ENERGY AND ECONOMY IN BRAZIL 1970 - 2000
João Antonio Moreira
SECTORIAL ANALYSIS REGARDING ENERGY CONSUMPTION, AGGREGATED VALUE (AV) AND GROSS DOMESTIC PRODUCT (GDP) - from 1970 to 2000.
Notes on methodology:
a) Final energy consumption data, except for the residential sector and the non-energy uses, were taken from the National Energy Balance and the Lower Calorific Value for electricity is considered as 1kWh=860 kcal. Energy consumption of non-metallic products corresponds to cement and ceramics. Energy consumption of other services corresponds to commerce and public service. Energy consumption of the metallurgical sector corresponds to pig iron and steel, ferroalloys and other metallurgical items. The “other industries” group aggregates construction industry and the other industrial sectors that were not analyzed separately.
b) The GDP and AV data are from IBGE with the following adjustments: (i) new calculation methodology for the GDP from 1990 to 2000 and normalization of the period from 1970 to 1989, considering the real product indexes relative to 1980 and (ii) AV of the energy sector combines fractions of the mineral extraction, public services, petroleum refining and coke production.
The following table presents employment data (Financial Balances from Gazeta Mercantil), participation in the GDP of the Aggregated Value (AV) of each economic sector, participation in the total consumption of the energy consumption of each sector and the energy intensity, measured by the ratio of energy consumption to the GDP and the AV.
As can be observed, there are sectors of the economy that are large job creators, others that are large contributors to the GDP and others that are large energy users.
As a general rule, the sectors that are large job creators are also large contributors to the GDP (other services, agriculture and husbandry, other industries) and the sectors that are small job creators are also the larger energy users (metallurgy, energy sector, paper and cellulose, etc.).
Therefore, on one hand there are sectors that are intensive in energy and capital and little intensive in manpower and on the other hand, there are sectors less capital-intensive, less energy-intensive and large job creators. The participation and evolution of these sectors in the economy of a country can determine its development level.
According to data from the publication “Key World Energy Statistics” of the International Energy Agency, 70% of the world economy is concentrated on countries that are 45% dependent on external energy and with a per capita income of US$ 24,000. Countries responsible for 72% of energy exports represent only 8% of the world economy and have an average per capita income of US$ 2,800. Brazil seeks self-sufficiency in petroleum.
Based on data from the Brazilian Institute of Steel Industry and the Brazilian Aluminum Association, most of the developed countries are highly dependent on imports of steel and aluminum. Brazil is a steel and aluminum exporter and has allocated a good part of hydraulic generation for these sectors.
The following graphic shows the GDP formation in decreasing order, the participation of sectors in the GDP structure and in the final energy consumption.
Only three sectors, services (SERV), other industries (O. IND) and agriculture and husbandry (AGRI) present a larger participation in the GDP formation. The remaining sectors present a larger participation in energy consumption. Transport (TRAN) is the largest energy consumer followed by metallurgy (MET) and energy sector (ENER), the latter largely influenced by automotive alcohol production. The textile (TEX) and mining (MIN) sectors have a low representation in the GDP and energy consumption.
The food and beverage sector (F&B) is influenced by sugar production that is an energy-intensive product and little intensive in aggregated value. In an analysis to be carried out next, the effects of these products on the indexes will be calculated.
The next graphic shows in decreasing order the energy intensities by economic sector measured by the ratio of the energy consumption of the sector to its respective AV.
The transport sector is the one that presents the largest energy intensity, followed by metallurgy, non-metallic products and paper and cellulose.
The analysis of these Brazilian economic sectors in the period from 1970 to 2000 shows the decline of most indexes presented in this document when compared to those of developed countries. Due to reasons that were imposed or not by the world economy, industrial development based on export of products with low aggregated value (steel, aluminum, ferroalloys, pellets, cellulose, etc), the investments aiming at petroleum self-sufficiency and the long market reserve concerning computers have contributed to an economic growth of only 2% in the two last decades.
In the seventies, when Brazil presented good conditions concerning debt and access to external capital, the economy had an excellent performance (8% annually on the average), however, after the second large increase of world petroleum prices in 1979 the external conditioning factors acquired a larger influence on the Brazilian economy that then had a larger debt and had to cope with the increase of international interest rates. From the eighties on, different economic plans were implemented but not any of them could be sustained. No plan could face the interest of oligopolies and of big capitalists. It is no wonder that the last results of IBGE’s census indicate that Brazil continues to have a perverse income distribution.
The two following graphics show in decreasing order the indexes relative to energy consumption, AV and GDP based on the year 1970.
It can be noticed that the energy-intensive sectors such as paper and cellulose, metallurgy and mining presented an energy consumption increase larger than the GDP but an economic growth (AV) lower that the GDP, which is not favorable from the point of view of income distribution. Other sectors, that are also energy-intensive, such as the energy sector, chemistry and transport sectors presented a growth higher than that of the GDP relative to energy consumption and to AV.
The service sector, of low energy intensity, has also presented a growth higher than the GDP in what concerns energy consumption and AV.
The best performance is that of other industries group that even with an economic growth (AV) higher than that of the GDP presented a growth lower than the energy consumption. This group effectively represents the industries that should support the economic growth of a country – high level of jobs, low energy consumption and low investment.
In the United States, the other industries group represents 19% of energy consumption (not including residential consumption and non-energy uses), in Japan, 11%, in Europe, 6.8% and in Brazil, only 3.9%.
It should also be mentioned that the country presents a large potential for agricultural expansion associated with food and beverage industries, besides expansion of the service sector, notably tourism.
The next graphic points out in decreasing order the increase and decrease of energy intensities in the period from 1970 to 2000.
