Model Results and Simulations Performed

The normalized co-integration vector for the series of sales is shown in Table 1. As highlighted by Margarido (2004) in the VEC model all variables are on the same side of the system, that is, there would be independent and dependent variables, normalizing one of the variables (LSALES_SA), the other would be deemed independent and therefore, their signals would be reversed. Thus, the interpretation of the signs of the coefficients is made of inverted manner.

Co-integration vector normalized to LSALES_SA

Source: From the results obtained with the software EViews 5.0.

Since the variables are in logarithmic form on both sides of the equation, it is possible to interpret them as elasticity in relation to sales (vehicles demand).

Therefore, one can make the following findings:

a) The increase of 1% in prices causes a reduction of approximately 2.09% in vehicle sales;

b) The increase of 1% in credit leads to an increase of approximately 0.49% in vehicle sales;

c) The increase of 1% in income leads to an increase of approximately 3.66% in vehicle sales.

Thus, it is important to note that sales are very sensitive to changes in prices and average income of workers employed and are less sensitive to changes in credit facilities in the analyzed period, that is, between June 2002 and December 2012. The income is the variable that most impacted sales throughout the period and this is probably due to the appreciation of the minimum wage policy promoted in Brazil in recent years.

The results seem to confirm the ones obtained by DIMAC (2009) for low elasticity of sales on credit. The difference of this work in relation to DIMAC (2009) is that the latter used Ordinary Least Squares (OLS) and a sample from June 2003 to June 2009, while the first used the Vector Error Correction Model (VECM).

Concerning the study by Alvarenga et al. (2010a, 2010b), income also showed greater impact in the long run. However there are some differences between this study and the ones performed by those authors regarding the data and methodology. The first is that the data here used cover a longer period (from June 2002 to December 2012). The second is in relation to income since they used GDP as a proxy for income and in this work was used the average nominal income of employed persons. The third is that here sales are inelastic to credit and for Alvarenga et al. (2010a, 2010b) sales are elastic to credit.

Vehicle sales (cars and light commercials) were simulated for two periods. The first period covers the first reduction of the IPI on national vehicles, between January 2009 and March 2010. The second period includes the second reduction of IPI on domestically produced vehicles, from June 2012 to December 2012. The sales performance was simulated for three hypothetical scenarios.

In the first scenario sales behavior was simulated⁵ given an increase of 5.85%⁶ in seasonally adjusted prices and demand. The impact of IPI reduction was first calculated for the period between January 2009 and March 2010.

While in months as November 2009, December 2009, January 2010 and February 2010, sales without IPI reduction approached the sales with IPI reduction in other months as March 2009, June 2009, September 2009 and March 2010 sales with reduction of the IPI were significantly higher. The justification is that on these months consumers anticipated purchase fearing a possible increase in rates.

During this period around 771,000 vehicles were sold, because of the IPI reduction representing about 23.2% of vehicle sales during the period. This clearly shows that the IPI reduction policy was very important for the resumption of sales after the 2008 crisis, confirming the hypothesis of this study as shown in Figure 3.

If only considered the sales between January and November 2009, nearly 620,000 vehicles have been sold because of the IPI reduction, an impact of approximately 18.6%, which is slightly lower than that found by Alvarenga et al. (2010a 2010b)⁷.

However considering the period between January and June 2009, 387,000 vehicles were sold due to the lower tax rates for cars and light trucks, representing approximately 11.7% of vehicle sales during this period. Therefore, the impact of IPI reduction found in this study was lower but close to what was found by other authors especially in the period between January and November 2009.

Source: Elaborated by the authors.

Source: Elaborated by the authors.

In the second scenario, there is a 5% increase on the series of seasonally adjusted credit. First, was simulated the period between January 2009 and March 2010 and was found that between January and July 2009 the credit impact was greatly reduced. The impact of credit was more significant from in August 2009 and from November 2009 until February 2010.This may be due to the fact that after October 2009 IPI rates were increased gradually while loans still held favorable because of the IOF reduction on credit for individuals.

Source: Elaborated by the authors.

In both periods it can be seen that the impact of credit with IPI reduction is smaller than the impact of reduced IPI. Thus, we can say that for a share of consumers, if only there was a reduction in the IPI there would be a greater willingness to purchase vehicles as the consumption decisions of these individuals were mainly based on the opportunity to purchase vehicles at a lower price.

Source: Elaborated by the authors.

The 3rd scenario considers an increase of 5% in the volume of credit and seasonally adjusted prices of 5.85% and seeks to find out the impact of the credit if did not occur the IPI reduction.

Source: Elaborated by the authors.

It is noticed that, as well as for the period between January 2009 and March 2010, for the period between June and December 2012, sales obtained were lower than the ones obtained by simulation of sales with reduction IPI and credit enhancement. This is different from the results obtained by Alvarenga et al. (2010a, 2010b), as in that study the impact of credit was increased without reducing the IPI. The differences may be due to the fact that this work performs simulations between January 2009 and March 2010, while the one by Alvarenga et al. (2010a, 2010b) the period used is smaller, between January 2009 and November 2009¹⁰; and also because the income variable used in this study is the average income of employed persons and not GDP.

