Policy Research Working Paper 11023 Does Competition from Informal Firms Encourage the Formal Firms to Obtain Quality Certificates? Mohammad Amin Caroline Nogueira Development Economics Global Indicators Group January 2025 Policy Research Working Paper 11023 Abstract This study investigates the impact of competition from the “legalist” model of informality, whereby the positive informal or unregistered firms on the likelihood of formal impact of informal competition on the likelihood of having manufacturing small and medium-size enterprises obtain- a quality certificate is significantly larger in countries where ing internationally recognized quality certificates. The the business environment is less favorable to operating in sample includes 16 countries in Latin America and the the formal versus informal sector due to factors such as the Caribbean, one of the regions with the highest levels of weaker rule of law and greater regulatory burden on formal informality in the world. The study uncovers a positive firms. The paper provides several layers of checks against impact, with a one standard deviation increase in informal omitted variable bias, reverse causality, and measurement competition leading to an increase in the probability of errors. The findings also show that, as expected, there is no having a quality certificate by 2.9 to 3.6 percentage points statistically significant impact of informal competition on across the different specifications. This effect is large, given the likelihood of having a quality certificate among large that only 10.4 percent of small and medium-size enterprises manufacturing firms. have a quality certificate. These findings are consistent with This paper is a product of the Global Indicators Group, Development Economics. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at mamin@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper 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. Produced by the Research Support Team Does Competition from Informal Firms Encourage the Formal Firms to Obtain Quality Certificates? By: Mohammad Amin* and Caroline Nogueira** Keywords: Informal competition, Certification, Quality certificates, Latin America JEL Codes: L15, O17, O54 * Corresponding author. Senior Economist, Enterprise Analysis Unit, DECEA, World Bank, Washington, DC. Email: mamin@worldbank.org. ORCID: https://orcid.org/0000-0002-9451- 3629 ** Economist, Enterprise Analysis Unit, DECEA, World Bank, Washington, DC. Email: cnogueira@worldbank.org The findings, interpretations, and conclusions expressed in this paper 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. We thank the Enterprise Analysis Unit of the Development Economics Global Indicators Department of the World Bank Group for making the data available. All remaining errors are our own. 1. Introduction Studies find that competition that formal or registered firms face from informal or unregistered firms (henceforth, informal competition) affects formal firms’ functioning and performance in terms of employment growth (Amin 2023), access to finance (Distinguin et al. 2016), innovation (Mendi and Costamagna 2017, Amin 2024), and total factor productivity (Wen et al. 2021). This body of work is still in its infancy, and several important issues remain unexplored. The present paper attempts to fill this gap in the literature by analyzing how informal competition impacts the proclivity to acquire internationally recognized quality certificates by formal manufacturing small and medium enterprises (SMEs) in the Latin America and Caribbean (LAC) region. The results show that this impact is positive and consistent with the “legalist” view of informality, much stronger in countries that have weaker rule of law and more burdensome business regulations. Our findings have important policy implications that are discussed. An important motivation for focusing on informal competition is the large size of the informal sector in the developing world (see Loayza 2016, ILO 2018). For instance, on average, for the 16 LAC countries included in this study, an estimated 61 percent of total employment and 34 percent of gross domestic product (GDP) originate in the informal sector.1 Thus, the informal sector can exert a large impact on the growth and development of the formal private sector through, for example, providing competition to the formal sector firms. Competition is generally considered beneficial for economic efficiency and economic development (see Nickell 1996, Van Reenen 2011, and Backus 2020). However, the literature largely ignores the potential of informal 1 The estimates are averages over 2015-2020. For Nicaragua, the employment figure is not available. The estimates are based on the World Bank’s Informal Economy Database and compiled by Elgin et al. (2021). Available at https://www.worldbank.org/en/research/brief/informal-economy-database 2 competition to contribute to the development of the private sector and the overall economy. There are only a handful of studies that analyze the issue (Section 2.1 reviews these studies). The size of the informal sector alone is not sufficient. Much depends on the interface between formal and informal firms. There are competing views here varying from no competition, as posited by the “dual economy” and the “structuralist” views, to significant competition between formal and informal firms, as argued by the “legalist” or “parasite” views (Section 2.2 reviews these models). Our findings reveal that the “legalist” view holds and that most formal manufacturing SMEs in LAC experience informal competition. Thus, it is important to analyze informal competition’s impact on the private sector in the LAC region. Another source of motivation for us is the rapid expansion in the use of internationally recognized quality certificates (henceforth, quality certificates or certification), such as the International Organization for Standards (ISO) series and the Social Accountability (SA) series over the last three decades or so. The certification process involves stipulating management practices that firms must follow and an independent verification of them. The management practices cover several areas, such as those related to product quality, working conditions, environment, and social responsibility. For example, one of the most well-known certificates, ISO 9001, stipulates management practices to ensure high product quality to meet buyer requirements and expectations (ISO 2002). The key benefit of certification is that it serves as a credible signaling device, which helps in bridging the informational asymmetry between the firm and external entities like buyers and creditors. The informational asymmetry arises because firms have superior information about their products and management practices than outsiders. This information asymmetry can lead to market failures and be a stumbling block for the growth and development of the private sector. Thus, it is not surprising that studies find that certification boosts firm 3 performance through increased sales and profit margins, encourages investment and innovation efforts, improves corporate image, and enables better access to finance (Section 2.3.2 reviews this literature). Obtaining quality certificates comes with substantial costs, which includes employee training, documenting production and management procedures, and fees for auditing by independent third parties. The decision to pursue such certifications is contingent upon a balance of the costs and benefits, influenced by several factors. While only a handful of factors, such as corruption, firm size, and innovation, have been identified in the literature (see Section 2.3.3), several other potential factors remain unexplored. We focus on one such potential factor, which is informal competition. The broader literature on competition and certification suggests several ways in which informal competition may impact a formal firm’s decision to acquire quality certificates. First, greater informal competition can put pressure on formal firms to innovate to improve product quality and develop new products. Certification goes hand in glove with quality upgrades and new products (as shown in Wen et al. 2022 and Medase and Basit 2023). One caveat here is that the opposite result is also possible. That is, more competition can discourage innovation by increasing the uncertainty associated with excessive rivalry between firms, reducing the appropriability of outcomes from investment in innovation, and decreasing the rate of return from innovation – the “Schumpeterian” effect (Schumpeter 1942). Second, greater informal competition can affect firms’ productivity, growth, and access to finance. While these effects are likely to alter the cost-benefit calculus of certification, it is difficult to predict whether they will increase or decrease the incentives for certification. Third, in the absence of complete information, external entities may apply the average product quality or standard in the industry to all firms. Thus, formal firms that 4 compete against informal firms may be viewed as having low quality. Certification can help high- quality formal firms distinguish themselves from low-quality firms. Fourth, firms often use low sales volumes and high prices to signal high product quality. The rationale here is that the low- quality firms have lower marginal and average costs of production, and therefore they prefer large volumes more than the high-quality firms. Informal competition may frustrate such signaling because it is likely to lower the profit-maximizing sales volumes of low-quality