusively on hydro resources and thermal power. Since hydro plants generate most of the electricity in Colombia, power generation contributes only 8.5% of the total CO2 emissions in Colombia [57], a low value compared to the 2012 world average of 42% emissions from power generation and heating [58]. Colombia has focused its incentives for renewables on the electrification of rural off-grid zones, as emissions from the power sector are low [59], and the costs of renewable technologies (RETs) are high

compared to conventional technologies [60]. As summarized in Table 2, the provisions for RETs in Colombia's interconnected system are limited to tax exemptions and other alternatives used in other countries, such as feed-in-tariffs, and obligations [61] are not considered [62]. The consequence of this structure is that renewables other than small hydro play a minor role in the Colombian electricity supply. The institutional design has not taken RETs into account, which means they must compete against conventional generation technologies whose environmental and social costs are not fully internalized [63,64]. In general, the mechanisms for integrating RETs with the interconnected system are not clearly defined. This is a problem for wind and other power sources, because the lack of a methodology and rules for calculating their contribution to firm energy means that these sources cannot compete for reliability payments in the firm energy auctions [65]. As discussed, high investment costs are one of the barriers for large scale generation with RETs. While costs of wind and PV have decreased in recent years [61], their decrease has not yet facilitated their adoption in Colombia, as, e.g. in the residential sector, the benefits of PV do not outweigh their costs [66]. The outlook for the future of renewables in Colombia is uncertain, but several steps have been taken that could influence the future adoption of RETs making them more attractive as a generation technology. In May 13th, 2014, a new Law (1715/2014) was issued in order to integrate non-conventional energy into the national power system. This law provides a new legal framework to promote investment, research, development, and use of non-conventional energy sources including: wind, geothermal, PV, and biomass. The law creates tax incentives for developing renewable generation projects beyond the reliability framework of the OEF. The exact mechanism is still not fully agreed, so currently we cannot foresee the effect it will have on the future generation mix. Considering this uncertainty, the best signal is given by the Colombian Energy and Planning Unit, (UPME) in its 2015 generation and expansion plan. The plan is not mandatory, and thus provides only an indication of possible future scenarios. According to the scenarios, RETs' share in the 2028 generation matrix might be between 6% and 15% of installed capacity [36]. As Fig. 5 shows, proposed wind projects (scenarios 9 and 10) could substitute between 24% and 86% of the 1050 MW of coal capacity required in the base highdemand scenario (5). The results in Fig. 5 are consistent with previous research on the competitiveness of wind power in the Colombian wholesale market [64,65]. From a reliability point of view, increasing the share of renewable technologies that do not depend on water might help during dry seasons [51,59,67]. Furthermore, developing wind resources in the Guajira region could relieve some of the current transmission constraints [67]. Generation from sugar cane and African palm biomass is currently competitive [64,36], and further developing this technology would diversify energy supply, although at a smaller scale than wind power. Spot prices are expected to decrease with a higher share of renewables, which would particularly affect thermal generation, implying that an increase in renewables will make future thermal expansions harder. Finally, expansion of renewables is expected to be facilitated by the development of smart grid technologies. Smart grids in Colombia are envisioned as support for energizing isolated rural areas using renewable generation [68], for integrating renewables to the national energy system [69], and for developing an electric car sector [67]. The challenge for regulators is to integrate smart grids with the wholesale market. To date, most smart grid projects have been demonstration projects. InterAmerican Development Bank, IADB, is developing a roadmap for the promotion of smart grids but results are not yet available.

Y. Olaya et al. / Renewable and Sustainable Energy Reviews 56 (2016) 563–571

569

Table 2 State and incentives for renewables in Colombia until 2014.

Wind Biomass – Cogeneration Biomass – Other Geothermal PV a b c

2014 installed capacity (MW)

FIT

Electric utility quota/obligation

Tax exemptions

Surplus energy and energy sales

20a 206a 26b 0 9–11c

No No No No No

No No No No No

Yes Yes Yes Yes Yes

At market price At market price At market price Regulated prices

Interconnected capacity. Data for 2010 [55]. Isolated capacity. Sources: [55,36].

Fig. 5. Alternatives for capacity expansions 2019–2028 in a high-demand scenario. RETs: geothermal, solar (PV) and Wind. Source: elaborated from [36].

