By Elbia Melo
Note: This is an abridged and updated version of a paper written by Elbia Melo, Executive President of ABEEólica (the Brazilian Wind Energy Association), for the Instituto de Estudos Avançados da Universidade de São Paulo (Institute of Advanced Studies of the University of São Paulo). Original print version ISSN 0103-4014
Wind energy has experienced exponential growth in Brazil. Between 2009 and 2012, wind projects participated in six energy auctions, securing contracts to supply 7GW of power. Installed wind capacity in Brazil will reach over 8.4GW by 2017, 3.5 times current levels – attracting investment of over R$25 million.
Historically, the Brazilian Government provided support for wind power under the PROINFA programme, which began in 2004. PROINFA was based around finance to support regional development policies and resulted in the commissioning of 1,422.9MW of wind power, at subsidised prices.
Wind first participated in competitive auctions in 2009 and became firmly established in the Brazilian energy matrix in 2011, with 2,905MW contracted – more than sufficient, in terms of critical mass, to support a supply chain (which would require annual new demand of around 2GW). Prices fell to average levels of R$100.00/MWh that year, establishing wind as the second most competitive energy source in the country.
In 2012 Brazil passed the 2GW mark of installed capacity, ending the year with 2.5GW of installed wind power, or 2% of the overall energy matrix. During 2012, 40 new wind farms were installed, involving a total of 108 projects, and adding 1GW to the system – as much capacity in one year as had been installed over the previous 13 years (1998-2011). Wind power in Brazil now has the capacity to supply four million homes.
Wind energy is clean, renewable and generates jobs and income for Brazil. The industry created 15,000 direct jobs in 2012 and there are now eleven equipment manufacturers based in Brazil. In 2012 some R$7 billion was invested in the sector and investment is forecast to reach R$50 billion by 2020. Job creation in less developed regions is amongst the socioeconomic benefits of wind power – as is the leasing of land in areas including the semi arid regions, where landowners are guaranteed an income for at least 20 years.
Brazil has an estimated potential capacity of 300GW of wind power. The country commissions around 6GW of new power annually through the energy auction system, when GDP growth is running at average levels, and there is enormous potential for wind energy to meet some of this growth in demand.
Brazil’s energy matrix is dominated by clean and renewable energy, mainly hydro power, which complements wind power. Of Brazil’s total energy, 45% comes from sources that do not emit CO2, compared with a global average of 20%, and its options for the cost effective generation of clean energy include hydro power, cogeneration, biomass and wind power. Prices in the energy auctions are an important factor determining future energy sources for Brazil.
Recent market share gains for wind in Brazil can be attributed to a combination of factors, including:
- Technological advances in the industry – such as increases in the height of wind turbine towers (from 50 metres to 100 metres) and increases in the diameter of the blades and rotors, which favour the particular characteristics of Brazilian winds.
- The introduction of attractive financing terms for projects securing contracts through the national energy auction system.
- The international economic crisis, which resulted in a reduction of investment in renewable energy/cuts in renewable energy subsidies in Europe and the US. This led to increased competition in Brazil, with project developers and suppliers prepared to accept lower returns, since it was one of the few regions still attracting investment, in order to establish market share.
The table below shows the results of the auctions and the subsequent commissioning of the wind power energy contracted in the auctions held since 2009.
Figure 1 – Presence in auctions
In terms of prices, the 7GW contracted since 2009 was significantly lower than the prices achieved during the first projects, developed under PROINFA. Prices for the 1.4GW contracted through PROINFA were up to three times higher than those in the most recent auction, as the following graph shows.
Figure 2 – Average of prices for wind
Industry competitiveness is illustrated by a decline of almost 50% in the average level of investment (total CAPEX) over the last eight years. Capital expenditure was R$6 million per MW installed under PROINFA: it has fallen to R$3.5 million per MW installed in recent projects. This can be explained by improvements in technology in the industry but also – significantly – by the influx of wind turbine manufacturers. The number of turbine manufacturers in Brazil jumped from just two in 2009 to eleven in 2012.
An annual demand level of 2GW of installed capacity, the average contracted in the auctions since 2009, implies a market of over R$7 billion per year. This takes into account only domestic market potential and assumes production of 1,000 turbines, 1,000 towers and 3,000 blades annually. Based on the assumption of around 15 jobs per MW installed, throughout the supply chain, there is scope for the creation of 280,000 direct and indirect jobs by the end of 2020.