firms more than those of high-quality firms. As a result, high-quality firms in the formal sector may resort to other signaling devices, like quality certificates. It is important to note that the mechanisms discussed in the previous paragraph are motivated by the literature on competition between formal firms. Therefore, an additional layer of complexity arises because the competition that a formal firm faces from informal firms is different from the competition that it faces from other formal firms. Perhaps the most important difference is that unlike competition between formal firms (henceforth, formal competition), competition that formal firms face from informal firms occur on an uneven playing field. For example, in a celebrated study, Farrell (2004) argued that the cost advantages of avoiding taxes and regulations help informal firms compete “unfairly” with formal firms and “steal” market share from them. The impact of such “unfair” competition on certification by formal firms cannot be inferred from existing studies that focus on formal competition. Another distinction between formal and informal competition is that while formal firms may use quality certificates to “escape competition” from other formal firms, it is not clear how much quality differences matter for formal firms to escape competition from informal firms. For instance, competition between formal and informal firms may be more about delivery methods and customized service than product quality. Furthermore, there may be limited overlap in the product space between formal and informal firms. 5 We focus on LAC to keep the sample relatively homogeneous and because the region has one of the highest levels of informality in the world. As Figure 1 shows, output originating in the informal sector as a percentage of GDP is much higher in LAC and Sub-Saharan Africa (over 33 percent) than in the other regions (between 11.2 and 27.1 percent).2 Nevertheless, we show below that our main result of a significant positive relationship between informal competition and formal manufacturing SMEs’ likelihood of acquiring a quality certificate holds in other regions for which we have data. However, these results should be treated with caution as they do not pass some of the endogeneity tests (details below). Our study makes several important contributions to the small but growing literature on the effects of informal competition on formal firms and to the literature on certification. First, we extend the literature on informal competition to another important aspect of formal firms’ functioning, namely obtaining, and using quality certificates. To the best of our knowledge, this is the first study to do so. Concomitantly, we contribute to the literature on the drivers of certification by highlighting informal competition and several nuances affecting firms’ decisions on certification. Second, we explore how the impact of informal competition varies with the level of rule of law and the regulatory burden on formal firms. The analysis yields rich insights that can help design future policies for the informal sector and quality certification. Third, we pay due attention to endogeneity concerns. Our measure of informal competition is based on the average of all manufacturing SMEs in the country-year, which is relatively exogenous to any given firm. We also use the level of informal competition faced by the service firms in the country as a relatively exogenous proxy for the informal competition faced by the manufacturing SMEs. All our regressions account for time-invariant country-specific factors, which eliminates several 2 The data source is the World Bank’s Informal Economy Database mentioned in footnote 1. 6 sources of omitted variable bias. Following, among others, Rajan and Zingales (1998) and Buccirossi et al. (2013), we test for the theoretical predictions of the “legalist” model of informality, which are unlikely to hold if the results suffered from reverse causality, omitted variable bias, or measurement errors. Our results reveal that higher informal competition has a positive and statistically significant impact on the likelihood of obtaining quality certificates by the formal manufacturing SMEs. According to our baseline estimates, a one standard deviation increase in informal competition leads to a 2.9 to 3.6 percentage point increase in the probability of acquiring quality certificates by the formal manufacturing SMEs. Note that only 10.4 percent of the firms have quality certificates. This positive impact of informal competition is found to be much larger in countries with weaker rule of law and greater regulatory burden on formal firms. 2. Conceptual framework and literature review 2.1 Impact of informal competition A few studies have analyzed the impact of informal competition on the performance and functioning of formal firms. These studies find that higher informal competition leads to lower job creation by the formal firms (Amin 2023), a higher chance of being credit constrained (Distinguin et al. 2016), higher total factor productivity and more efficient energy use (Wen et al. 2021), lower output (Rozo and Winkler 2021), and higher likelihood of R&D activity and innovation (McCann and Bahl 2017, Mendi and Costamagna 2017, Amin 2024). Some studies find that informal competition has no significant effect. For example, Avenyo et al. (2021) use the World Bank Enterprise Surveys (WBES) data for five Sub-Saharan African countries collected in 2011-2014. 7 They find no significant impact of informal competition on the likelihood of R&D activity by formal firms. There are some suggested mechanisms or channels through which informal competition affects formal firms. First, informal competition leads to lower product prices and a loss of market share for the formal firms. As a result, formal firms experience lower profits and reduced growth. Second, lower profit margins due to higher informal competition can put pressure on the formal firms to cut costs, invest in new technology, innovate, and improve productivity. However, the opposite result is also possible. That is, smaller market shares and lower profit margins can lower the returns from investment, innovation, and other cost-cutting measures. Thus, it is possible that the incentive to invest, innovate, and adopt cost-cutting measures may decline with higher informal competition. Third, formal firms operating in industries or regions with high levels of informality may be perceived by buyers and banks as having low productivity and selling low-quality products (see Distinguin et al. 2016). As a result, such formal firms may have difficulty obtaining finance, with consequent effects on their functioning and performance. 2.2 Models of informality The early view of informality, the “dual economy” view, is that of a near complete segmentation of the formal and informal economies. In this model, informality is a product of a lack of adequate jobs in the formal sector resulting from imbalances between the growth rates of the population and of modern industrial employment and a mismatch between people’s skills and the structure of modern economic opportunities. Thus, informal firms are a product of “necessity” or involuntary “exclusion” from the formal economy (“necessity” or “exclusion-driven” firms). These informal firms are involved in marginal activities that are distinct from those of the formal firms. They are 8 too small and unproductive to provide any competition to the formal firms. In short, in the “dual economy” model, there is little or no impact of the informal sector on the formal sector. See, for example, La Porta and Shleifer (2008, 2014), Chen (2012), Rothenberg et al. (2016), and Dell’Anno (2022). The “structuralist” view of informality is that informal firms are subordinate economic units that serve large formal firms by supplying them with cheap goods and services (see Moser 1978, Castells and Portes 1989). The capitalist system, the process of industrialization, and globalization create competitive pressure on the formal firms to cut costs. At the same time, organized labor, high taxes, and stringent government regulations make cost-cutting difficult. Thus, formal firms are forced to rely on cheap inputs from informal firms. In the “structuralist” model, formal and informal economies are inherently linked, but there is no competition between the two in the product markets. According to the “legalist” view, informality is an outcome of a rational decision by entrepreneurs to “exit” the formal sector and seek profitable opportunities in the informal sector (De Soto 1989, Maloney 2004, La Porta and Shleifer 2014, Chen 2012, Dell’Anno 2022). High taxes, high corruption, and stringent business regulations make it unprofitable for some firms to operate in the formal sector. Hence, these firms “exit” into the informal sector. Also, some firms may “exit” because of the desire to gain an “unfair” advantage by not having to pay taxes and comply with costly business regulations. Most “exit-driven” or “opportunity” firms operate on the fringes of the formal-informal divide and are large and productive enough to compete with the formal firms. We caution that not all informal firms enjoy the “unfair” advantage, and that the extent of such an advantage may vary substantially across different contexts. Further, the “unfair” 9 advantage of being informal may be countered, at least partly, by the benefits of being formal, including better access to finance, infrastructure, and markets. 