Katia Rocha

Thermopower generation investment in Brazil--economic conditions

Abstract One of the main questions in electricity market deregulation is the aptitude of private capital for investments in power generation. This is especially important in Brazil, whose load has a strong growth trend (≈6% per year). Thermopower is an attractive alternative for expanding generation, as it is complementary in many aspects to hydropower, which supplies most Brazils power at a very low price most of the time, but makes the system vulnerable to seasonal water variations. This paper studies the competitiveness of thermopower generation in Brazil under current regulations; assesses under the real options theory approach the conditions for investments in thermopower generation, and finally presents and discusses a hydropower generation schedule model.

The Term Structure of Sovereign Spreads in Emerging Markets: A Calibration Approach for Structural Models

This paper proposes a simple structural model to estimate the term structure of sovereign spreads and the implied default probability of a selected group of emerging countries, which accounts for more than 50% of the J. P. Morgan EMBIG index. The real exchange rate dynamics, modeled as a pure diffusion process, are assumed to trigger default event. By relaxing the hypothesis of market completeness, the calibrated model generates sovereign spread curves consistent with market data, giving average deviations below 30 (Mexico, Russia and Turkey) or 60 (Brazil) basis points over time. We show the robustness of the model and argue that the criticism of structural models for underestimating the magnitude of market spreads should be reconsidered. The results suggest that the market tends to overprice the spreads for Brazil, whereas for Mexico, Russia and Turkey the model reproduces the market behavior. Este trabalho apresenta um modelo estrutural para estimar a estrutura a termo do spread soberano e a probabilidade implicita de default em um grupo de paises emergentes que compoe mais do que 50% do indice EMBIG do JP Morgan. A dinâmica da taxa real de câmbio evolui de acordo com um processo de difusao simples, e representa a variavel indicativa do evento de default. Relaxando-se a hipotese de mercado completo, o modelo calibrado reproduz a estrutura a termo dos spreads de forma consistente com a observada no mercado, gerando desvios absolutos menores que 30 (Mexico, Russia e Turquia) ou 60 (Brasil) pontos-base. O modelo proposto e robusto e, portanto, a critica a respeito dos modelos estruturais subestimando a magnitude dos spreads deve ser reconsiderada. Nossos resultados revelam que o mercado esta sobreestimando os spreads para o Brasil, enquanto para Mexico, Turquia e Russia o modelo reproduz o comportamento do mercado.

The Term Structure of Sovereign Spreads in Emerging Markets

A simple structural model to estimate the term structure of sovereign spreads and the implied default probability of a selected group of emerging countries can account for more than 50% of the J.P. Morgan EMBIG index. The real exchange rate dynamics, modeled as a pure diffusion process, are assumed to trigger default events. Relaxing the hypothesis of market completeness, the model generates sovereign spread curves consistent with market data. The robustness of the model indicates a need to reexamine the criticism that structural models underestimate market spreads. The results suggest that the market tends to overprice the spreads for Brazil by 120 basis points on average, but it reproduces the market behavior for Mexico, Russia, and Turkey. These findings are supported by credit rating agency upgrades of the ratings for Brazil in the second half of 2004, just after the study period.

Term structure of sovereign spreads: a contingent claim model

Este trabalho propoe um modelo estrutural para estimar a estrutura a termo e a probabilidade implicita de default de paises emergentes que representam, em media, 54 % do indice EMBIG do JPMorgan no periodo de 2000–2005. A taxa de câmbio real, modelada como um processo de difusao simples, e considerada como indicativa de default. O modelo calibrado gera a estrutura a termo dos spreads consistente com dados de mercado, indicando que o mercado sistematicamente sobre-estima os spreads para o Brasil em 100 pontos base na media, enquanto para Mexico, Russia e Turquia reproduz o comportamento do mercado.

