Grupo de Economia da Energia

Brazilian light vehicle fleet: expansion of flex-fuel and CNG

In biofuels, ethanol on 19/04/2010 at 02:02

By Luciano Losekann and Thais Vilela (*)

The Brazilian energy matrix is characterized by a significant ethanol and natural gas participation as automotive fuel alternative to oil byproducts. Having knowledge of the evolution and characteristic of the automotive fleet is a key step to estimate the fuel consumption and guide policies related to the replacement of oil byproducts as well as the mitigation of CO2 emissions.

However, since 1986, after cancellation of Single Automotive Tax, the estimates of the national fleet circulating in Brazil, made by various institutions, have very different results. DENATRAN, National Traffic Department, publishes data on the fleet by vehicle type and states from the registration data conducted in Detran (State Department of Transit).

However, in addition to disregard a large portion of scrapping of vehicles, interstate transfers tend to drive oversized estimates.  In order to better quantify the national fleet, several estimates were made by experts (Mattos & Correia, 1996, and Alvim, 2000, for example). This article provides an updated estimate of the fleet circulating in Brazil, with the expansion of flex-fuel vehicles and vehicles converted to CNG.

To estimate the fleet, it is necessary to identify a scrapping function of automotive vehicles in Brazil. The only reference available on the age profile of the Brazilian fleet is the 1988 PNAD (National Household Sample Survey), which included a supplement with information on car ownership. According to the literature, we estimate a function that describes better the disposal of vehicles by comparing the ownership data with sale data, published by Anfavea (National Association of Vehicle Manufacturers) [1].

From this information, the light vehicles scrap rate in 1988 is calculated.  To determine a function that describes better the light vehicles scrapping process in 1988, three functions were tested, Logistic, Gompertz and Weibull, the latter being used in the international literature. The Gompertz function was chosen for corresponding better the 1988 PNAD data.

There are some controversies regarding the adoption of the 1988 scrapping function to calculate the current fleet. According to Santos (2008), the estimated parameters based on 1988 data do not reflect the current facts anymore, being necessary to adopt other methods of estimation.

Within this context, Scandiffio & Tosi (2007) assume hypotheses as to the scrap rate to project the evolution of the light vehicle fleet in 2025. According to the authors, the flex-fuel scrap rate is 3.5% in the first five years, 5% from 2007 to 2012 and 10% by 2025 [2]. Regarding the vehicle scrap rate using the compressed natural gas as fuel, it is assumed that this is 7% by 2012 and 10% by 2025. Following a similar methodology, the National Association of Industry Vehicle Components, Sindipeças, assumes an average mortality rate of 1.5% per year for the light line, considering total loss due to accidents and theft without recovery.

However, considering that international data from countries with characteristics similar to Brazil’s shows similar scrapping processes, which were not introduced in Brazil, policies that promote fleet renewal and there is no statistical support to update the scrapping function, we chose to use the scrapping function determined according to the 1988 PNAD data, with some adjustments.

To complete the task of determining the fleet by fuel profile in Brazil, we have also to include in the analysis the light vehicles process for converting to CNG. The CNG conversion data is from GasNet. By hypothesis, the scrap rate for vehicles converted to CNG is twice the vehicle not converted.

Besides the assumption on the scrapping curve, two hypotheses are needed. Since the available information does not distinguish the vehicles by type and by converted fuel, it is assumed that diesel and ethanol vehicles are not converted. Furthermore, we assume that the occurrence of CNG on light commercial vehicles is twice the occurrence on cars.

Thus, we calculate the fleet of cars and light commercial vehicles by fuel in 2009.

Table 1 – Fleet by Type and by Fuel in 2009 (thousand units)

Gasoline Flex-Fuel Diesel Ethanol NGV
Cars 11.126 8.025 1 1.190 1.085
Light Commercials 1.542 911 1.144 170 264

The estimation for total light vehicle fleet, in Brazil, in 2009, is 25.4 million units. The estimates obtained by some institutions for the light vehicle fleet are: (i) Anfavea, in 2007, 25.6 million: (ii) Sindipeças, in 2008, 26.2 million, and (iii) DENATRAN, in 2009, 36.5 million.

The strong growth of the vehicle fleet converted to CNG since 1996 (Figure 1) may be explained by governmental incentives to consumption of this fuel, for example, value of IPVA (Vehicle Tax) decreasing in Rio de Janeiro and increasing the gas pipeline in Brazil. However, due to the gas supply crisis, the vehicle fleet converted to CNG is decreasing since 2007. The continued growth of the natural gas market depends upon specific public policies with respect to the price of gas.

Figure 1 – Light Vehicle Fleet Converted to CNG (unit)

Source: Authors’ calculations based on data provided by Gasnet

Since its introduction in 2003, sales of flex-fuel vehicle increased rapidly in Brazil until 2007, when it begins to show lower rates of growth.  In 2009, sales of flex-fuel vehicles accounted for approximately 88% of total sales of light vehicles in Brazil. The growth of the fleet, on the other hand, shows no signs of stabilization. In 2009, the flex-fuel vehicle fleet already accounted for 35% of the total light vehicle fleet. The trend is that this condition will increase significantly for the next years.