2.3 Quality certificates 2.3.1 Why do firms acquire quality certificates? With the emergence of quality certificates, a substantial body of research has developed examining why firms acquire quality certificates. The research essentially uncovers two main types of economic benefits for firms. First, firms often experience substantial operational enhancements internally because of the certification process. This process forces firms to examine, evaluate, improve, and systematize their production and distribution processes, which can lead to cost reductions and heightened efficiency, as demonstrated in several studies such as Romano (2000), Sun (2000), Bayati and Taghavi (2007), Sampaio et al. (2009), and Otrachshenko et al. (2023). A key question that the literature does not answer is why firms need costly certification to improve their production and distribution processes if they can do so even without certification.3 Second, as mentioned above, the main rationale for obtaining quality certificates that are recognized on an international level is that they are vital for signaling purposes: firms may strategically obtain quality certificates to indicate to external entities that their products or services adhere to specific standards. Previous research has suggested that quality certificates can serve as globally decentralized institutions that are particularly influential in developing countries, where institutional voids are more common and quality concerns are more pressing, as highlighted by Montiel et al. (2012), Goedhuys and Sleuwaegen (2013), and Paunov (2016). In such 3 Meemken et al. (2017) suggest that firms may use certification as a commitment-enhancing device when they lack the self-control to implement changes in their production and distribution structures. However, the authors do not provide any evidence to support their claim. 10 environments, quality certifications are viewed as an economical method for signaling a firm’s unobservable quality performance (Anderson et al., 1999; Clougherty and Grajek, 2008; Potoski and Prakash, 2009; Yang et al., 2023). 2.3.2 Impact of certification on firm performance Several studies find positive effects of certification on firm performance. That is, firms that acquire quality certification experience higher total factor productivity (Goedhuys and Sleuwaegen 2013), higher sales and lower input costs (Starke et al. 2012), a higher growth rate of sales (Goedhuys and Sleuwaegen 2013), better operating performance or profit margins (Naveh and Marcus 2005), more innovation (Ullah 2022, Medase and Basit 2023), improved financial performance (Siougle et al. 2019), higher exports (Riillo et al. 2022, Goedhuys and Sleuwaegen 2016), higher energy efficiency (Otrachshenko et al. 2023), and better access to finance (Ullah 2020, Minard 2016).4 2.3.3 Drivers of quality certification Despite the benefits, the use of quality certificates is limited among firms in developing countries. Thus, it is important to understand the sorts of factors that affect firms’ decisions regarding certification. Only a few studies have explored such factors. These studies show that obtaining quality certificates is less likely when firms are financially constrained (Pietrovito 2020), employees resist change (see Ferreira et al. 2021), there is high corruption in the country (Paunov 2016), and firms are relatively small (Guasch 2007, Pekovic 2010). The use of quality certificates is more common among firms that export, those that have better learning capacity and human capital endowment, and more resources available (Darnall 2003, Montiel and Husted 2009, Fikru 4 There are several other studies that show benefits to farmers from using quality certificates to credibly signal product quality and sustainable agricultural practices. 10 2014). Evidence also shows that the incentive to obtain quality certificates is higher among firms that belong to a larger organization, innovating firms, and firms that are keener on cutting costs, improving product quality, and increasing consumer satisfaction (Pekovic 2010, Hudson and Orviska, 2013). We are not aware of any previous study on how informal competition affects firms’ likelihood of obtaining quality certificates. In fact, there is very little evidence on how product market competition in general impacts firms’ incentives to obtain quality certificates. 3. Methodology for addressing endogeneity Identifying the causal effects of informal competition on certification is challenging because of the potential problems of reverse causality, omitted variable bias, and measurement errors. The empirical sections below provide several checks against these problems. The motivation for these checks is as follows. 3.1 Cell averages In the firm-level surveys of formal firms (WBES) that we use, each firm was asked if it competes against informal or unregistered firms. Responses to the question cannot be directly used as explanatory variables in the regressions, as they may be affected by a firm’s decision to obtain quality certification (reverse causality) and/or correlated with firm characteristics that influence the certification decision (omitted variable bias). One solution suggested in the literature is to proxy firm-level responses with the average responses of firms within a given cell (country, industry) – the “cell average” method. The strategy of using cell averages to mitigate endogeneity concerns is extensively used in firm-level studies (see Ozler 2000, Dollar et al. 2006, Fisman and Svensson 2007, Aterido et al. 2011, Distinguin et al. 2016, Mendi and Costamagna 2017, Avenyo 11 et al. 2021, Amin and Soh 2021, and Amin 2023). For instance, Distinguin et al. (2016) use the WBES data to analyze how informal competition affects formal firms’ access to finance. They instrument informal competition faced by a firm by its country-level average. Applied to our study, the strategy of using cell averages assumes that formal manufacturing SMEs within a cell, a country-year pair in our case, face a similar level of informal competition. The assumption holds in our case.5 Regarding endogeneity concerns, it is highly unlikely that a single firm’s R&D activity will have any effect on the chances of formal firms in the entire country- year competing against informal firms (the cell average). Therefore, when using cell averages, reverse causality is unlikely to be a problem. Similarly, a single formal firm’s own characteristics that may affect its R&D activity are unlikely to be correlated with the cell average or the chances of all the formal firms in the country-year facing informal competition (the omitted variable bias problem). For instance, Ozler (2000) estimates the impact of the exports-to-sales ratio on a plant’s share of female workers in Türkiye. It uses the average value of the exports-to-sales ratio across all plants within an industry (cell average) instead of a plant’s own exports-to-sales ratio. Regarding the endogeneity problem, Ozler (2000, page 1241) notes that: “It would be difficult to argue that one plant would be causing the export share of output of an industry to increase unless the industry had only a couple exporting plants.” Cell averages also help to control for potential measurement errors if some formal firms choose not to respond or misreport informal competition (see Paunov 2016). 3.2 Testing for theoretical predictions 5 The correlation between informal competition at the firm level and its cell average at the country-year level (defined below) is high, equaling 0.38. 12 Our next defense against the endogeneity problem involves testing for the mechanisms or ways in which informal competition affects certification. The idea here is to theoretically identify the conditions under which, unlike other determinants of certification, informal competition has a larger or smaller impact on certification than is the case otherwise. Since the differential effect so identified is specific to informal competition, it is unlikely to hold in the data if informal competition were a mere proxy for other determinants of certification, causality ran from certification to informal competition, or there was a serious measurement error problem with our estimation. Therefore, demonstrating the differential effect empirically will increase our confidence against endogeneity concerns. This way of testing against endogeneity is extensively used in the literature (see Rajan and Zingales 1998, Duchin et al. 2010, Buccirossi et al. 2013, Vu and Glewwe 2022, Amin 2023). For example, Buccirossi et al. (2013) estimate the impact of competition policy on the total factor productivity (TFP) of firms. In reference to endogeneity checks, the authors note that (page 1327): “We search for situations where we expect competition policy to have a differential effect on productivity as compared to other omitted factors or policies. If we were to observe this kind of behavior in the data, this would enhance our confidence that the estimated nexus between the quality of a competition policy regime and TFP growth can be interpreted in a causal way.” Also, Rajan and Zingales (1998) estimate the impact of financial development on industry-level growth. They note that (page 560): “One way to make progress on causality is to focus on the details of theoretical mechanisms through which financial development affects economic growth, and document their working. Our paper is an attempt to do this. Specifically, theorists argue that financial markets and institutions help a firm overcome problems of moral hazard and adverse selection, thus reducing the firm’s cost of raising money from outsiders. So financial development should disproportionately help firms (or industries) typically dependent on external finance for their growth. Such a finding could be the “smoking gun” in the debate about causality.” 