Source: Elaborated by the authors.

Table 2 summarizes the results of the three scenarios simulated for the first period (between January 2009 and March 2010) and the second period (between June and December 2012).

Results of impacts for IPI reduction, credit with IPI reduction and credit without IPI reduction for the two periods simulated

This study analyzed the impacts of public policies on markets using as a study case two policies adopted on the national automotive market: reduction of rates on industrialized products (IPI) for nationally manufactured vehicles and reduction of tax on financial transactions (IOF) to stimulate credit to individuals wishing to purchase vehicles.

The investigation covered two periods: between January 2009 and March 2010, which was the first period of IPI reduction, when the world was facing the 2008 financial crisis, and between June and December 2012, which was the second period of IPI reduction.

An econometric model of Vector Error Correction (VEC) for sales series, prices, credit and income was used in logarithmic form and covered the period between June 2002 and December 2012. Results showed that an increase of 1% in prices causes a reduction of approximately 2.09% in vehicle sales; an increase of 1% in credit leads to an increase of approximately 0.49% in vehicle sales and that an increase of 1% in income leads to an increase of approximately 3.66% in vehicle sales.

Thus, sales are very sensitive to changes in prices and average income of workers and are less sensitive to changes in credit facilities in the analyzed period. The highest long-term impact of income on sales may be partially due to minimum wage appreciation policy promoted in Brazil in recent years but may be also due to the income effect of tax reduction policies.

Finally, sales behavior simulations were performed in order to obtain the impacts with reduced IPI only and using bank credit with and without IPI reduction.

From the results obtained from the simulations it can be said that the reduction of IPI was important for the recovery of vehicle sales in both periods. In fact, between January 2009 and March 2010, 23.2% of sales occurred because of the IPI reduction and between June and December 2012, 16.8% of sales were due to the IPI reduction.

Therefore, the IPI reduction was more important for the period immediately after the height of the financial crisis of 2008. Moreover, in 2012, unlike the period of the first IPI reduction of IPI, a significant portion of consumers already had new vehicles acquired because of the first IPI reduction. These consumers would be reluctant to purchase a new vehicle even with a further reduction of IPI, because they were still paying the installments of the financing of the first vehicle, which consumed important part of the monthly income.

In relation to credit with IPI reduction, it also showed a greater impact during the first period of IPI reduction. For this period the impact was 13.2% and in the second period the impact was 11.2%. However, considering the scenario without IPI reduction, the impact of the credit reached 9.2% in the first period and 6.7% in the second period.

Thus, this study showed that policies adopted by the Brazilian government, as the IPI reduction rate on vehicles and IOF reduction on credit extensions to individuals were effective to promote the increase of vehicle domestically produced sales.

So, it can be said that public policies cause effects on markets both on consumers and producers side and also to the government. For the government the tax reductions are in fact a given subsidy which impacts on tax revenues and traffic jam problems in urban cities to be solved. On the consumers side the tax reduction gives an incentive to buy new cars changing purchase possibilities as the tax reductions represents a lowering in prices. On the producers side the market interventions of tax reductions generate an increase on sales that lead to profit increase that may induce a change on production planning even in a bad economic condition as it may wrongly signalize to production increase. This will lead to more pressure on the government to maintain or adopt rate reductions to increase sales and keep employees leading to a vicious cycle.
References

Alvarenga, G. V. et al (2010a), ‘Indústria automobilística e políticas anticíclicas: lições da crise’, DISET/Revista Radar: Tecnologia, Produção e Comércio Exterior, Brasília, n.7, pp. 9-14.

______ (2010b), ‘Políticas anticíclicas na indústria automobilística: uma análise de cointegração dos impactos da redução do IPI sobre as vendas de veículos’, Discussion Paper IPEA, n. 1512, Rio de Janeiro, pp 1-21.

Alves, L. S. & Wilbert, M. D. (2014), ‘Redução do Imposto sobre Produto Industrializado e a venda de automóveis’,In: Congresso De Iniciação Científica Em Contabilidade, São Paulo, USP, n.7. Anais... URL: Available at http://www.congressousp. fipecafi.org , Accessed in 2015-01-23.

ANFAVEA (2012), ‘Séries temporais: dados referentes a quantidade de automóveis e comerciais leves licenciados’.URL: Available at http://www.anfavea.com.br/ tabelas.html , Accessed in 2015-01-25.

______(2014), ‘Anuário da Indústria Automobilística Brasileira’, São Paulo, 2014. URL: Available at http://www.anfavea.com.br/anuario.html , Accessed in 2015-03-08.

Banco Central do Brasil (2014) ‘Concessões consolidadas das operações de crédito com recursos livres referenciais para taxa de juros :aquisição de bens veículos - Pessoa física’. URL: Available at https://www3.bcb. gov.br /sgspub/ consultarvalores /, Accessed in 2015-01-23.

