EntErprisE survEys 65118 EntErprisE notE sEriEs informality Comparing Informal Firms in Buenos Aires and Chaco 2011 Mohammad Amin T his note highlights differences between informal businesses in two regions of Argentina— Buenos Aires and Chaco. Labor productivity is much higher in Buenos Aires than Chaco. This difference is partly due to higher sales and partly due to lower employment in firms in Buenos Aires. Relative to Buenos Aires, firms in the Chaco region are more likely to use machinery and vehicles in the production process and they also face larger seasonal fluctuations in sales. Firms are more likely to report various benefits from registering—including better access to finance, markets and government subsidies, less bribes to pay and being able to issue receipts—in Chaco EntErprisE notE no. 25 than in Buenos Aires. Important firm-characteristics, such as the level of education of the owner of the firm, and the perception of corruption and crime as obstacles to business, also differ sharply across the two regions. These findings suggest a greater need to design policies towards the informal sector at the local level rather than at the national level. The survey excludes firms in the agricultural sector Introduction but contains a roughly equal mix of manufacturing Recently, the Enterprise Analysis Unit of the World and service sector firms. Firms were classified into 20 Bank conducted a survey of informal or unregistered different industries within the manufacturing and service firms in two regions of Argentina1—Buenos Aires sectors. The representation of these industries shows (Greater Buenos Aires metropolitan area) and Chaco some variation between Buenos Aires and Chaco. The (Greater Resistencia area). The Greater Buenos Aires most noticeable difference within manufacturing is that area is the commercial, industrial and financial center of compared with Chaco, Buenos Aires has a much larger Argentina, with a population of about 12.8 million per proportion of firms making handcrafts (23.6 percent vs. the 2001 Census and GDP per capita of U.S. $28,200 (in 12.2 percent) and fewer firms making furniture (6.4 percent 2006, purchasing power parity adjusted). In contrast, the vs. 24.5 percent). Within the service sector and compared Greater Resistencia area is relatively less developed with a with Chaco, Buenos Aires has more firms engaged in population of about 0.36 million (Census 2001). The city construction (7.6 percent vs. 0 percent) and less selling of Resistencia is considered to be one of the poorest cities of food or groceries (17.4 percent vs. 41.2 percent). All in Argentina. results discussed below are robust to industry fixed effects; World Bank Group The survey provides a rich source of information on that is, to differences in industry composition between the various aspects related to the structure, conduct and two cities. performance of the firms. This note uses data from this Informal firms are known to be extremely small, often survey to highlight some of the important differences run by the owner alone. Such small firms target the local between informal firms located in Buenos Aires and Chaco. market for their sales and the purchase of inputs. Hence, This is important given that the informal economy is large local factors are likely to play a key role in the structure, and little understood. Recent estimates suggest that for the conduct and performance of informal businesses. For world as a whole, between 22.5 percent and 34.5 percent example, in one study focusing on informal manufacturing of all economic activity occurs in the informal economy; firms in India, Mukherjee (2005) finds wide variations in for countries in the lowest quartile of GDP per capita, the labor productivity levels across different regions of the estimates range between 29 percent and 57 percent (La country. Regional differences in the quality of governance Porta and Shleifer 2008). and infrastructure availability are two possible explanations for these observed differences. In another study, Kim and a move from Chaco to Buenos Aires even after taking Kang (2009) find that the size of the informal activity into account the sector of the firm (manufacturing varies significantly across different regions of Russia vs. service), gender of the largest owner, whether the depending on the delay in implementing reforms and the business operates from inside or outside of household quality of governance at the regional level. premises, whether a firm uses machinery or not, age of This note extends the regional analysis of informal firms. the firm and firm-size measured by the total number It is important to note that due to lack of proper sampling of employees in a regular month. The figure clearly frames, the survey used in this note is based on a random illustrates that none of the listed variables can explain selection of informal firms. The sample is therefore not away the difference in labor productivity across the two necessarily representative of the informal economy at regions.2 the country level or even the city level. Hence, the results presented below pertain to the structure of the informal Firm-size does not show a consistent pattern firms surveyed rather than the informal economy per se. across Buenos Aires and Chaco Using two separate measures of firm-size, number of Labor productivity and firm-size employees and total sales in a regular month, the survey Firm-efficiency as measured by the ratio of total sales shows that there is no consistent difference in firm-size (in U.S. dollars) to employment in a regular month (labor across the two regions. In terms of employment, an productivity) is almost twice as high in Buenos Aires as average firm in Buenos Aires is about 90 percent of the in Chaco. That is, the median value of labor productivity size of an average firm in Chaco and this difference is for firms in Buenos Aires equals U.S. $258 versus a mere significant; like labor productivity, it does not appear to U.S. $129 for firms in Chaco. It is possible that differences be due to differences in the composition of the sample in the