13 Our main result is that informal competition has a positive impact on formal firms’ likelihood of obtaining quality certificates. This result is consistent with the “legalist” view of informality, which asserts that informal firms can and do compete against formal firms and often “unfairly” so, because unlike formal firms, informal firms do not pay taxes nor comply with costly business regulations. Our first endogeneity check assumes that the “unfair” advantage, and therefore the intensity with which informal firms compete against formal firms, is higher when the business environment is less conducive to operating in the formal sector relative to the informal sector because of factors such as a weaker rule of law and more burdensome (for the formal firms) business regulations. A weaker rule of law reduces the probability of detection of informal firms, allowing them to expand production and compete more vigorously against formal firms. More burdensome business regulations adversely affect the performance of the formal firms but not the informal firms. As a result, informal firms compete more vigorously with formal firms. To summarize, we can expect that the positive impact of higher informal competition on formal firms’ certification decisions is magnified by weaker rule of law and more burdensome business regulations. Note that this prediction is quite stringent because not all informal firms enjoy the “unfair” advantage, and for the ones that do, the degree may vary significantly across contexts. The idea that informal firms compete more vigorously and, therefore, affect formal firms more when the business environment is less conducive to operating in the formal vs. informal sector has been discussed in the literature. It has been empirically tested and confirmed by Distinguin et al. (2016) and Amin (2023). For instance, Distinguin et al. (2016) note (page 19): “A strong rule of law, however, raises the likelihood of informal activity detection; hence informal firms have to keep their operations small or to cease their activities completely, weakening their capacity to adversely affect formal SMEs’ ventures. Moreover, complicated tax rules and a high level of corruption and bureaucracy constrain firms from entering or staying in the formal sector 14 as they both lower the benefits and increase the costs associated with formality. High tax rates may also dissuade formal firms from innovating and investing leading them to lose their competitive edge against informal firms.” Thus, based on the predictions above, we state our next endogeneity test under the following hypothesis: Hypothesis 1: Higher informal competition has a larger positive impact on formal manufacturing SMEs’ probability of obtaining quality certificates in countries with weaker rule of law and more burdensome (to formal firms) business regulations. 3.3 Falsification test The next test against endogeneity concerns is a falsification test. It is reasonable to assume that informal competition has a much smaller impact or no impact on large firms compared to SMEs. Large firms operate in different product markets than informal firms. Thus, the informal competition faced by large firms is likely to be of low intensity and therefore unlikely to impact their certification decisions much. We test for this prediction. We caution that while the falsification test helps raise our confidence against endogeneity concerns, it does not completely rule out the possibility that omitted factors could have a similar impact on large firms vs. SMEs. Hence, we should interpret the falsification test with due caution. Summarizing, the falsification test is as follows: Hypothesis 2: The positive impact of higher informal competition on formal manufacturing SMEs’ likelihood of obtaining quality certificates is nonexistent or smaller for large firms than for SMEs. 15 3.4 Informal competition in the service sector Our empirical analysis excludes the service sector (retail and other services) because information asymmetry about the quality of the product and management practices in the service sector are very different from those in the manufacturing sector. However, countries with a high level of informality in manufacturing are also likely to have a high level of informality in the service sector. For the countries in our sample, the correlation between the proportion of formal manufacturing SMEs and formal service sector SMEs in the country that compete against informal firms is 0.71. So, we use the proportion of formal service sector SMEs in a country-year (cell average) that compete against informal firms as a proxy for the informal competition faced by the formal manufacturing SMEs in the country-year. The advantage of using this proxy is that the informal competition faced by service sector SMEs is relatively exogenous to manufacturing firms. Regarding reverse causality, it is unlikely or inconceivable that the certification decision of a manufacturing SME has any impact on the level of informal competition faced by service sector SMEs. Regarding the omitted variable bias problem, it is unlikely that the unobserved or omitted characteristics of a manufacturing SME that affect its certification decision are correlated with the informal competition faced by service sector SMEs. 4. Data, main variables, and estimation methodology 4.1 Data sources Our main data source is the World Bank Enterprise Surveys, or WBES, which are firm-level surveys of private firms conducted by the World Bank. The surveys are nationally representative of the private sector excluding agriculture, extractive industries, and some service sectors like finance, education, and health care. The WBES covers only formal or registered firms with five or 16 more employees. The sampling methodology used is stratified random sampling, with stratification done by firm size (small, medium, and large, defined below), industry, and location within a country.6 Our baseline sample includes all small and medium enterprises (SMEs) in the manufacturing sector in LAC. Following the definition used by WBES for stratification purposes, SMEs are all firms with fewer than 100 full-time permanent workers. The WBES also surveys large firms, defined as those with 100 or more workers. These large firms are excluded from the baseline sample because they operate in very different markets and sell different products than informal firms. Thus, informal competition is unlikely to have a significant effect on these firms.7 Our sample includes 4,853 manufacturing SMEs in 16 countries in LAC, for which data are available. Sampling weights are provided by WBES and used throughout. The distribution of the sample by country is provided in Table A1 in the Appendix. We complement the WBES with other data sources, including the World Development Indicators (WDI) and Worldwide Governance Indicators (WGI), both from the World Bank. 4.2 Estimation model Our baseline results are based on the following logistic equation: (ijkt = 1) 1 = (1) 1 + -(a+/31/nformal Competitoinkt+CFEk+YFEt+/FEj+Firm Controlsijkt+Country Controlskt+uijkt) 6 The firm-level WBES data that we use are publicly available at https://www.enterprisesurveys.org/en/enterprisesurveys. All the other data sources that we use are also in the public domain. 7 Results for the sample of large firms obtained from WBES are provided separately. 17 where the subscript i denotes the firm, j the industry (at the 2-digit ISIC Rev. 4 level), k the country where the firm operates, and t the year the WBES was administered in the country. p(.) is the probability of success conditional on the explanatory variables. Y is a dummy variable equal to 1 if the firm has an internationally recognized quality certificate and 0 otherwise; Informal Competition is a measure of the level of informal competition faced by the formal firm; CFE is a set of dummy variables for the country where the firm operates (Country fixed effects); YFE is a set of dummy variables for the year the ES was administered in the country (Year fixed effects), and IFE is a set of dummy variables for the firm’s industry (Industry fixed effects). Firm controls account for various firm characteristics. Country controls include controls for country characteristics, and u is the error term. The parameters in equation (1) are estimated by applying maximum likelihood estimation to the following transformed log odds equation: (ijkt) ቆ ቇ 1 − (ijkt) = + 1 kt + k + t + j + ijkt + kt + ijkt (2) As discussed above, we go beyond and explore how the relationship between informal competition and the likelihood of certification depends on country characteristics (rule of law and business regulations). This heterogeneity analysis is conducted by adding interaction terms to equation (2) as follows: 18 (ijkt) ቆ1 − (ijkt)ቇ = + 1 kt + 2 kt ∗ kt + 3kt + k + t + j + ijkt + ijkt + ijk (3) Unlike equation (2), equation (3) includes the interaction term between informal competition and country characteristics of interest captured by Z. It also includes as controls the interaction terms between informal competition and country and firm characteristics. These controls ensure that our main interaction term in the equation (involving informal competition and Z variables) is not spuriously picking up the differential impact of informal competition depending on factors such as income (GDP per capita) and firm size. Equation (3) is estimated using maximum likelihood estimation method. All regressions use Huber-White robust standard errors clustered on the country-year. In the online Appendix, a formal definition of all the variables is provided in Table A2 and summary statistics are provided in Table A3. 