Introduced at the end of the 1970s to study the impacts of structural changes on electricity consumption by industry, index decomposition analysis techniques have been extended to various other areas to help in the formulation of energy policies, notably in developed countries. However, few authors have applied these techniques to study the evolution of energy consumption in developing countries. In Brazil, the few available studies have focused only on the industrial sector. In this article, we apply the decomposition technique called the logarithmic mean Divisia index (LMDI) to electricity consumption of the Brazilian residential sector, to explain its evolution in terms of the activity, structure and intensity affects, over the period from 1980 to 2007. The technique is sufficiently robust and flexible to perform this analysis, by disaggregating residential consumers by consumption classes and regions of the country. Among the main results is measurement of the impact of government programs for income transfer and universal service on variations in residential consumption, typical of developing countries.
Income inequality among countries has long attracted interest in the academic literature (Barro, 2000; Dollar and Kraay, 2002; Deininger and Squire, 1996; De Ferranti et al., 2004; Lopez and Servén, 2006). Likewise, income inequalities within the same country have also attracted a great deal of attention. In the Brazilian case, among the many socioeconomic studies are those of Barros et al., 2001, Barros. et al., 2007, Ferreira et al. (2006), Hoffmann, 2005, Hoffmann, 2007, Langoni (2005), Levy and Villela, 2006, Peliano (2006), Schwartzman (2006) and Soares (2006).
In general, studies of income inequalities involve two main themes, as explained by Rocha (1998). The first involves inequality among people, associated with levels of life quality and welfare of the population, normally measured by the income variable. In this respect, the average labor income per person has been used as an indicator, according to which Brazil can be considered somewhere in the middle in relation to other countries. However, investigation of income distribution reveals huge disparities. According to the Brazilian Institute of Geography and Statistics, the official census bureau (IBGE, 2008a), in 2007 the average yearly income from all labor sources1 of persons2 was around US$ [2000] 3245.00.3Of these people, 58% belonged to the lowest income classes (those receiving up to twice the minimum monthly wage),4 and they earned only 28% of all labor income, a share very near that of the richest 3% of the working population (those receiving more than ten times the minimum wage), earners of 25% of total income.
The second theme is regional inequalities, which are still severe in Brazil despite longstanding efforts by the government to reduce them.5 According to the IBGE (2008a), again based on total labor income, in 2007 the Southeast region, where 43% of the occupied people live, was responsible for 52% of this income, while the Northeast was responsible for only 15% of income despite having 26% of the occupied population. The average yearly labor earnings in the Southeast and Northeast for 2007 (at 2000 buying power) were US$ 3970.00 and US$ 1871.00, respectively.6
Analogously to income, household electricity consumption is one of the indicators used to measure the level of comfort and welfare of people and, in this sense, inequalities in consumption by households belonging to different income classes and region has motivated some studies about this issue. In the Brazilian case, among the main studies are those of Achão (2003), Lenzen and Schaeffer (2004), Cohen et al. (2005) and Wachsmann et al. (2009).
The data disclosed in MME/EPE (2008)7 and EPE (2008a) show a significant expansion in residential power consumption over the 1980–2007 period, from 23.3 to 90.9 TWh, an annual growth rate of 5.2%. Average residential consumption (i.e., consumption per domicile) increased from 132 to 145 kWh/domicile/month in the same period, according to the EPE (2008b).
This growth, however, was not uniform, either at the regional level (aggregate) or among households belonging to different income classes. According to the EPE (2008a), the Northeast region, for example, which had 25% of residential consumers in 2007, was responsible for only 16% of residential electricity consumption, while the Southeast, containing 48% of residential consumers, was responsible for 54% of total residential consumption that same year.
The disparities in access to and use of electricity become even more glaring when examining households belonging to different income classes. Information contained in EPE (2008b) shows that households classified as low-income consumers8 represented nearly 34% of all residential consumers in 2007 and consumed 15% of residential electricity that year. In contrast, households not classified as low-income (called “conventional consumers”) represented 66% of residential consumers in 2006 and were responsible for 85% of residential LV consumption.
The aim of this article is to examine the evolution of residential energy consumption in Brazil based on explanatory effects, combining the two themes mentioned. We initially investigate inequality among regions regarding access to electricity, more specifically consumption of electricity by households, for the period from 1980 to 2007. We then extend the analysis to the disparities between households belonging to different income classes, more specifically between those classified as low-income and conventional consumers. The analysis covers the period from 1997 to 2007, since the criteria for classification of households as being low-income were only unified and regulated as of 1995.9
To analyze the evolution of household energy consumption we employ the logarithmic mean Divisia index I, or LMDI I method (Ang, 2004), a decomposition technique originally conceived in the late 1970s to study the impacts of structural changes on energy use by industry and since then widely used to analyze energy consumption and CO2 emissions in OECD countries, but little used in non-OECD countries. In Brazil, the use of such techniques is limited to the industrial sector (Machado and Schaeffer, 2006; Worrell et al., 1997).
This article is divided into five sections including this introduction. Section 2 describes the evolution of electricity consumption in the Brazilian residential sector through the analysis of some aggregate indicators. Section 3 presents the data and decomposition technique utilized. The results are reported and discussed in Section 4. Finally, Section 5 presents the main conclusions.
If you are interested in this research article, you can download the full length article through the download page using the link below, both PDF file and Word file are provided.
No comments:
Post a Comment