Given the inexistence of market value for

Given the inexistence of market value for all firms in our database, we account for investment opportunity by two different types of proxies. Following the empirical literature, in terms of covariate, we control for variables that capture investment opportunity in sectorial and firms’ level. In sectorial terms, as suggested by Carpenter and Guariglia (2008), we include the actual impact of industry-level of value-added growth. We also incorporate industry-level investment growth. We believe that the dynamic of sectorial investment may influence directly the business decision to invest. At the firms’ level, we add sales growth considering the Euler equation approach of Bond and Meghir (1994) as discussed in Section 2. Although the main aim of the paper is to evaluate whether Brazilian manufacturing firms are credit constrained, we also investigate whether there are differences in the cash flow coefficient across the following groups of firms: (i) by size, evaluated by number of employees; (ii) whether they are listed or not on the stock market; and (iii) the level of firms’ exports, given by the ratio between exports and sales revenue. All these classifications allow coefficients of control variables to differ across observations in the distinct sub-samples and it rifampicin will indicate the level of financial constraints faced by firms. Thus, cash flow variable may be interacted with different dummies variables:
Data and summary statistics In order to evaluate the link between cash flow and investment, we use four different sources. SERASA is the main source as it contains balance sheet information for more than 28 thousand Brazilian firms with annual revenue over R$ 10 million (around US$ 3 million). SERASA is a company that compiles firms’ financial statements and analyzes these information to create credit scores. In summary, SERASA dataset comprehends the information of all active firms in Brazil which are able to collect information. From this dataset, we use different measures: capital, investment, cash flow and sale’s revenue. Capital is the fixed assets value of each firm and investment is its variation. Cash flow is measured by Earnings Before Interest, Tax, Depreciation and Amortization (EBITDA). The dataset comprehends all sectors of the economy from 2006 to 2010, yet focus here will be given to manufacturing sector from 2007 to 2010. Two other sources are utilized for this investigation. Number of employees of each firm from the Annual Social Information Report (Relação Anual de Informações Sociais — [RAIS]) of the Ministry of Labor is used to control for size. Additionally, information of the Foreign Trade Secretary (Secretaria de Comércio Exterior — [SECEX]) of the Ministry of Industrial Development and Foreign Trade regarding how much each firm has exported is considered. To control for potential influence of outliers, we exclude firms with observations in the one percentile of each investment and cash flow variables. Finally, if firms’ information is missing in any year from the period, we dropped it. Our final data consist of a balanced panel with 3343 firms related to all manufacturing sectors. We divide our sample in three ways: listed on the Brazilian stock market (Listed); and large firms compared to Small and Medium Enterprises (SME); and their export status. This dataset covers 27% of total revenues and 43% of total employment in the manufacturing sector. Furthermore, Terminal redundancy is important to emphasize that SERASA database consists mainly in non-listed firms, covering a huge range of size. In this way, our study appears more suited for an investigation of credit restrictions than the ones focused on firms listed on stock markets, which are larger than the average firms in a developing country, like Brazil. Additionally, it is worth to mention that there are correct incentives for firms to provide true information to SERASA, because this information may be disclosed to the banking system and contributing to potential access to credit in more favorable conditions.