4.3 Dependent variable The dependent variable is a dummy variable equal to 1 if the firm had an internationally recognized quality certificate at the time the WBES was administered and 0 otherwise (Quality Certificate). The data source for the variable is WBES. About 10.4 percent of the firms in the baseline sample have a quality certificate. 19 4.4 Main explanatory variable Our main explanatory variable is a proxy for the competition that a formal firm faces from informal or unregistered firms. The WBES asked firms if they competed against informal firms. As discussed in detail in Section 3.1, we use the “cell average”, or the proportion of all manufacturing SMEs in a country-year cell that compete against informal firms, as our main measure of informal competition (Informal Competition). The mean value of Informal Competition equals 0.70, and the standard deviation is 0.16. As mentioned above, we also use the proportion of all SMEs in the service sector in a country-year cell as a proxy for the level of informal competition faced by the manufacturing SMEs in the country-year. 4.5 Controls We control for several potential covariates of informal competition that may also have direct effects on quality certification. This helps raise our confidence against the omitted variable bias problem. The motivation for the various controls is based on the existing literature. Table A2 in the Appendix provides a detailed definition of the controls mentioned below. 4.5.1 Baseline controls We employ several controls to minimize the possibility of omitted variable bias in our main result. In the baseline model, we begin by controlling for dummy variables for the country to which the firm belongs (Country fixed effects). This ensures that our results are unaffected by time-invariant country characteristics and that the identification of our main result comes from differences in informal competition across time (WBES surveys) for a given country rather than from across countries. Next, we control for a set of dummy variables for the year the WBES was administered 20 (Year fixed effects) and a set of dummy variables for the firm’s industry (2-digit ISIC Rev. 4). Thus, annual global shocks to certification are accounted for, as are industry specific requirements for certification. Based on the literature on the determinants of certification decisions reviewed in Section 2.3.3, we include several other controls. At the country-level, we control for overall economic development that is typically associated with greater use of certification and lower informality. The control is the log of GDP per capita (PPP adjusted and at constant 2007 Int’l $). We also control for country-size, proxied by the log of total population in the country-year to capture possible economies of scale in certification. The data source for both country-level controls is WDI (World Bank). In the robustness section, we include several other country-level controls. Firm-level controls (taken from the WBES) included in the baseline mode are as follows (see Table A2 in the Appendix for a detailed description): labor productivity defined as the log of a firm’s annual sales (in USD and deflated to 2009 prices) divided by the number of workers employed at the firm; firm-size proxied by the log of the total number of workers employed at the firm; log of the age of the firm; log of the number of years of experience the top manager of the firm has in the industry; competition between formal firms proxied by the Herfindahl index of sales concentration defined at the country-year-industry level (Herfindahl index), with higher values of the index implying less competition between the formal firms; annual sales made directly abroad as a proportion of total annual sales of the firm (Exports); the percentage of the firm that is owned by foreign entities (Foreign ownership); a dummy variable equal to 1 if the firm has a woman top manager and 0 otherwise; annual growth rate of the firm proxied by the percentage increase in the number of workers over the last 3 years (Annual employment growth); dummy variable equal to 1 if the firm introduced a new product or a vastly improved one over the last 3 years and 0 otherwise (New Product); dummy variable equal to 1 if the firm has a loan or a line of 21 credit from a financial institution and 0 otherwise (Line of Credit); quality of physical infrastructure proxied by the number of power outages experienced by the firm in a typical month during the last year; dummy variable equal to 1 if the firm uses its own website to communicate with clients and suppliers and 0 otherwise (Website); and dummy variable equal to 1 if the firms uses technology licensed from a foreign company and 0 otherwise (Foreign Technology). 4.5.2 Additional controls Additional country-level controls included in the robustness section are as follows (data source in brackets): annual growth rate of GDP (WDI, World Bank); the rule of law measure and the control of corruption measure (WGI, World Bank); and country-year level average of the number of visits or required meetings with the tax officials reported by the firms (WBES). Firm-level controls taken from WBES are: dummy variable equal to 1 if the firm is part of a larger establishment and 0 otherwise (Multi establishment firm); dummy variable equal to 1 if the firm had formal training programs for its full-time permanent workers and 0 otherwise; dummy variable equal to 1 if the firm was registered when it started operations and 0 otherwise; severity level on a 0-4 scale of tax rates as an obstacle for the firm’s current operations; the severity level on a 0-4 scale of labor laws as an obstacle for the firm’s current operations; and the severity level on a 0-4 scale of inadequately educated workers as an obstacle for the firm’s current operations. 5. Base regression results The base regression results are provided in Table 1. All the specifications control for industry fixed effects, year fixed effects, and country fixed effects. The remaining baseline controls are added sequentially. Panel A in the table contains the estimated log odds ratios, and the marginal effects 22 are provided in Panel B. In all the specifications, there is a large and positive relationship between the likelihood of having a quality certificate (log odds and marginal effects) and informal competition. This relationship is statistically significant at less than the 1 percent level or close to it. Without any other controls (except for country, year, and industry fixed effects), a unit increase in informal competition is associated with an increase in the probability of a firm having a quality certificate by 17.4 percentage points (the marginal effect in Panel B, column 1), which is significant at the 1 percent level. This impact increases to 21.2 percentage points when we add the remaining controls to the specification (see columns 2-4 in Panel B). Alternatively, a one standard deviation increase in informal competition leads to an increase in the probability of certification by 2.9 to 3.6 percentage points across the different specifications. For the final baseline specification (column 4, Panel B), the corresponding increase is 3.6 percentage points. This is an economically large impact given that only 10.4 percent of the firms in the sample have a quality certificate. Regarding the controls, the likelihood of certification (marginal effect and log odds ratio) is significantly higher for firms that are larger, more productive in terms of sales per worker, older, have foreign owners, use technology licensed from a foreign company, and have their own website. Higher exports are also associated with an increase in the likelihood of certification, but this becomes weak and statistically insignificant when some of the other controls are included in the specification. 6. Robustness 6.1 Additional controls 23 Regression results with the additional controls are provided in Table A4 in the Appendix. As discussed above, the additional controls include, among others, firm- and country-specific characteristics like training provided by a firm or not, whether the firm was registered when it started operations, the growth rate of GDP per capita, the rule of law, and corruption. The relationship between informal competition and the likelihood of certification remains positive and significant at the 1 percent level with the additional controls. Quantitatively, it is bigger than in the baseline estimation. 6.2 Sub-samples of small and medium firms We repeat the baseline estimation, but separately for the small firms and the medium firms. The regression results are provided in tables A5 and A6 in the Appendix. These results show that our main result of a positive and statistically significant relationship between informal competition and the likelihood of certification holds among small firms and medium firms. Quantitatively, the relationship is larger among the medium firms than the small firms. 6.3 Excluding one country at a time Next, we check to see if our results are unduly affected by any single country. For this, we repeat the baseline regressions excluding one country at a time. Our main result remained intact. For example, for the final baseline specification, Figure 2 shows the marginal effect of a one standard deviation increase in informal competition when a country is excluded from the sample. The marginal effect varies, ranging between 2.8 percentage points (when Mexico is excluded) and 5.4 (when Colombia is excluded). However, all the marginal effects shown are economically large and statistically significant at the 1 percent level. 