Figure 3 – Employment in wind sector
Source: Simas, Moana (2012)
The potential is massive – but there are many challenges, and a responsible approach on the part of the industry is essential. The wind industry faced significant bottlenecks throughout 2012, including logistics, for transport and transmission (notably the delays in the ICGs). Currently in Brazil, there has been a mismatch between the construction of wind farms and the construction of the transmission grid for the energy generated. This situation compromises the effective development of the sector. The industry was also hit by the revision of the criteria for ‘FINAME’ funding by BNDES (which requires specific levels of industrialisation), successive delays in auction dates, and the publication of MP 579, which determines the energy reallocation in existing portfolios of power distribution companies, compromising the calculations of these companies for future markets.
Despite the challenges, in 2013 Brazil is set to rise in the global rankings from number 16 to number 10, in terms of installed capacity of wind power.
Challenges to the Brazilian wind market
Brazil ended 2012 with 2.5GW of installed wind power, or 2% of the total Brazilian energy generating capacity, and by the end of 2017, taking into account the power contracted up to 2012, it will reach 8.7GW of installed capacity – equivalent to 5.5% of the national total. However, there was a relative slowdown in growth. Only one A-5 auction took place, in December 2012 – and just 574.3MW of energy was contracted in that auction, well below the average of 6GW required annually to meet the growing demand for energy in Brazil. Just 281.9MW of the energy contracted was wind power.
Low demand also resulted in very low prices: supply was 28 times greater than demand and average prices in the auction were just R$90.00 per MWh. This was very disappointing for the wind industry. Wind had previously supplied around 2GW/year through the auctions since 2009 – and a supply chain has been established to support that level of demand.
The 2012 auction should not be considered an indication of the future of the industry. Prices for wind power had fallen in 2011 to levels of R$105.00 per MWh (in current prices). Fierce competition, in 2011, had driven prices to an absolute minimum, curtailing profit margins. Prices below R$100.00 per MWh may not be sufficient to support the industry.
The electricity distribution companies are required to submit projections for demand for future auctions (either A-3 or A-5) based on a three to five year view. They had been submitting projections for more energy than they would require since 2010, ahead of the auctions for large hydro electric projects (Jirau, Santo Antônio and Belo Monte). Since these big projects are auctioned in one large batch, utilities have projected energy quotas which are higher than they need in order to secure energy supply from these projects, with competitive energy prices. Combined with slower than expected GDP growth, distributors had been contracting for more energy than they required. Lastly, Provisional Measure 579, by announcing a reallocation of the existing energy portfolios of distributors, led to considerable uncertainties about their quotas, affecting the calculation of the distribution for the future market.
The FINAME/BNDES financing model also contributed to uncertainty and volatility in the sector throughout 2012. BNDES introduced major changes to the structure by which it agrees to offer finance for wind energy, resulting in uncertainty surrounding the security of some recently contracted projects.
BNDES suspended six of the eleven manufacturers of wind turbines for failing to meet the required level of local production under the FINAME rules. Although within its rights to do so, the impact on a still immature industry resulted in a substantial reduction in the number of suppliers, decreasing the domestic supply of equipment and reducing competition. The biggest issue, however, was that the BNDES measures resulted in risks to projects with contracts already signed through the A-5 and A-3 auctions since 2009 – projects which were already under construction and in most cases were impossible to be reversed.
Lower demand for both energy and equipment, combined with fewer suppliers, have resulted in a major adjustment to the wind energy market, bringing it back into equilibrium. The influx of manufacturers over the last few years has raised questions about the ability of Brazil to absorb the production potential, raising doubts over whether there is space for so many players over the medium term.
Consolidation and sustainable growth of the Brazilian wind energy industry
The Brazilian wind industry is passing through a crucial consolidation phase. It is clear that, although there are eleven manufacturers, each with average production capacity of 500MW per year of equipment, this does not necessarily mean the supply of 5,000MW of machines to the market, for three reasons.
- Not all of the eleven manufacturers are fully established in Brazil, so their respective manufacturing capacity varies, depending upon the speed at which they are expanding – which depends upon the orders they have contracted.
- There is effectively no market for manufacturers which do not comply with the FINAME regulations and whose equipment cannot therefore be financed by BNDES. Interest rates on other financing sources are around four percentage points higher – highly significant in such a competitive industry.
- The wind characteristics in different regions of Brazil vary significantly. For any given site, there is an optimal design (layout) associated with specific equipment. A decision to work with a different equipment provider is possible, but is increasingly difficult as a project advances and the deadline for the supply of energy to the grid approaches. Turbines are not commodities – one machine is not necessarily a perfect substitute for another and the project developer cannot switch manufacturer easily. The industry therefore demonstrates the structure of a ‘differentiated oligopoly’.