5. Discussion and conclusion Large hydropower plants dominate the portfolio for power generation in Colombia. The expansion in thermal capacity observed between 1995 and 1999 resulted from government planning and market intervention rather than from market rationality. With the advantage of hindsight, the need for thermal power was overestimated, as evidenced by the low share of production by thermal plants, and by the number of units retired during the 2000s (more than 400 MW). Electricity demand is expected to continue growing as the country develops, and adding hydropower capacity keeps carbon emissions and electricity prices low – provided that the market behaves competitively. The hydropower plants expected to enter between 2014 and 2019 are supported by the new capacity mechanism, known as reliability charge or firm energy. Although the mechanism allows the installation of any technology that increases resource adequacy, the conditions in Colombia: uncertainty in gas markets and a large base of hydro, favor large hydropower plants. Along with the new large hydropower plants, a boom of small hydro plants has been observed during the 2000s. These small plants take advantage of their low capital cost, low environmental impacts and the fact that all power plants under 20 MW are always dispatched in the market (guarantee of generation). The additions of hydro power that were awarded long-term capacity contracts in 2008 displace thermal generation to the right of the supply curve, increase resource adequacy, and make economic sense. However, hydropower has environmental costs that are not fully accounted for, such as the impacts on communities and the local environment, and it is also vulnerable to weather changes. A more diverse mix of generation technologies could make the power system more reliable, for example, by using the increase of winds during dry seasons to offset the decrease in hydro power [70]. Adding intermittent technologies such as wind or solar power to the interconnected system is one of the challenges for the next regulation periods. The experiences in other

countries suggest that renewable technologies require non-market incentives such as feed-in-tariffs, but there are proposals for supporting these technologies with capacity payments similar to the reliability charges [65,70]. The power market created in the 1990s has greatly improved the financial position of companies in the Colombian power sector. There is evidence of the ability of the power system to satisfy demand with efficiency. Capacity payments, such as the ones applied in Colombia, have been effective in the sense that they have added generating capacity, thereby avoiding black outs. The goal of diversifying the technology mix for generation in order to make the Colombian power system less vulnerable to changing weather has not been achieved by the process of deregulation. Although technology diversification can be attained in competitive markets, the high uncertainty in costs, demand and technologies, and the imperfections of power markets may cause actual choices to differ from optimal choices. Regulators are expected to set up clear and stable rules for investing in new generation capacity. Regulators often intervene in the market with incentives that are intended to move the system towards what is, at that moment, seen as the desirable state. Many interventions, however, lead to unintended consequences [71] as the market might not always respond in the foreseen way, e.g. in the case of Colombia, we can actually observe that twenty years after deregulation the system has less thermal generation than in the initial period. It is also important that the regulator does not constantly intervene in the system as this creates additional uncertainty and risk for long-term investments. This is important because, as the Colombian case evidences, investors respond to regulatory uncertainty by delaying decisions until new information is available. The impact of the different periods of regulation in Colombia shows that the power system has improved, the debt levels are far from what we observed before the deregulation, the companies are relatively healthy today and, so far, there is enough capacity to satisfy demand. From this point of view we conclude that the changes in

570

Y. Olaya et al. / Renewable and Sustainable Energy Reviews 56 (2016) 563–571

regulatory regimes have had at least some desirable effects, in that the last change prompted a further round of investments. However, from the reliability perspective there might still be a number of uncertainties, as the change in regulation has not created the desired change in the capacity mix. There is always a tradeoff between the reserve margin, cost, and reliability, e.g. if Colombia has a significant reserve margin mainly based on hydro then reliability might not be an issue; however, the consumers will have to pay for this excessive reserve margin. On the other hand, if the regulator manages to increase the thermal generation capacity through capacity or reliability payments, this might also increase the cost. The tradeoff between these two options has to be carefully weighed to decide which is the most cost-efficient for the same reliability.