24 Our sample includes two small island countries: Barbados and Suriname. Excluding both of these countries did not change our main result qualitatively. 7. Heterogeneity results 7.1 Rule of Law Regression results for the interaction term between informal competition and the rule of law are provided in Table 2. These results show that the interaction term is negative and statistically significant at the 1 percent level in all the specifications. That is, consistent with our prediction based on the “legalist” view of informality, the impact of informal competition is much larger (more positive) when the rule of law is weaker. Note that the estimation results here rule out the possibility of the Rule of Law spuriously picking up the effects of overall economic development because the estimation controls the heterogeneous effect of informal competition with respect to GDP per capita (see column 4). To get a sense of the magnitude involved and based on a specification that includes all the baseline controls (column 3 in Table 2), a unit increase in informal competition is associated with an increase in the log odds of having a quality certificate by 4.6 points at the 25th percentile value of Rule of Law and by a much lower 3.2 points at the 75th percentile value of Rule of Law. Thus, the first endogeneity test is passed. For robustness, we use an alternative measure of the rule of law. This equals the proportion of manufacturing SMEs in the country-year that report the functioning of courts as a major or very severe obstacle as opposed to moderate, minor, or no obstacle (Courts: Major obstacle). The regression results using this measure instead of the Rule of Law are provided in Table A7 in the Appendix. As expected, the interaction term between informal competition and courts functioning as a major obstacle is positive and statistically significant, implying that informal competition has 25 a larger positive impact on the likelihood of having a quality certificate when the court’s functioning is poorer. 7.2 Regulation Regression results for the interaction term between informal competition and regulatory burden on the firms proxied by the average number of visits by tax inspectors in the country are provided in Table 3. Without any controls (except for country, year, and industry fixed effects), the interaction term is close to zero and statistically insignificant (column 1). Controlling for firm size and GDP per capita causes the interaction to increase in value to 2.97, which is significant at less than the 5 percent level (not shown). We treat these controls for firm size and GDP per capita as basic controls and condition our result for the interaction term between regulation and informal competition on them. Adding the remaining controls causes the interaction term to become even larger, and it remains significant at less than the 5 percent level (see columns 2-4). The significant positive interaction term confirms our earlier prediction that greater regulatory burden on the formal firms allows informal firms to compete more vigorously with the formal firms, which magnifies the positive impact of competition from informal firms on the likelihood of having a quality certificate by the formal firms. Thus, the endogeneity test is passed. To get a sense of the magnitude involved and based on a specification that includes all the baseline controls (column 3, Table 3), a unit increase in informal competition is associated with an increase in the log odds of having a quality certification by 6.5 points at the 25th percentile value of Inspections and by a much higher 8.8 points at the 75th percentile value of Inspections. Thus, the endogeneity test is easily passed. For robustness, we repeated the estimation after replacing Inspections with an alternative measure of regulatory burden. This equals the share of 26 manufacturing SMEs in the country-year that report obtaining licenses and permits as a major obstacle or very severe obstacle for their operations as opposed to a lesser obstacle (Licenses: Major obstacle). However, this did not change the results qualitatively (see Table A8 in the Appendix). 7.3 All interactions simultaneously The heterogenous effects of informal competition with respect to the Rule of Law and Inspections discussed above hold when considered simultaneously. See Table A9 in the Appendix for these results. 8. Other results 8.1 Excluding own firm from the cell average The cell average of informal competition used above is the proportion of all manufacturing SMEs in a country-year that compete against informal firms, including the firm in question. One may argue that there may be some feedback effect from the own firm’s certification decision to whether it faces informal competition. While it is highly unlikely that a single firm will have any significant impact on the group or country-year level average (as argued above), we check if this is indeed the case. Table A10 in the Appendix provides the results with own firm excluded from the cell average of informal competition. These results are like the baseline results discussed above. 8.2 Informal competition in the service sector as a proxy We repeat the baseline regressions using the proportion of SMEs in the service sector as an alternative measure of the level of informal competition. As discussed above, this serves as a useful 27 test against endogeneity concerns, given that the informal competition faced by service sector firms is relatively exogenous to manufacturing firms. Table 4 provides the results. They are like the baseline results, both qualitatively and quantitatively. 8.3 Falsification test Regression results for the sample of large firms are provided in Table A11 in the Appendix. As expected, the relationship between informal competition and the likelihood of having a quality certificate is weak and statistically insignificant in all the specifications considered. The finding helps bolster our confidence against endogeneity concerns. 8.4 Other regions The WBES data that we use covers other regions, including Sub-Saharan Africa, Eastern Europe and Central Asia, East Asia and the Pacific, the Middle East and North Africa, and South Asia. We checked if there is a similar relationship between informal competition and the likelihood of certification by firms in the other regions. Thus, we repeat our baseline estimation first for all the countries in regions other than LAC. Next, we focus on individual regions that have at least 10 countries to ensure adequate variation in our country-year level measure of informal competition. Two regions that satisfy this criterion are Sub-Saharan Africa and Eastern Europe and Central Asia. Regression results are provided in Table 5. For brevity and without any loss of generality, the regression results are shown without any baseline controls (except for country, year, and industry fixed effects) and with all the baseline controls. These results reveal a large, positive, and statistically significant impact of informal competition on the likelihood of obtaining quality certificates in all the regions considered. We caution that some of the other results, like the 28 heterogeneity with respect to rule of law and regulation discussed above, do not hold in any of the samples considered in Table 5. Thus, more work is needed to ascertain the causality issue in regions other than LAC. 9. Conclusion Competition in the product markets is generally regarded as beneficial to the long-term efficiency and growth of the private sector. At the same time, the competition that formal firms face from informal sector firms is typically viewed as “unfair” and harmful to formal firms. The possible benefits of informal competition to firms in the formal sector have been overlooked in the literature. There are only a handful of studies that show how informal competition can boost formal firms’ growth, efficiency, and innovation. The present paper contributes to this small but important body of work by analyzing how informal competition in LAC affects formal manufacturing SMEs’ decisions to obtain quality certificates. We find that this impact is positive and larger when the rule of law is weaker and business regulations are more stringent. Our results have several important implications for policy. First, policies to encourage informal firms to enter the formal sector are based on a careful analysis of the costs and benefits of formalization. Our results indicate that this cost-benefit calculus requires a readjustment to account for the benefit from informal competition through increased certification. Second, policies aimed at encouraging formal firms to obtain quality certificates should be tailored to and aligned with the level of informal competition. Third, our results suggest that in most cases, the optimal policy intervention to maximize the positive spillovers from the informal to the formal sector and minimize the negative spillovers depends on the quality of the business environment as reflected in the rule of law and the stringency of business regulations. Fourth, policies to encourage the use 29 of quality certificates by formal manufacturing SMEs need to move beyond providing subsidies and other such incentives to nurturing and promoting the interaction between formal and informal enterprises. Fifth, our findings challenge the conventional policy view that treats the informal sector as