In the short and medium terms, Brazil does not have sufficient manufacturing capacity to meet domestic demand, taking into account the different characteristics of equipment and limited flexibility for projects, already contracted under the auctions at extremely competitive prices, to be redesigned.
Most Brazilian wind equipment manufacturers have sought to raise the percentage of local content, to ensure that their clients qualify for financing from BNDES. However, this is a new industry and there are few local suppliers of the key components, sub-components and parts to guarantee entirely local production of wind turbines. Imports of certain key items in the supply chain therefore remain important.
BNDES is aware of this situation and recently established new rules under which wind turbines become eligible for funding. It has created a programme for the phased move to local production between 2012 and 2015, by when turbine models which incorporate a gearbox must demonstrate that the gearbox, generator and DFIG panel are produced locally. For models without a gearbox, local production of certain parts of the generator, the structure of the nacelle or inverter will be required by 2015. Technology transfer is a challenge in itself – and the creation and consolidation of a local supply chain is slow. However, the Brazilian Government’s industrial policy (which BNDES administers) should be fully implemented for the wind power industry by 2015.
BNDES requires manufacturers to submit a business plan for local manufacturing (by July 2013), to undertake the civil construction and purchase the machinery necessary for the manufacture of nacelles (by January 2014), and to be manufacturing the nacelles locally (by July 2014).
Brazil has the potential to install 2GW of wind power capacity annually, but there is neither time nor sufficient scale for the development of a complete supply chain. 2MW turbines that have been installed recently require innovative components, which must be imported. However, the significant potential opportunities for a local supply chain should be stimulated.
In mid-2008 there was just one wind turbine manufacturer in Brazil. Today there are eight manufacturers already producing, two with plants under construction and others at the planning stage – and there are now thirteen manufacturers active in the domestic market. Six wind tower factories are operational and one is being constructed. In addition to these, the second largest manufacturer of wind blades in the world, Tecsis, is a Brazilian company.
The Global Wind Energy Outlook from the Global Wind Energy Council (GWEC) concludes that Latin America has excellent quality winds with great potential for power generation. The chart below from that GWEC report outlines three growth scenarios. Under the moderate scenario, installed capacity in the region of 28GW is projected by 2020. Assuming that Brazil has capacity of 14GW by 2020, the remaining 14GW is represented by other countries – indicating significant investment and the need for a supply chain.
Figure 4 – Installed capacity in Latin America
Brazil leads the development of wind power in Latin America, but other countries are also investing. Chile had almost 170MW of capacity in operation in late 2009 with a wind potential of more than 40GW. Peru, with almost 150MW under construction in late 2009, has the potential of more than 20GW. Uruguay has a target of 500MW by 2015 and has potential for more than 1.5GW. There are a number of initiatives in Central America and the Caribbean, and other Latin American countries such as Argentina, Venezuela, Colombia, Ecuador and Bolivia are investigating the potential for wind, which could mean a significant potential market for Brazil as an equipment supplier – which would contribute to the longer term sustainability of this industry.
Brazil has already attracted eleven manufacturers of wind turbines and companies in the supply chain and may soon have production capacity in excess of domestic demand. The major investment needed for a manufacturing industry and economies of scale to attract manufacturers do not exist in other Latin American or Caribbean countries - so Brazil is well placed to become a major hub for the production and export of wind generating equipment. Brazil has a unique opportunity to attract investment capital and technology and reap the economic benefits that an industry of this scale can supply. To do so, however, it is essential to encourage innovation.
Innovation and competition
Economic theory demonstrates that competition is essential to sustain a strong, established, innovative, export oriented industry and the associated industrial structures which require large capital investment. To ensure competition, it is crucial for the Brazilian wind industry to remain open to imports. Economies of scale, cost advantages for incumbent firms and strong product differentiation within this highly concentrated industry require a competitive environment – which won’t exist if firms are protected by barriers to entry.
Economic theory accepts that innovation is stimulated by an intermediate level of market concentration. On the one hand, a highly fragmented market fails to leverage resources and companies are not interested in innovation; on the other, if monopolies are tolerated companies have no fear of being replaced and therefore fail to innovate. Clearly the potential threat of competition, at least, is important – and a structure of oligarchy is considered to be the best scenario to stimulate innovation.