Figure 2 – Evolution of the Fleet (units) and Sales (%) of Flex-Fuel Light Vehicles

Source: Authors’ calculations based on data from the IBGE and Anfavea

By considering the evolution of the light vehicle fleet by fuel type (Figure 3) you can see the fast progress of flex-fuel vehicle fleet and lower rate of growth for the gasoline light vehicle fleet.

Figure 3 – Light Vehicle Fleet Converted by Fuel (unit)

Source: Prepared from the data provided by Anfavea and IBGE

In fact, since the entry of bi-fuel light commercial vehicles and cars, in March 2003, participation of gasoline vehicles in the Brazilian automotive market is decreasing.

Table 2 – Participation in the Fleet by Fuel Brazilian Light Vehicles

Gasoline Flex-Fuel Diesel Ethanol GNV
2003 78,9% 0,2% 4,5% 13,0% 3,3%
2004 78,1% 1,7% 4,5% 11,7% 4,1%
2005 74,9% 5,3% 4,5% 10,5% 4,9%
2006 69,7% 10,9% 4,5% 9,1% 5,8%
2007 63,1% 18,6% 4,4% 7,7% 6,2%
2008 56,3% 26,8% 4,5% 6,4% 5,9%
2009 49,8% 35,1% 4,5% 5,3% 5,3%

Knowledge of fleet by fuel profile is therefore essential to estimate the consumption of these fuels and for the development of policies related to the replacement of oil byproducts as well as mitigation of greenhouse gases emissions. Brazil has a great advantage with respect to the new challenges posed by climate changes and the search for more sustainable economies. Although gasoline is still the predominant fuel, accounting for 49% of the fleet in 2009, the Brazilian fuel matrix is diverse, but as seen earlier the expansion of alternative fuels like CNG depends on pricing policies.


Alvim, Carlos Feu. (2000). Frota e consumo de veículos leves no Brasil. Economia e Energia, nº 21, julho/agosto. Available at: Acesso em: 17 de março de 2010.

Cavalcanti, Marcelo C. B. (2004). Ascenção do gás natural no mercado de combustíveis automotivos no Brasil. In: Anais do terceiro Congresso Brasileiro de Pesquisa e Desenvolvimento em Petróleo e Gás. Salvador, Bahia.

Chen, Cynthia e Niemeier, Debbie. (2005). A mass point vehicle scrappage model. Transportation research part B 401-415.

Mattos, João A Bastos e Correia, Eduardo Luiz. (1996). Uma nova estimativa da frota de veículos automotivos no Brasil. In: Anais do VII Congresso Brasileiro de Energia, p. 1267.

Ministério da Ciência e da Tecnologia. (2006). Emissões de gases de efeito estufa por fontes móveis, no setor energético. In: Primeiro inventário brasileiro de emissões antrópicas de gases de efeito estufa – relatórios de referências. Disponível em: Acesso em: 16 de março de 2010.

Santos, Antônio Carlos (2008). A influência do uso do etanol como combustível nas emissões dos gases do efeito estufa nos motores ciclo Otto. Dissertação de mestrado. São Caetano do Sul, SP.

Scandiffio, Mirna Ivonne Gaya e Furtado, André Tosi. (2007). Etanol: a riqueza adormecida. Revista eletrônica ComCiência, nº 86. Disponível em: Acesso em: 17 de março de 2010.

Sindicato Nacional da Indústria de Componentes para Veículos Automotores. (2008). Disponível em: Acesso em: 17 de março de 2010.

Wills, William. (2008). O aumento da eficiência energética nos veículos leves e suas implicações nas emissões de gases de efeito estufa – cenários brasileiros entre 2000 e 2030. Dissertação de mestrado. Programa de Planejamento Energético, COPPE/UFRJ, Rio de Janeiro

Zachariadis, Theodoros; Samaras, Zissis e Zierock, Karl-Heinz. (1995). Dynamic modeling of vehicle populations: an engineering approach for emissions calculations. Technological Forecasting and Social Change, v. 50, 135-149.

[1] Anuário estatístico 2009

[2] This methodology involves scrapping the fleet much faster than that used in this article..

(*) Associated Researcher at the Energy Economics Group

Click here to access other texts on this blog in English

About the Blog

About the Energy Economics Group

Deixe um comentário

Preencha os seus dados abaixo ou clique em um ícone para log in:

Logotipo do

Você está comentando utilizando sua conta Sair / Alterar )

Imagem do Twitter

Você está comentando utilizando sua conta Twitter. Sair / Alterar )

Foto do Facebook

Você está comentando utilizando sua conta Facebook. Sair / Alterar )

Foto do Google+

Você está comentando utilizando sua conta Google+. Sair / Alterar )

Conectando a %s