a homogeneous entity, revealing that the impact of informal firms on their formal counterparts varies based on the intensity of their competition. As a result, optimal policy design requires distinguishing carefully between the different types of informal firms. To conclude, a key takeaway message for policy makers is that informality is not necessarily detrimental to the growth and development of the formal private sector, and much depends on the interface between formal and informal firms and the underlying quality of the business environment. There are several issues that future research can explore. First, it would be interesting to examine how other aspects of firm performance and functioning are affected by informal competition. Some possible areas include sources of financing working capital, exports, and worker training. Second, data limitations did not allow us to track the mechanisms through which informal competition affects certification. This is an important gap in the literature that future research can address. Third, weaker rule of law and stringent business regulations were found to magnify the positive impact of informal competition on certification. The analysis above can be extended to aspects of the business environment such as, financial development, quality of physical infrastructure, and macroeconomic stability. Fourth, quality certificates are not all the same. It would be interesting to analyze how informal competition affects different types of quality certificates. Fifth, we treated the business environment in terms of the rule of law and the regulatory burden as common across all firms within a country. However, there may be differences in laws and regulations and in their enforcement across industries, sub-national regions, and firms. It would be interesting to incorporate these differences into future work. Sixth, the exercise in the present paper 30 can be extended to the service sector. Last, an indication from the results above is that the impact of certification on firm performance measures like productivity and growth may depend on the level of informal competition. This needs to be thoroughly examined. 31 References Amin, M. (2024). How Does Competition from Informal Firms Impact Research and Development by Formal Manufacturing Small and Medium Enterprises in the Developing and Emerging Economies? Forthcoming Kyklos. Amin, M. (2023). Does Competition from Informal Firms Hurt Job Creation by Formal Manufacturing SMEs in Developing and Emerging Countries? Evidence Using Firm-level Survey Data. Small Business Economics 60: 1659–1681. https://doi.org/10.1007/s11187-022-00672-z Amin, M. and Y. C. Soh (2021). Does Greater Regulatory Burden Lead to Higher Corruption: Evidence Using Firm-level Survey Data for Developing Countries. World Bank Economic Review 35(3): 812-828. https://doi.org/10.1093/wber/lhaa007 Anderson, S., J. Daly, and M. Johnson (1999). Why Firms Seek ISO 9000 Certification: Regulatory Compliance or Competitive Advantage? Production and Operations Management 8: 28–43. Aterido, R., M. Hallward-Driemeier, and C. Pages (2011). Big Constraints to Small Firms’ Growth? Business Environment and Employment Growth Across Firms. Economic Development and Cultural Change 59 (3): 609-647. Avenyo, E.K., M. Konte, and P. Mohnen (2021). Product Innovation and Informal Market Competition in Sub-Saharan Africa. Journal of Evolutionary Economics 31: 605–637. Backus, Matthew (2020). Why Is Productivity Correlated with Competition? Econometrica 88(6): 2415-2444. Bayati, A., and A. Taghavi (2007). The Impacts of Acquiring ISO 9000 Certification on the Performance of SMEs in Tehran. The TQM Magazine 19: 140–149. Buccirossi, P., L. Ciari, T. Duso, G. Spagnolo, and C. Vitale (2013). Competition Policy and Productivity Growth: An Empirical Assessment, The Review of Economics and Statistics, 95(4): 1324-1336. https://doi.org/10.1162/REST_a_00304 Castells, M., and A. Portes (1989). World Underneath: The Origins, Dynamics and Effects of the Informal Economy. In A. Portes, M. Castells, and L. Benton (Eds.), The Informal Economy: Studies in Advanced and Less Developed Countries. Johns Hopkins University Press. Chen, Martha (2012). The Informal Economy: Definitions, Theories and Policies. WIEGO Working Paper No. 1. WIEGO. Clougherty, J. and M. Grajek, M. (2008). The Impact of ISO 9000 Diffusion on Trade and FDI: A New Institutional Analysis. Journal of International Business Studies 39: 613–633. Darnall, N. (2003). Why Firms Certify to ISO 14001: An Institutional and Resource-Based View. Academy of Management Proceedings 2003(1): B1-B6. 32 De Soto, H. (1989). The Other Path. London: Harper and Row. Dell'Anno, Roberto (2022). Theories and Definitions of the Informal Economy: A survey. Journal of Economic Surveys 36(5): 610-1643. Distinguin, I., C. Rugemintwari, and R. Tacneng (2016). Can Informal Firms Hurt Registered SMEs’ Access to Credit? World Development 84(August): 18-40. Dollar, D., M. Hallward-Driemeier, and T. Mengistae (2006). Investment Climate and International Integration. World Development 34 (9): 1498-1516. Duchin, R., O. Ozbas, B. A. Sensoy (2010). Costly External Finance, Corporate Investment, and the Subprime Mortgage Credit Crisis. Journal of Financial Economics 97(3): 418–435. Elgin, C., M. A. Kose, F. Ohnsorge, and S. Yu (2021). Understanding Informality. CERP Discussion Paper 16497, Centre for Economic Policy Research, London. Farrell, D. (2004). The Hidden Dangers of the Informal Economy. McKinsey quarterly 3: 26-37. Ferreira, L.M. and C.J. Cândido (2021). Factors Influencing Firm Propensity for ISO 9001 Withdrawal: Evidence on Decertification Tendency and Antecedents. International Journal of Production Economics 233: p.108024. Fikru, M. G. (2014). Firm level determinants of international certification: Evidence from Ethiopia. World Development 64, 286–297. Fisman, R. and J. Svensson (2007). Are Corruption and Taxation Really Harmful to Growth? Firm Level Evidence. Journal of Development Economics 83 (1): 63-75. Goedhuys, M. and L. Sleuwaegen (2013). The Impact of International Standards Certification on the Performance of Firms in Less Developed Countries. World Development 47: 87-101. Goedhuys, M. and L. Sleuwaegen (2016). International Standards Certification, Institutional Voids and Exports from Developing Country Firms. International Business Review 25(6): 1344-1355. Guasch, J.L. (2007). Quality Systems and Standards for a Competitive Edge. World Bank Publications. Hudson, J. and M. Orviska (2013). Firms’ adoption of international standards: One size fits all? Journal of Policy Modeling 35 (2), 289–306. ILO (2018). Women and Men in the Informal Economy: A Statistical Picture. Third edition. Geneva: International Labor Office. International Labor Organization. 33 ISO (2002). The ISO survey of ISO 9000 and ISO 14000 certificates, Eleventh cycle – 2001. Geneva: ISO Central Secretariat. La Porta, R. and A. Shleifer (2014). Informality and Development. Journal of Economic Perspectives 28 (3): 109-126. La Porta, R., and A. Shleifer (2008). The Unofficial Economy and Economic Development. Brookings Papers on Economic Activity Fall: 275 – 352. Loayza, N. V. (2016). Informality In the Process of Development and Growth. World Economy 39(12): 1856–1916. Maloney, W. F. (2004). Informality Revisited. World Development 32 (7): 1159-1178. McCann, B. T. and M. Bahl (2017). The Influence of Competition from Informal Firms on New Product Development. Strategic Management Journal 38(7): 1518-1535. Medase, S.K. and S. A. Basit (2023). Trademark and Product Innovation: The Interactive Role of Quality Certification and Firm-level Attributes. Innovation and Development, 13(1): pp.1-41. Meemken, E.M., P.C. Veettil, and M. Qaim (2017). Toward Improving the Design of Sustainability Standards-A Gendered Analysis of Farmers’ Preferences. World Development 99: 285–298. https://doi.org/10.1016/j.worlddev.2017.05.021. Mendi, P., and R. Costamagna (2017). Managing Innovation Under Competitive Pressure from Informal Producers. Technological Forecasting and Social Change 114(January): 192 – 202. Minard, P. (2016). Signalling Through the Noise: Private Certification, Information Asymmetry and Chinese SMEs’ Access to Finance. Journal of Asian Public Policy 9(3): 243-256. Montiel, I. and B.W. Husted (2009). The Adoption of Voluntary Environmental Management Programs in Mexico: First Movers as Institutional Entrepreneurs. Journal of Business Ethics 88: 349-363. Montiel, I., B. Husted, and P. Christmann (2012). Using Private Management Standard Certification to Reduce Information Asymmetries in Corrupt Environments. Strategic Management Journal, 33: 1103–1113. Moser, C.O. (1978). Informal Sector or Petty Commodity Production: Dualism or Dependence in Urban Development? World development 6(9-10): 1041-1064. Naveh, E. and A. Marcus (2005). Achieving Competitive Advantage Through Implementing a Replicable Management Standard: Installing and Using ISO 9000. Journal of Operations Management 24(1): 1-26. 