References [1] Sioshansi FP, Pfaffenberger W. Why restructure electricity markets? In: Sioshansi FP, Pfaffenberger W, editors. Electricity market reform: an international perspective. 1st ed.. Amsterdam: Elsevier Science; 2006. p. 35–48 978-0-08045-030-8. [2] Chick M. Electricity and energy policy in Britain, France and the United States since 1945. 1st ed.. . Cheltenham, UK: Edward Elgar Publishing; http://dx.doi. org/10.4337/9781847206893. [3] OECD, IEA, NEA. Technology roadmap: nuclear energy. Paris: 2015. doi:10.1007/SpringerReference_7300. [4] Arango S, Dyner I, Larsen ER. Lessons from deregulation: understanding electricity markets in South America. Util Policy 2006;14:196–207. http://dx. doi.org/10.1016/j.jup.2006.02.001. [5] Roques FA. Market design for generation adequacy: healing causes rather than symptoms. Util Policy 2008;16:171–83. http://dx.doi.org/10.1016/j.jup.2008.01.008. [6] Roques FA, Newbery DM, Nuttall WJ. Investment incentives and electricity market design: the British experience. Rev Netw Econ 2005;4:93–128. [7] OECD/IEA. Power generation investment in electricity markets. Paris; 2003. [8] OECD/IEA. Tackling investment challenges in power generation in IEA countries. Paris; 2007. doi:10.1787/9789264030084-en. [9] Olsina F, Garcés F, Haubrich H-J. Modeling long-term dynamics of electricity markets. Energy Policy 2006;34:1411–33. http://dx.doi.org/10.1016/j.enpol.2004.11.003. [10] Euroelectric. Powering investments: challenges for the liberalised electricity sector; 2012. [11] Sweeney James L. The california electricity crisis. 1st ed.. Stanford, California: Hoover Institution Press Publication; 2002. [12] Dyner I, Larsen ER. From planning to strategy in the electricity industry. Energy Policy 2001;29:1145–54. http://dx.doi.org/10.1016/S0301-4215(01)00040-4. [13] Stoft S. Power system economics. New York: Wiley-IEEE Press; 2002. [14] Arango S, Larsen E. Cycles in deregulated electricity markets: empirical evidence from two decades. Energy Policy 2011;39:2457–66. http://dx.doi.org/ 10.1016/j.enpol.2011.02.010. [15] Moreno R, Barroso L a, Rudnick H, Mocarquer S, Bezerra B. Auction approaches of long-term contracts to ensure generation investment in electricity markets: lessons from the Brazilian and Chilean experiences. Energy Policy 2010;38:5758–69. http://dx.doi.org/10.1016/j.enpol.2010.05.026. [16] Hunt S. Making competition work in electricity. 1st ed.. New York: Wiley; 2002. [17] Adib P, Schubert E, Oren S. Resource adequacy: alternate perspectives and divergent paths. In: Sioshansi Fereidoon, editor. Competitive electricity markets. Design, implementation, performance. 1st ed.. Oxford: Elsevier Science; 2008. p. 327–62. http://dx.doi.org/10.1016/B978-008047172-3.50013-1. [18] Cramton P, Stoft S. A capacity market that makes sense. Electron J 2005;18:43– 54. http://dx.doi.org/10.1016/j.tej.2005.07.003. [19] UPME. Cobertura de Energía Eléctrica a 2014; 2015. [20] XM. Descripción del sistema eléctrico Colombiano. Informacion operativa y comercial, /http://www.xm.com.co/Pages/DescripciondelSiste maElectricoColombiano.aspxS; 2015 [accessed 23.01.15]. [21] CIA. The World Factbook. World Factb, 〈https://www.cia.gov/library/publica tions/resources/the-world-factbook/index.html〉; 2014 [accessed 20.10.14]. [22] DANE. Estimaciones de población 1985–2020 y proyecciones de población 2005–2020 total municipal por área, 〈http://www.dane.gov.co/index.php/ poblacion-y-demografia/proyecciones-de-poblacion〉; 2015. [23] IEA. Colombia: electricity and heat for 2012, 〈http://www.iea.org/statistics/sta tisticssearch/report/?country¼Colombia&product¼electricityandheat〉; 2015. [24] OLADE. Precios Promedio Anual, Sistema de Información Económica Energética, 〈http://sier.olade.org/consultas/visor_reportes_d3.aspx? oc¼ 360&or¼640&ss¼2〉; 2015. [25] Larsen ER, Dyner I, Bedoya L, Franco CJ. Lessons from deregulation in Colombia: successes, failures and the way ahead. Energy Policy 2004;32:1767–80. http: //dx.doi.org/10.1016/S0301-4215(03)00167-8. [26] Industria y banca. líderes de la inversión nacional en el exterior. El Tiempo 2014 /http://m.eltiempo.com/economia/sectores/industria-y-banca-lideres-de-lainversion-nacional-en-el-exterior/14288059/2S.