Credit Risk in the Pool - Implications for Private Capital Investments in Brazilian Power Generation

The new Brazilian Electric Sector Regulation issued in 2004 introduced two negotiation markets for the energy supply: the regulated pool (ACR) and the free market (ACL). Competition is enforced via energy auctions, where the winning generator has to sign long-term power purchase agreements (PPA) simultaneously with all distributors at the bidding-price. In order to estimate the appropriate credit risk spread of the pool, we implement a clustering methodology to rank and rate the distributors. The results show an average spread between 5.75% and 8.5% for the pool, which corresponds to a credit rating of B- (two levels below the Brazilian country risk). This estimation is at least 208 basis points higher than the credit rating Ba1/BB+ assigned to the distributors by the National Electric Energy Agency (Aneel) in the periodic tariff revisions, which implies an underpricing of risk. On average, distributors with higher risk/spread are located in the South-Southeast region, compared to the low risk/spread ones that are concentrated in the North-Northeast. This behavior is kept during the period of study (2001-2004). Due to the risk sharing mechanism of the pool, a generator that used to trade just in the North-Northeast market is now bearing a risk 480 basis points higher. We estimate the opportunity cost of capital in real terms in the range 13%-16% to account for the credit risk of the pool. Essential to determine the bidding price at the auctions, this estimation is considerably higher than the 11.26% opportunity cost estimated by Aneel in the 2003-2004 periodic revisions. Since efficient markets should correctly price the risks and returns, the pools credit risk has to be taken into consideration, especially for compensating private capital investments in Brazilian power generation.

Determinants of the Brazilian Sovereign Risk: A Two-Factor Structural Model

This paper develops a two-factor credit risk model based on the contingent claim structural models for pricing the Brazilian sovereign risk implicit in the Brazilian C-Bond. The underlying variable that captures the default probability is a measure of the macroeconomic fundamentals associated with the assumption that agents expectation changes smoothly over time. This approach incorporates both changes into the pricing model: the macroeconomic fundamentals and the creditors expectations. We identify the following applications for the proposed model: i) how macroeconomic policies affect the evolution of the credit spread; ii) estimation of the default probability and the associated risk premium; and iii) the credit spread forecast. We reach the following conclusions: i) the proposed model has a forecast ability higher than the random walk; ii) both quantities - fundamentals and expectations - are important to explain the credit spread evolution; iii) the risk premium is not constant over time; iv) at the margin, fundamentals are more important than the risk premium to forecast thespread; and v) the recent increase in the risk premium for Brazil indicates there is significant room to reduce the credit spread just as a consequence of an improvement over expectations.

A política fiscal e as taxas de juros nos países emergentes

This paper analyzes the role of fiscal policy sustainability on the determinants of domestic interest rate of 18 emerging market countries, in the period 1996-2008. This issue deserves attention since countries in the sample present a great level of heterogeneity relating to inflation target and exchange rate regime, political system, foreign reserves and saving rates; and such differences may also affect interest rate. Despite the heterogeneity between countries, the result shows that is not possible to reject the hypothesis that fiscal policy sustainability decreases domestic interest rate in emerging markets. An increase of 1% at primary budget decreases interest rate between 50 to 100 basis point; a figure tree times higher than the one estimated by Aisen & Hauner (2008) for emerging economies. This fact illustrates the importance of fiscal policy in determining interest rate in emerging economies and therefore the limitation of central bank in the conduction of monetary policy.

Attraction of Investments in Power Generation and the Brazilian Regulatory Framework

The privatization policy for power generation system in Brazil needs that investments on new plants should be profitable. This depends on regulation rules: a) the cost of energy supplied on deficit situation; b) the normative value of energy that is the maximum value of energy for long-run contracts; and c) the uncertainty of price of natural gas (in Real). This paper connect parameters rules to the condition to private investment. To analyze this problem we admit that investors are rational, which means that it is represented by real option theory. The stochastic profit of the termogeneration plant that is the asset associated to the investment, depends on energy price on long-run contracts and on spot market price. This considering that energy dispatch is centralized due regulations rules.

Competitiveness of Thermopower Generation in the Operation and Expansion of the Brazilian Power System

The power system regulation order that the energy dispatch is centralized by a agent - ONS - that should use a stochastic dynamic optimization model that maximize the energy is accumulated in the hydro plants. This model consider only the uncertainty of water affluence. Our first issue is develop another model that do the same for a simplified system - just one hydro and one termo plant - but considering the uncertainty of de-mand and of gas price. Our second target is evaluate the relationship between system expansion cost and termo generation participation. Our third target is to measure the effect of operation flexibility of the power plants on then system operation cost. To analyse this questions was developed a dynamic stochastic control model that do the optimal dispatch.