Electricity and macroeconomics in Brazil: the end of the golden age

The fourth new energy auction in Brazil took place last Wednesday (July 26), with contracts for delivery starting in 2010. Its results (Table 1) displayed some features that deserve attention:  



(i)    There was no submission of bids by hydro plants, even though two large hydro plants and one PCH had pre-qualified to participate;

(ii)    The thermo plants (12 in total) were all fuel-oil;

(iii)    101.8% of the auction demand was attended and, according to EPE President Mauricio Tolmasquim, the new capacity contracted in the auction, together with the one from the June 18th renewable energy auction, will be enough to satisfy 2010 demand and “hold deficit risk to 5%”; but

(iv)    The auction’s average final price (R$ 134.67/MWh, reflecting the single participation of thermal plants) was higher than the average prices of the previous new energy auctions (R$122.05, R$ 128.12, and R$ 128.90 per MWh, respectively for the first, second, and third new energy auctions). 

        Rowe Michels and Stewart Ragar, from Bear Sterns, have spotted a trend of rising prices as a result of higher expected generation prices around the end of the decade, as the auctions moved ahead in time, currently focusing on the electrical energy to be delivered in that period. Maybe an expectation of even higher prices at subsequent auctions or with sales to free consumers motivated the pre-qualified hydro suppliers to end up deciding to stay out.        

         What explains this stepping up of electricity prices? Does it share any common factors with the current tightening of energy supply in Chile and Argentina?            

        Brazil has benefited in the last few decades from a relatively cheap energy matrix, because of the weight of hydropower in electricity supply. The installed capacity of hydro plants was about 70% of the total in 2004, whereas gas (10%), oil (4.8%), biomass (3.1%) nuclear (2%), coal (1.4%), wind (0.03%), and imports (8.15%) accounted for the remainder.

        Together with a higher use of sugar cane-based ethanol as vehicle fuel in the last few years, that hydro share has also implied a relatively clean total energy matrix, as compared to other countries, both in terms of local and global pollutions. The share of renewable energy is significantly high (Table 2): 43.9% in 2004, against 13% in the world.


             Source: www.centroclima.org.br

        The increase of marginal costs of energy at the end of the current decade, as apparently reflected in the new energy auctions heretofore held, reflects two main factors. One is the low rhythm of hydro investments in the last decades, investments that have obviously a longer maturation period than alternative sources. The scarcity of hydro investments is starting to bind again as a restriction, after the efficiency gains accrued in the post 2001- rationing have all been incorporated. It is key to keep in mind that that low pace did not derive from exhaustion of potential hydro resources, as it has occurred in many other parts of the world, but from newly enshrined environment concerns, and mainly from financial and regulatory reasons (an issue for another blog).   

          Furthermore, risks associated to gas supply that increased with the age of “resource nationalism” in Latin America are also at play. It is not by chance that all thermal plants were oil fuel-based in Brazil’s auction last Wednesday, whether or not the absence of new gas thermal plants is to continue. Gas-supply risks have also acquired a high profile in Argentina and Chile, in which – as in Brazil – gas consumption has been growing at double-digit rates for several years. And oil-based thermal energy cannot be relied upon as a cheap alternative in the following years.

          Notice as well that the above-mentioned trends imply a less clean Brazilian energy matrix at the margin along the remainder of the decade. Less hydro and more oil-fuel plants mean higher CO2 emissions and consumption of non-renewable energy.            

          What about the 2010s?               The landscape might change after 2011 if the pipeline of already mapped potential hydro investments comes to fruition and the phase of low hydro investments is left behind. Alternative sources of power (wind, solar, biomass etc.) tend to rise as items of the energy matrix but the bulk of the absolute increase in electricity supply along the next decade will most likely come from a combination of hydro, gas-fired and oil-fuel plants. Among the latter, hydro electricity offers the most attractive combination of (low) carbon emissions and risk-adjusted forward costs, despite its peculiar costs of compliance to environment safeguards and upfront investment outlays.

           Take for instance the case of the two Madeira River’s dams that have recently received clearance by Brazilian environmental regulators – Santo Antonio and Jirau – which together might generate 6,450 MW from 2012 onwards, provided that the plants start being built in mid-2008 (for the sake of comparison, recall that Itaipu’s capacity is of 14,000 MW). Most energy pundits believe that the price floor at auctions for the energy generated in those two plants will be lower than the average hydro prices that have recently prevailed, despite costs of compliance with the series of new safeguards required by regulators, as well as the high costs of transmission from the Amazon to consumer areas. Nonetheless, it is reasonable to expect the old age of ultra-cheap energy costs to be definitely gone in Brazil.

           Thus, assuming that the energy-security premium and other forward costs of gas-fired energy plants will not come down substantially in the foreseeable future, the future of Brazil’s electricity costs will hinge upon its ability to overcome regulatory and financial barriers to hydro investments, as well as to comply in an efficient way with the heavy environmental safeguards required by hydro plants. By the way, one should keep the mind two environment-related points:

·        Well-crafted large dams usually have a smaller environment impact than the one derived from a number of small hydro plants generating the equivalent amount of energy; and

·        Given that it is unrealistic to expect alternative new sources of energy to be capable of fully substituting for conventional sources in the near future, the benchmark for the opportunity costs of avoiding hydro plants must be built in terms of higher use of nuclear and fossil fuels.            

          What are the macroeconomic implications of more costly electricity?          

          At the level of relative prices, the comparative advantage of energy-intensive production processes tends to be reduced, affecting particularly segments of natural-resource processing activities. Before any reader or fellow blogger spots any conflict of this conclusion with my previous blog on the “Dutch Disease” – where I argued that natural resource-based activities tend to be prevail as comparative advantages of the region in the current international labor division – let me remind her/him that the Brazilian edge in terms of lower energy costs will still remain, even if a return to past levels cannot happen. This also applies to the other technologically more highly graded and skilled-labor intensive activities that can be developed around the core of nature-related comparative advantages.    

        Let us assume that real exchange rates play the balance-of-payments adjustment, whatever be the actual impact of higher energy costs on comparative advantages. Then one can realize that the end of the age of very low-cost sources of electricity will extract a toll on Brazil’s economic growth:

(i) Ceteris paribus, the shift toward sources of domestic energy supply that require more capital and labor per MW imply a negative shock on the country’s Total Factor Productivity (TFP); and

(ii) Ceteris paribus, a shift toward imports of gas or oil would have a similar effect, using old-time supplies as a benchmark, particularly if the balance-of-payments adjustment to higher imports requires a devaluation of domestic TFP in foreign-currency terms.

         That toll may obviously be more than compensated by other positive growth factors, including technological progress in the same energy sector. However, the less able to tap remaining potential hydro opportunities – in the appropriate environmental manner – be Brazil, the higher will be that toll. It is worth reminding that current decisions on energy-dependent investments, whose maturity will span beyond 2010 and thus necessarily take future energy scenarios into account, are already frontloading the latter as a current macroeconomic performance-shaping factor.

       Present energy and environment policies also matter for current economic growth. In this blog, we should always keep our sight beyond – and underneath – the evolution of short-run macroeconomic variables.