[27] Sandoval AM. Monografía del sector de electricidad y gas colombiano: condiciones actuales y retos futuros. Archivos de Economía, Doc. 272, 1-121 (Bogotá), 2004. [28] Banco de la República de Colombia. Deuda Publica Externa de Mediano y Largo Plazo. Banco La República Colomb Finanz Públicas, 〈http://www.banrep. gov.co/series-estadisticas/see_finanzas_publi.htm〉; 2011. [29] Ochoa Franco FJ. Incidencia de los problemas financieros en los racionamientos de energía eléctrica. Lect Econ 1986;19:33–96. [30] Banco de la República de Colombia. PIB, producción, salarios y empleo; 1998. [31] Garay LJ, Quintero LF, Villamil JA, Tovar J, Fatat A, Gómez S, et al. Colombia: estructura industrial e internacionalización 1967–1996. Bogotá, D.C.: Biblioteca Virtual del Banco de la República; 1998. p. 2004. [32] ISA. Plan de Expansión de Referencia Generación-Transmisión 1993 Revisión. Medellín: UPME ;1993. [33] Congreso de la República. Ley 142 de 1994. Por la cual se establce el régimen de los servicios públicos domiciliarios y se dictan otras disposiciones. Colombia: Diario Oficial no 41.433; 1994. [34] Congreso de la República. Ley 143 de 1994. Por la cual se establece el régimen para la generación, interconexión, transmisión, distribución y comercialización de electricidad en el territorio nacional, se conceden una autorizaciones y se dictan otras disposiciones en materia energ. Colombia: Diario Oficial no 41.342; 1994. [35] Bacon RW, Besant-Jones J. Global electric power reform: privatization and liberalization of the electric power industry in developing countries. Annu Rev Energy Environ 2001;26:331–59. http://dx.doi.org/10.1146/annurev.energy.26.1.331. [36] UPME. Plan de Expansión de Referencia Generación-Transmisión 2014–2028; 2015. [37] Caicedo E. El fenómeno del Niño y su posible impacto en Colombia. Bogotá, D.C.; 2007. [38] Mejía Mazuera J. APAGÓN VALE $ 342.000 MILLONES. El Tiempo; 1992. [39] CREG. Resolución 116 de 1996. Colombia: Comisión de Regulación de Energía y Gas; 1996. [40] CREG. Resolucion 113 de 1998. Colombia: Comisión de Regulación de Energía y Gas; 1998. [41] CREG. Resolución 111 de 2000. Colombia: Comisión de Regulación de Energía y Gas; 2000. [42] Rivier M, Vásquez C, Enamorado JC, Pérez Arriaga JI. Documento Final Estudio Cargo por Capacidad realizado por la U. Bogotá: Comillas de Madrid; 2000. [43] TERA. Escisión del SIC-Bolsa de Energía Eléctrica de las Actividades Comerciales de Interconexión Eléctrica S.A. E.S.P y definición del Esquema Institucional para las demás funciones del Actual Centro Nacional de Despacho. Bogotá, D.C.; 2001. [44] Cramton P, Stoft S. Colombia firm energy market. In: Proceedings of the 40th Annual Hawaii International Conference on System Science. HICSS 2007. IEEE; 2007. p. 124. doi:10.1109/HICSS.2007.133. [45] CREG. Resolución 071 de 2006. Colombia: Comisión de Regulación de Energía y Gas; 2006. [46] Restrepo EMI, Arango S, Vélez ALG. La confiabilidad en los Sistemas Eléctricos Competitivos y el Moelo Colombiano de Cargo por Confiabilidad. Cuad Econ 2012;31:199–222. [47] Ausubel LM, Cramton P. Using forward markets to improve electricity market design. Util Policy 2010;18:195–200. http://dx.doi.org/10.1016/j.jup.2010.05.004. [48] Arango S, Castañeda J a, Larsen ER. Mothballing in power markets: an experimental study. Energy Econ 2013;36:125–34. http://dx.doi.org/10.1016/j. eneco.2012.11.027. [49] CREG. Análisis de Asignación de OEF 2014–2015. Bogotá, D.C.; 2010. [50] Frontier Economics. Diagnóstico fallas del mercado de gas natural de Colombia. Bogotá, D.C.; 2010. [51] García H, Corredor A, Calderón L, Gómez M. Análisis costo beneficio de energías renovables no convencionales en Colombia; 2013. [52] Klimpt JÉ, Rivero C, Puranen H, Koch F. Recommendations for sustainable hydroelectric development. Energy Policy 2002;30:1305–12. http://dx.doi. org/10.1016/S0301-4215(02)00092-7. [53] Caetano de Souza AC. Assessment and statistics of Brazilian hydroelectric power plants: dam areas versus installed and firm power. Renew Sustain Energy Rev 2008;12:1843–63. http://dx.doi.org/10.1016/j.rser.2007.04.005. [54] Sternberg R. Hydropower: dimensions of social and environmental coexistence. Renew Sustain Energy Rev 2008;12:1588–621. http://dx.doi.org/ 10.1016/j.rser.2007.01.027. [55] CORPOEMA. Formulacion de un Plan de Desarrrollo para las Fuentes no Convencionales de Energía en Colombia, PDFNCE. Volumen 2, Diagnóstico. vol. 2. Bogotá, D.C.; 2010. [56] Arango S, Larsen ER. The environmental paradox in generation: how South America is gradually becoming more dependent on thermal generation. Renew Sustain Energy Rev 2010;14:2956–65. http://dx.doi.org/10.1016/j. rser.2010.07.049. [57] IDEAM. Segunda Comunicación Nacional ante la Convención Marco de las Naciones Unidas sobre Cambio Climático. Colombia; 2007. [58] International Energy Agency (IEA). CO2 Emissions from fuel combustion highlights. Paris; 2014. [59] UPME. Integración de las energías renovables no convencionales en Colombia Integración de las energías en Colombia; 2014. [60] GEA Writing Team. Global energy assessment. toward a sustainable future. 1st ed.. Cambridge: Cambridge University Press; http://dx.doi.org/10.1017/ CBO9780511793677. [61] Foley T, Thornton K, Hinrichs-rahlwes R, Sawyer S, Sander M, Taylor R, et al. Renewables 2015. Global Status Report. Paris: REN21; 2015.