We can conclude that, if domestic equipment manufacturers are close to the technological frontier throughout the industry globally, they can react to incentives and changing conditions. They will not be forced out of their comfort zones without competition. A highly inefficient national industry would not be able to act, so there is a clear trade-off. Permitting imports risks ‘killing off’ the national industry: but equally, imports result in productivity improvements and lower production costs. Banning imports, on the other hand, ensures that the national industry survives, but in a ‘locked-in’ scenario and in a state of equilibrium combining low productivity with high prices.
It is worth pointing out that the global wind energy industry has been developing economically viable technology with significant levels of technology investment only since the mid-1990s.
Figure 5 – World historical evolution of wind technology
Source: EPE, 2012
For the Brazilian wind power industry, technological improvements are critically important since costs per MW installed are very high relative to China and Europe, as shown below:
Figure 6 – Comparison of installation prices
*Prices of construction of towers per MW are still high compared to global standards
Source: Roland Berger, 2012.
High wind turbine costs in Brazil are influenced by local material costs, the scale of production as well as the ‘Brazil Cost.
Reductions in turbine costs offer the potential to bring down the total cost of wind energy production – and the high costs underline the need to encourage technological progress, either in terms of direct investment in research and development and innovation, or by ensuring that Brazil remains relatively open to receiving foreign investments. Brazil still has a long way to go to create a competitive market and ensure the long term sustainability of a wind power manufacturing industry – and one factor will be to make wind turbines in Brazil more competitive.
High CAPEX costs in Brazil have not prevented the wind industry from creating the most competitive prices per MWh in the auctions. There are two explanations for this.
The first is the global context of economic crisis and the sharp decline in investment in renewable energy (see above). With no orders coming in from their main markets in the west and with a full inventory, equipment manufacturers had to seek alternative markets in developing countries, and in particular in the BRIC countries. While China, the fastest growing wind energy market, might have appeared to be an attractive alternative, it is dominated by local suppliers. Thus, European and North American manufacturers of wind turbines focused their sales efforts on new markets such as South America. Brazil, offering a scenario of sustainable economic growth and steadily increasing demand for power (60,000MW of installed power demand forecast over the next ten years), became the focus of attention as a location for investment by manufacturers of equipment.
A large number of manufacturers have arrived in Brazil since 2009 – with a corresponding drop in turbine prices as companies took the strategic decision to enter the market aggressively and supply equipment from stock at low prices (with the intention, in the future, to establish a manufacturing plant). Initially there was a tendency to supply technologically second rate equipment. However, since the auctions in 2010, and especially since the 2011 auctions, this has changed and manufacturers now supply the most up-to-date equipment.
The second factor contributing to the disparity between the low contracted prices of 2011 and 2012 and the capital expenditure costs is related to the Brazilian auction system. Many countries encourage the adoption of wind energy through feed-in tariff models, based on a fixed energy price and variable revenues. As feed-in tariffs are generally very high, investors seek to maximise the energy produced by the turbine – building increasingly larger turbines. This ensures the maximum potential use of local wind available, but implies a high unit cost of investment. Large generators often operate for a long time below average capacity. The project therefore returns a higher ratio in terms of MW by local area but a lower capacity factor.
Feed-In tariff schemes offer a clear incentive to increase energy production, which is remunerated by attractive tariffs. Thus, they naturally encourage technology development. In contrast, under the Brazilian auction system the price depended upon the competition, especially since 2009 until the present day. Operators received a ;fixed; revenue, based on the expected annual energy production. To achieve low auction prices, developers were forced to minimise their investment, focusing on the deployment of smaller turbines or a lesser number of turbines per project. Turbines with larger blades and smaller generators are cheaper, but do not maximise the potential use of the local wind. The typical Brazilian configuration ensures that generators operate more of the time near the average capacity than larger ones. This implies a lower ratio of MW per local area and higher capacity factor.
So Brazil’s contracting mechanism, in contrast to feed-In tariff systems (or indeed the former Brazilian PROINFA model), did not necessarily optimise the use of the wind resources.
The wind industry is in its early days, both in Brazil and globally, and there is great potential for growth, underlining the case against the imposition of high barriers to entry, because the absence of competition could restrict the potential for the development of the market.
For a long term energy mix, we must think in terms of a model that prioritises renewable and clean energy, taking into account the issue of security of supply. In this sense, it is important to reconsider an auction model where price is the sole criterion.
After a difficult year in 2012, the Brazilian wind industry is expected to resume its growth rate in 2013, underpinned by GDP growth of around 4% and a resurgence of demand for power contracts. The industry should achieve its goal to contract 2GW for this year and in future years, creating a sustainable industry for the longer term.