composition of the sample across the two regions across the two regions.3 However, looking at monthly could be responsible for the observed difference in sales as the measure of firm-size, the opposite result labor productivity. However, on closer inspection this holds. The median firm in Buenos Aires is 1.7 times larger explanation does not appear to be too strong, although than the median firm in Chaco. As with employment, the it cannot be ruled out completely. For example, figure 1 difference in total sales is large and significant and cannot shows that labor productivity (log values) increases with be explained away by differences in the composition of the sample across the two cities.4 Figure 1 Higher labor productivity in Buenos Use of machinery and other equipment is Aires cannot be explained away by more common in Chaco compared with differences in sample composition Buenos Aires 0.60 The survey provides information on the use of machinery, 0.40 use of own vehicle or other means of transportation and (logs, quartile averages of residuals) use of cell phones in the business activity. For all these three Average labor productivity 0.20 variables, equipment usage is much more common in Chaco 0.00 than Buenos Aires—figure 2 illustrates the point for the use -1.50 -1.00 -0.50 0.00 0.50 1.00 1.50 -0.20 of machinery across various subsamples within each region. For vehicle usage, 17.6 percent of the firms in Buenos Aires -0.40 compared with 30 percent in Chaco use their own vehicle -0.60 or other means of transportation. The corresponding Proportion of firms in Buenos Aires (quartile averages of residuals) figures for cell phone usage are 36.7 percent and 47 percent, respectively. These differences across the two regions are Source: Enterprise Surveys. significant and robust. Notes: (a) The figure is a partial scatter plot of the relationship between labor productivity (logs) and the dummy for Buenos Aires and it is based on quarterly averages of the residuals There are some differences between the two obtained from two separate linear regressions of the (log of) regions in how businesses finance day-to-day labor productivity for the Y-axis and the dummy for Buenos operations, use family labor and in seasonal Aires for the X-axis on the following variables: location of fluctuations in total sales business (inside vs. outside), gender of the largest owner (female vs. male), age of the firm (log), number of employees Economic development and financial development are in a regular month (log), industry (20 industries within known to be highly correlated. Hence, one might suspect that manufacturing and service sectors). (b) The positive relationship compared with Chaco, firms in Buenos Aires are less likely in the figure is statistically significant at the 5 percent level, computed using Huber-White robust standard errors. to use internal or their own funds and more likely to borrow 2 Figure 2 Use of machinery is much Figure 3 Use of family labor is more more common in Chaco than common in Chaco than Buenos Aires Buenos Aires ■ Buenos Aires ■ Chaco ■ Buenos Aires ■ Chaco Percentage of firms that use machinery 80 All firms 70 60 Manufacturing 50 40 Service 30 Firm uses machinery 20 10 No machinery used 0 All firms Manu- Service Operate Operate Female- Male- Single Multiple Female-owned facturing outside inside owned owned employee employees household household Male-owned Source: Enterprise Surveys. Single employee Multiple employees from external sources. In both of the regions, a roughly 0 10 20 30 40 50 60 70 80 90 similar percentage of firms (close to 90 percent) report internal Percentage of firms that use family labor funds as the most important source of finance. However, Source: Enterprise Surveys. compared with Buenos Aires, a much larger percentage of firms in Chaco use internal funds (97 percent vs. 75 percent) In the case of some firm-characteristics, the difference as well as external funds (18 percent vs. 5.3 percent), where between Chaco and Buenos Aires holds within certain external funds include credit from suppliers or advances categories of firms but not in the full sample. For example, from customers and loans from micro finance institutions, the percentage of firms with the largest owner having banks and moneylenders. This suggests a greater tendency secondary education or higher is much larger in Buenos among firms in Chaco than in Buenos Aires to diversify Aires than in Chaco but only among firms located outside across internal and external funds. household premises (63 vs. 48 percent) and not otherwise In the sample under study, the ratio of total sales of a firm (61 vs. 54 percent). in the busiest to the slowest month averages 3.1. However, the ratio is significantly higher in Chaco at 3.7 compared with Corruption and crime are bigger obstacles to 2.9 in Buenos Aires. The higher seasonality in Chaco holds business in Chaco than Buenos Aires within the various subsamples, such as manufacturing and service firms. The survey reports firms’ perception of whether factors Informal business is often associated with the use of family such as crime, corruption and poor access to finance are a labor. Unable to find suitable jobs in the formal sector, many severe obstacle to doing business or not. Across the two unemployed prefer to join the family informal business. As regions, there is not much difference in how firms report one might expect, the data confirm the percentage of firms on access to finance. However, a much larger percentage that use family labor is much higher in the relatively less of firms perceive corruption as a severe obstacle in Chaco developed Chaco region (figure 3). This difference between than in Buenos Aires (53 percent vs. 27 percent). The the regions persists when accounting for various firm same holds for crime (62 percent vs. 38 percent). These characteristics including the ones highlighted in figure 3. differences are large, robust to various checks and also hold within various subsamples. These findings suggest Firms in Buenos Aires and Chaco are similar in that the quality of institutions and governance