Ball, R. (1995), ‘Interest Groups, Influence and Welfare’, Economics & Politics. Volume 7, Issue 2, pages 119–146.

Becker, G.S., (1983). “A Theory of Competition Among Pressure Groups for Political

Influence,” Quart. J. Econ. 98, 371-400.

BRASIL (2012), ‘Decreto nº 7726, de 21 de maio de 2012’, Diário Oficial [da] República Federativa do Brasil, Poder Executivo, Brasília, DF,22 maio 2012. URL: Available at http://www.planalto.gov.br/ ccivil_03/_Ato2011-2014/2012/ Decreto/ D7726.htm, Accessed in 2015-02-22.

______. ‘Medida Provisória nº 451, de 15 de dezembro de 2008’, Diário Oficial [da] República Federativa do Brasil, Poder Executivo, Brasília, DF,16 dezembro 2008. URL: Available at http://www. planalto.gov.br/ccivil_03/_ ato2007 -2010/2008/mpv/451.htm , Accessed in 2015-02-22.

Bueno, R. D. L.S. (2011). Econometria de séries temporais, 2nd ed., São Paulo, Cengage Learning.

Bullock, D. (2005), ‘Should we expect Government Policy to be Efficient?’, In: American Agricultural Economics Association Annual Meeting, Providence, Rhode Island. URL: Available at http://purl.umn.edu/19444, Acessed in 2015-09-20.

DIMAC, Diretoria de Estudos Macroeconômicos (2009). Impactos da Redução do Imposto sobre Produtos Industrializados (IPI) de Automóveis, Technical Note n.15 Aug. 2009. URL: Available at http://www.ipea.gov.br/portal/index.php? option=com_content& view= article& id=624&Itemid=3 , Accessed in 2015-01-23.

Enders, W. (2004), Applied Econometric Time Series, 2nd. ed. New York: Wiley.

Engle, R.F., Granger, C.W.J (1987). ‘Co-integration and error correction: representation, estimation and testing’, Econometrica, Chicago, 55(2), 251-276.

Fundação Getúlio Vargas (2015) ‘IPA origem - produtos industriais’. URL: Available at: http://www.ipeadata.gov.br/, Accessed in 2015-01-25.

______. ‘IPC - geral – índice’. URL: Available at: http://www.ipeadata.gov.br/ , Accessed in 2015-01-25.

IBGE (2015). ‘Rendimento médio nominal do trabalho principal, efetivamente recebido no mês de referência... fev.2002-nov.2014’. URL: Available at:http://www.ibge.gov.br/ home/estatistica/ indicadores/ trabalhoerendimento/pme_nova/ , Accessed in Jan. 25, 2015.

Johansen , S. & Juselius, K. (1990). ‘Maximum Likelihood Estimation and Inferences on Cointegration with applications to the demand for money’. Oxford Bulletin of Economics and Statistics, 52, 169–210.

Lazzarini, S.C.(2011), Capitalismo de Laços: Os donos do Brasil e suas Conexões. Rio de Janeiro: Elsevier.

Lerch, S. (2004). ‘Impacts of Tax Exemptiion: An Overview’. Washington State Institute for Public Policy Full Report, 3-33.

Poguetto, P. D. (2009). ‘Produção de veículos despenca 47,1% em dezembro’. G1 CARROS, São Paulo. URL: Available at: http://g1.globo.com/Noticias/ Carros /0,,MUL 947714 -96 58, 00 -PRODUCAO +DE+VEICULOS+DESPENCA+ EM+ DEZEMBRO.html. Accessed in Jan. 25, 2015.

Margarido, M. A.,(2004) ‘Teste de Co-Integração de Johansen utilizando o SAS’, Revista Agricultura em São Paulo, São Paulo, 51(1), 87-101.

Shaffer, B.(1995). Firm level responses to government regulation: theoretical and research approaches. Journal of Management, 21(3), 495-514.

Varian, H.R.(2006), Microeconomia Uma abordagem Moderna. Rio de Janeiro: Elsevier.

Vieira, D.M. & Gomes, R.C. (2014). ‘Mudança Institucional Gradual e Transformativa: A Influência de Coalizões de Advocacia e Grupos de Interesse em Políticas Públicas.’ Revista Organizações e Sociedade, Salvador, 21(1), p. 679-694.

Xavier, W.G., Bandeira-de-Melo & R. Marcon,R. (2014). Institutional Environment and Business Groups’s Resilience in Brazil. Journal of Business Research, 67(5), 900-907.

Zimmermann, A., Egan, M., Poli, P. (2011). ‘Industry Overview: Automotive in Brazil and On Line Perspective’. Havas Digital Insight. URL: Available at http://www.havasmedia.com/documents_library/insights/hd_automotivebrazil_genericsept11.pdf, Accessed in 2015-10-25.