The textile and mining sectors were those that presented the largest increment of energy intensity, the first one due to modernization and the second one due to growth of the pellet industry that is mainly responsible for energy consumption in the sector. Since these two sectors are the less representative in energy consumption in the GDP, the energy intensity increase has little effect on the economy.
The intensity increase of the paper and cellulose sector is due to the larger increase of cellulose production relative to paper; however it should be pointed out that this sector produces about 75% of the energy it consumes which attenuates the efforts concerning the expansion of public supply.
The energy intensity index of metallurgy shows once again that it is difficult to support the sector, causing large efforts concerning public energy supply. This sector consumes 15% of the country’s energy (not including non-energy uses and residential sector), contributes only with 2.7% to the GDP and produces about 15% of the energy it consumes.
In the United States the participation of the metallurgy, paper and cellulose and mining sectors in energy consumption is only 4%, in Europe, 12%, in Japan, 18%, in Mexico, 9% and in Brazil, 20%.
The continuing loans of public money to these sectors should be reevaluated in a context where investment priority should be given to other sectors, for example the other industries one, that increase the capital productivity.
The services sector had also an increase in energy intensity due mainly to the crescent use of electricity in air conditioning and computers.
The transport, other industries, food and beverage and agriculture and husbandry sectors decreased their energy intensity.
Considering the temporal series from 1970 to 2000 of the ratio of the annual energy consumption to the GDP rates – income elasticity of the sectorial energy consumption – and calculating the standard deviation of each series, one finds the results shown in the following graphic.
Except for the services, the largest oscillation in the energy behavior relative to the GDP occurs in the energy-intensive sectors, and sectors that have a low AV and that export their products – metallurgy, sugar and paper and cellulose. In the services sector, the energy cost is insignificant relative to the Production Value, the cause of large oscillations of energy and GDP. Moderate oscillations occur in the mining, energy sector and non-metallic sectors. The lowest oscillation occurs in the agriculture and husbandry, textile, transport, chemistry and other industries sectors. The synergy among these sectors and that of food and beverage (excluding sugar) moves the economy and orients its direction.
The following graphics present for each economic sector the variations that occurred in the GDP, AV and energy consumption in the period from 1970 to 2000, taking the year 1970 as its basis.
During this entire period, metallurgy presented the AV curve behavior below that of the GDP and the energy curve behavior above that of the GDP. The largest distance between the AV and energy curves occurs in the period from 1983 and 1990, the period of largest expansion of steel and aluminum production aiming at exports.
The paper and cellulose AV curve, except for 1990 and 1991, has also a behavior below the GDP curve. The largest cellulose production relative to paper production and international prices oscillations are the reasons of the distances between the AV and energy curves, mainly from 1991 on.
The mining sector presents growing distances between the AV and energy curves with small variations in the period.
Chemistry presents low variation between the AV and energy curves and both are above the GDP curve. There are large variations from 1991 on, probably due to international price oscillations of chemical products.
In the energy sector, due to the increase of alcohol production from 1976 on, the energy curve deviates from the AV curve and becomes stable from 1988 on due to production stagnation. The largest AV increase from 1981 on is due to growth of petroleum production.
The expansion of large commercial centers, the computers boom and the expansion of public illumination should justify the increase of energy intensity in the services sector from 1987 on. It should not be discarded the possibility of AV underestimation of this sector due to informal economy.
The non-metallic industry in years of good income distribution, as during the Cruzado Plan in 1986 and the Real Plan in 1994, shows an approximation between the AV and energy curves caused by the growth of construction industry that increased the energy-intensive cement production. In years of low economic growth, the AV of the sector tends to grow below the GDP.
If in 1970 some planner would have projected that the industrial energy consumption in the next 30 years would grow according to the GDP, he would have hit the bull’s eye.
It is a mere coincidence, since in Japan in the same period the industrial energy consumption relative to the GDP had a reduction of 50%.
The AV of other industries behaves in a very similar way as the GDP but it is always accelerated, increases faster when the GDP increases and decreases faster when the GDP decreases. The drop of the energy curve in 1980 can be a consequence of governmental restrictions concerning fuel oil use that caused the more efficient use of energy in the sector as it occurred recently with electrical energy.
The food and beverage sector presents an energy consumption curve always below the AV curve even when it is impaired by the sugar production that is energy-intensive.
In agriculture and husbandry the energy and AV curves are below the GDP curve during the whole period and the energy curve has the smallest growth. In this sector, diesel and electricity present a large growth; however, firewood decreases due to urbanization and industrialization that reduces its use in handicraft production (manioc flour, fabrication of tiles and bricks, grain drying, manufacture of sugary products and rations, etc).
The transport sector presents the AV curve above that of the GDP and the energy consumption curve is below that of the GDP practically during the whole period. As in the non-metallic sector, in this sector the energy curve tends to come close to the AV curve when there is an increase of the purchase power of the population – Cruzado and Real Plans – as a result of the increase of gasoline, alcohol and aviation kerosene consumption.
a) In the last two decades, the Decennial Plans for Expansion of the Electric Sector have always projected annual economic growth above 4%, income elasticity of about 1.2. Reality has shown an economic growth of 2% annually and an income elasticity of 2.5 for electricity.
b) The larger the sensitivity of the models for energy forecasting in the long term concerning the questions presented in this document, the better the possibilities of prediction coming close to reality.
c) The macroeconomic scenarios with previsions of expansion of energy- and capital-intensive sectors that are sufficiently clear and detailed will improve the quality of forecasting and will permit the deviations from energy forecasting to have a behavior closer to the deviations from GDP forecasting.
Graphic Edition/Edição Gráfica:
Tuesday, 11 November 2008.