may be a number of important characteristics much more important to informal firms located in the relatively less developed cities. This is also consistent with Many important firm-characteristics show roughly the fact that the benefits from registration appear to be similar levels in Buenos Aires and Chaco. Some examples greater in Chaco than Buenos Aires (discussed below). include the number of years of managerial experience, whether the firm has a single owner or multiple owners, More firms in Chaco perceive positive effects marital status of the largest owner (single vs. married, of registration divorced or widowed), percentage of firms that maintain accounts of the business separately from household About 56 percent of the firms in Chaco compared with expenses, percentage of firms that report internal funds a much lower 37 percent in Buenos Aires report wanting as the most common source of finance for day-to-day to register their business. The higher percentage in Chaco operations and the education level of the largest owner’s compared to Buenos Aires is robust to a number of parents. controls and holds within various subsamples such as manufacturing firms (47 percent vs. 35 percent), service 3 firms (65 percent vs. 38 percent), female-owned firms Argentina. These differences relate to various aspects of (55 percent vs. 28 percent), male-owned firms (57 percent firm-performance, firm-characteristics, business climate vs. 46 percent), firms that operate from inside (53 percent and the perceived benefits of registering. Policies aimed at vs. 29 percent) and outside (71 percent vs. 42 percent) improving income levels of those in the informal sector household premises. are likely to yield better results if they take into account The survey provides valuable information on the local factors or regional differences such as the ones potential benefits from registering as perceived by the discussed above. firms. These benefits include better access to finance, better access to raw materials, infrastructure and Notes government services, less bribes to pay and being able 1. The Enterprise and Informal Surveys implemented in Latin to issue receipts. For each of these potential benefits, America and Caribbean countries, are jointly conducted by the the percentage of firms reporting in the affirmative is World Bank and the Inter-American Development Bank for this geographic region. much higher in Chaco than Buenos Aires. This finding is 2. Regression analysis shows that the listed variables explain less than robust and holds within various subsamples. For example, 11 percent of the gap in labor productivity (log values) between taking the percentage of firms that report positively on Buenos Aires and Chaco. The significantly higher labor productivity the various benefits listed and taking the average over in Buenos Aires compared with Chaco also holds within various these percentage figures, the results show that this average subsamples, although in the subsamples (e.g., businesses operating from inside household premises) it is somewhat weak without equals 56 percent in Chaco compared with a much lower controlling for whether a firm uses machinery or not. 29 percent in Buenos Aires. This difference holds within 3. That is, the difference in employment level is not due to differences various subsamples and is robust to a number of controls. across cities in the industry composition (20 industries within One might wonder if other unreported potential manufacturing and service) of firms, belongs, age of the firm, benefits from registering may be more important to firms gender of the largest owner, whether the business operates from in Buenos Aires than Chaco. This seems unlikely, as the inside or outside household premises, whether the firm uses machinery or not, proportion of workers that are family members survey also reveals that a much larger percentage of firms of the largest owner and the education level of the largest owner. in Buenos Aires (45 percent) report no potential benefit 4. That is, the stated difference in monthly sales across the two cities from registering (as perceived by them) as the main reason is robust to the factors listed for employment in the previous for not registering, compared to only 22 percent of firms footnote. However, the difference is small and insignificant in some in Chaco. This difference is robust and holds within the subsamples such as firms that operate from outside household premises. various subsamples. In contrast to the findings on the benefits from References registering, various elements of the cost of registering do Kim, Byung-Yeon, and Youngho Kang. 2009. “The Informal not show any significant difference between Chaco and Economy and the Growth of Small Enterprises in Russia.� Buenos Aires. That is, roughly the same percentage of Economics of Transition 17(2): 351-376. firms in both regions report the following as the primary La Porta, Rafael, and Andrei Shleifer. 2008. “The Unofficial Economy reason for not registering: time, fees and paper work and Economic Development.� Tuck School of Business Working required for registering, taxes that registered businesses Paper No. 2009-57. have to pay, inspections and meeting with government Mukherjee, Dipa. 2005. “Productivity in the Informal Manufacturing Sector: Regional Patterns and Policy Issues,� in Industrialization, officials that follow registration and bribes that registered Economic Reforms and Regional Development: Essays in Honor of Professor businesses need to pay. A.K. Mathur, edited by Sukhadeo Thorat, Jaya Prakash Pradhan and There are significant differences between informal Vinoj Abraham. Shipra Publications. firms located in the Buenos Aires and Chaco regions of The Enterprise Note Series presents short research reports to encourage the exchange of ideas on business environment issues. The notes present evidence on the relationship between government policies and the ability of businesses to create wealth. The notes carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this note are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. 4