Y. Olaya et al. / Renewable and Sustainable Energy Reviews 56 (2016) 563–571

[62] Zuluaga MM, Dyner I. Incentives for renewable energy in reformed LatinAmerican electricity markets: the Colombian case. J Clean Prod 2007;15:153– 62. http://dx.doi.org/10.1016/j.jclepro.2005.12.014. [63] Ruiz B, Rodríguez-Padilla V. Renewable energy sources in the Colombian energy policy, analysis and perspectives. Energy Policy 2006;34:3684–90. http://dx.doi.org/10.1016/j.enpol.2005.08.007. [64] Caspary G. Gauging the future competitiveness of renewable energy in Colombia. Energy Econ 2009;31:443–9. http://dx.doi.org/10.1016/j. eneco.2008.12.007. [65] Botero BS, Isaza CF, Valencia A. Evaluation of methodologies for remunerating wind power's reliability in Colombia. Renew Sustain Energy Rev 2010;14:2049–58. http://dx.doi.org/10.1016/j.rser.2010.02.005. [66] Radomes A a, Arango S. Renewable energy technology diffusion: an analysis of photovoltaic-system support schemes in Medellín, Colombia. J Clean Prod 2015;92:152–61. http://dx.doi.org/10.1016/j.jclepro.2014.12.090. [67] UPME. Plan Energético Nacional Colombia: Ideario Energético 2015. Bogotá, D.C.; 2015.

571

[68] Gaona EE, Trujillo CL, Guacaneme J a. Rural microgrids and its potential application in Colombia. Renew Sustain Energy Rev 2015;51:125–37. http: //dx.doi.org/10.1016/j.rser.2015.04.176. [69] Naranjo C, Alzate JD, Salazar J. Future scenarios of smart grids in Colombia and their impact on climate change. In: Proceedings of the 2011 IEEE PES Conference on Innovative Smart Grid Technologies (ISGT Latin America). IEEE; 2011. p. 1–8 (UNFCC: United Nation Framework for Climate Change). doi:10.1109/ISGT-LA.2011.6083200. [70] Vergara W, Deeb A, Toba N, Cramton P, Leino I. Wind energy in Colombia. A framework for market entry. Washington, D.C.; 2010. doi:10.1596/978-0-82138504-3. [71] Sioshansi FP. Electricity market reform-progress and remaining challenges. Competitive electricity markets. Design, implementation, performance, vol. 1987. Oxford: Elsevier Ltd.; 2008. p. 1–23. http://dx.doi.org/10.1016/ B978-008047172-3.50004-0.