34 Nickell, L, S. (1996). Competition and Corporate Performance. Journal of Political Economy, 104: 724–46. Otrachshenko, V., C.A. Hartwell, and O. Popova (2023). Energy Efficiency, Market Competition, and Quality Certification: Lessons from Central Asia. Energy Policy 177: 113539. Ozler, S. (2000). Export Orientation and Female Share of Employment: Evidence from Turkey. World Development 28(7): 1239-1248. Paunov, C. (2016). Corruption’s Asymmetric Impacts on Firm Innovation. Journal of Development Economics 118(January): 216 – 231. Pekovic, S. (2010). The Determinants of ISO 9000 Certification: A Comparison of the Manufacturing and Service Sectors. Journal of Economic Issues 44(4): pp.895-914. Pietrovito, F. (2020). The Impact of Credit Constraints on International Quality and Environmental Certifications: Evidence from Survey Data. Journal of Risk and Financial Management 13(12): p.322. Potoski, M. and A. Prakash (2009). Information Asymmetries as Trade Barriers: ISO 9000 Increases International Commerce. Journal of Policy Analysis and Management 28: 221–238. Rajan, R. G., and L. Zingales (1998). Financial Dependence and Growth. American Economic Review 88(3): 559-586. Riillo, C.A., I. Mijatovic, and H.J. de Vries (2022). Certification to Compensate Gender Prejudice – Analysis on Impact of Management System Certification on Export. Applied Economics 54(33): 3777-3794. Romano, P. (2000). ISO 9000: What is its Impact on Performance? Quality Management Journal 7: 38–56. Rothenberg, A.D., A. Gaduh, N. E. Burger, C. Chazali, I. Tjandraningsih, R. Radikun, C. Sutera, and S. Weilant (2016). Rethinking Indonesia’s Informal Sector. World Development 80(April): 96– 113. Rozo, S., and H. J. Winkler (2021). Is Informality Good for Business? The Impacts of IDP Inflows on Formal Firms. Journal of Human Resources 56(4): 1141-1186. Sampaio, P., P. Saraiva, and A. G. Rodrigues (2009). ISO 9001 Certification Research: Questions, Answers and Approaches. International Journal of Quality and Reliability Management 26: 38– 58. Schumpeter, J. A. (1942). Capitalism, Socialism and Democracy. Harper. New York. NY. 35 Siougle, E., S. Dimelis, and C. Economidou (2019). Does ISO 9000 Certification Matter for Firm Performance? A Group Analysis of Greek Listed Companies. International Journal of Production Economics 209: 2-11. Starke, F., R.V. Eunni, N. Manoel Martins Dias Fouto, and C. Felisoni de Angelo (2012). Impact of ISO 9000 Certification on Firm Performance: Evidence from Brazil. Management Research Review 35(10): 974-997. Sun, H. (2000). Total Quality Management. ISO 9000 Certification and Performance Improvement. International Journal of Quality & Reliability Management 17: 168–179. Ullah, B. (2020). Signaling Value of Quality Certification: Financing Under Asymmetric Information. Journal of Multinational Financial Management 55: 100629. Ullah, B. (2022). The Impact of Quality Certification on SME Innovation and the Role of Institutions. Research in International Business and Finance 62: 101748. Van Reenen, J. (2011). Does Competition Raise Productivity through Improving Management Quality? International Journal of Industrial Organization 29(3): 306–316. Vu, K., and P. Glewwe (2022). Maternity Benefits Mandate and Women’s Choice of Work in Vietnam. World Development 158: 105964. Wen, H., N. Li, and Chien-Chiang Lee (2021). Energy Intensity of Manufacturing Enterprises Under Competitive Pressure from the Informal Sector: Evidence from Developing and Emerging Countries. Energy Economics 104(December): 105613. Wen, W., C. Forman, and S.L. Jarvenpaa (2022). The Effects of Technology Standards on Complementor Innovations: Evidence from the IETF. Research Policy 51(6): 104518. Yang, Z., P. Liu, and L. Luo (2023). Growing Exports Through ISO 9001 Quality Certification: Firm-level Evidence from Chinese Agri-food Sectors. Food Policy 117: 102455. 36 Figure 1: Output of the informal economy across regions 40 35.5 economy (% of official GDP, average over 34.3 35 Output originating int he informal 30 27.1 25.4 25 21.6 21.6 2015-2020) 20 15 11.2 10 5 0 Sub Saharan Latin South Asia Eastern East Asia Middle East North Africa America Europe and and Pacific and North America and Central Asia Africa Caribbean Source: Authors’ own calculations based on World Bank’s Informal Economy Database as compiled by Elgin et al. (2021). 37 Figure 2: Marginal effect of informal competition excluding one country at a time quality certifcate when informal competition 0.35 0.33 Increase in the probability of a firm having a increase by one standard deviation unit 0.30 0.27 0.25 0.24 0.24 0.25 0.23 0.22 0.22 0.22 0.22 (percentage points) 0.21 0.22 0.20 0.19 0.20 0.18 0.17 0.15 0.10 0.05 0.00 ARG BOL BRB COL CRI DOM ECU GTM HND MEX NIC PER PRY SLV SUR URY Country excluded from the sample Note: The marginal effect is obtained using logit estimation. The specification includes all the baseline controls (as shown in column 4 in table 1 below). All the marginal effects shown are statistically significant at the 1 percent level. 38 Table 1: Base regression results Dependent variable: Quality (1) (2) (3) (4) Certificate Y:1 N:0 Panel A: Log odds ratios Informal Competition (CY avg.) 2.104** 3.056*** 3.276*** 3.361*** (0.819) (0.780) (0.708) (0.737) GDP per capita (logs) -0.075 0.272 0.471 (0.924) (0.892) (1.268) Labor Productivity (logs) 0.485*** 0.438*** 0.411*** (0.117) (0.127) (0.133) Number of workers (logs) 1.182*** 1.133*** 1.007*** (0.186) (0.185) (0.185) Age of firm (logs) 0.296** 0.325** 0.350** (0.140) (0.146) (0.141) Manager experience (logs) -0.281 -0.252 -0.254 (0.176) (0.186) (0.184) Herfindahl index -0.287 -0.348 (1.026) (0.994) Exports (proportion of sales) 0.864** 0.619 (0.394) (0.391) Foreign ownership (%) 0.013*** 0.013*** (0.004) (0.004) Women top manager Y:1 N:0 -0.238 -0.279 (0.329) (0.307) Annual employment growth (%) -0.008 -0.008 (0.007) (0.006) New Product Y:1 N:0 0.062 (0.204) Line of Credit Y:1 N:0 0.001 (0.002) Number of power outages (per -0.017 month) (0.019) Foreign Technology Y:1 N:0 0.008*** (0.002) Website Y:1 N:0 0.006*** (0.002) Total population (logs) 0.687 (4.563) Industry fixed effects Yes Yes Yes Yes 39 Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 4,853 4,853 4,853 4,853 Panel B: Marginal effect Informal Competition (CY avg.) 0.174*** 0.212*** 0.221*** 0.222*** (0.067) (0.053) (0.048) (0.048) Controls (as above) Yes Yes Yes Yes Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 40 Table 2: Interaction with Rule of Law (log odds ratios) Dependent variable: Quality Certificate (1) (2) (3) (4) Y:1 N:0 Informal Competition (CY avg.) *Rule of Law -4.737*** -4.888*** -5.828*** -5.133*** (0.673) (0.812) (0.903) (1.166) Informal Competition (CY avg.) -0.063 0.668 0.255 7.055 (0.438) (0.700) (0.559) (13.491) Rule of Law 3.751*** 4.002*** 3.409*** 3.166*** (0.616) (0.940) (0.944) (0.998) GDP per capita (logs) -1.125 -0.335 0.313 (1.495) (1.364) (1.902) Labor Productivity (logs) 0.474*** 0.407*** 0.400*** (0.118) (0.133) (0.127) Number of workers (logs) 1.185*** 0.998*** 0.251 (0.188) (0.186) (0.924) Age of firm (logs) 0.290** 0.356** 0.346** (0.140) (0.140) (0.136) Manager experience (logs) -0.301* -0.260 -0.256 (0.176) (0.183) (0.185) Herfindahl index -0.287 -0.323 (1.029) (1.084) Exports (proportion of sales) 0.610 0.649 (0.400) (0.407) Foreign ownership (%) 0.014*** 0.013*** (0.004) (0.004) Women top manager Y:1 N:0 -0.267 -0.248 (0.306) (0.306) Annual employment growth (%) -0.007 -0.007 (0.006) (0.006) New Product Y:1 N:0 0.054 0.052 (0.204) (0.203) Line of Credit Y:1 N:0 0.001 0.001 (0.002) (0.002) Number of power outages (per month) -0.019 -0.018 (0.018) (0.018) Foreign Technology Y:1 N:0 0.008*** 0.008*** (0.002) (0.002) Website Y:1 N:0 0.006*** 0.006*** (0.002) (0.002) Total population (logs) 12.372*** 10.620** (4.342) (4.958) 41 Informal Competition (CY avg.)*GDP per -1.045 capita (logs) (1.449) Informal Competition (CY avg.)*Number of 1.062 workers (logs) (1.186) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 4,853 4,853 4,853 4,853 Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 42 Table 3: Interaction with Inspections (log odds ratios) Dependent variable: Quality (1) (2) (3) (4) Certificate Y:1 N:0 Informal Competition (CY avg.) 0.040 3.548** 4.044*** 3.722** *Inspections (CY avg.) (1.039) (1.534) (1.482) (1.559) Informal Competition (CY avg.) 3.096*** 3.918*** 3.961*** 8.649 (0.764) (0.981) (0.952) (11.599) Inspections (CY avg.) -0.832 -4.198*** -4.532*** -4.272*** (0.948) (1.336) (1.230) (1.311) GDP per capita (logs) 2.701** 3.389*** 3.972*** (1.203) (0.980) (1.010) Labor Productivity (logs) 0.481*** 0.414*** 0.405*** (0.118) (0.133) (0.128) Number of workers (logs) 1.194*** 1.010*** 0.093 (0.189) (0.186) (0.840) Age of firm (logs) 0.289** 0.347** 0.337** (0.140) (0.142) (0.140) Manager experience (logs) -0.298* -0.265 -0.264 (0.174) (0.185) (0.186) Herfindahl index -0.126 -0.198 (1.026) (1.123) Exports (proportion of sales) 0.658* 0.701* (0.397) (0.408) Foreign ownership (%) 0.012*** 0.012*** (0.004) (0.004) Women top manager Y:1 N:0 -0.276 -0.250 (0.312) (0.310) Annual employment growth (%) -0.007 -0.007 (0.006) (0.006) New Product Y:1 N:0 0.125 0.121 (0.209) (0.207) Line of Credit Y:1 N:0 0.001 0.001 (0.002) (0.002) Number of power outages (per month) -0.017 -0.016 (0.019) (0.019) Foreign Technology Y:1 N:0 0.008*** 0.008*** (0.002) (0.002) Website Y:1 N:0 0.006*** 0.006*** (0.002) (0.002) Total population (logs) 1.096 1.292 (2.647) (2.367) Informal Competition (CY -0.942 avg.)*GDP per capita (logs) 43 (1.121) Informal Competition (CY 1.308 avg.)*Number of workers (logs) (1.075) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 4,853 4,853 4,853 4,853 Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 44 Table 4: Using informal competition in the service sector as a proxy Dependent variable: Quality Certificate (1) (2) (3) (4) Y:1 N:0 Panel A: Log odds ratios Informal Competition (service sectors, 2.515** 3.794** 3.574** 4.996*** CY avg.) (1.273) (1.707) (1.801) (1.873) GDP per capita (logs) 0.994 1.181 0.235 (1.227) (1.240) (1.164) Labor Productivity (logs) 0.482*** 0.432*** 0.407*** (0.121) (0.131) (0.135) Number of workers (logs) 1.185*** 1.139*** 1.018*** (0.188) (0.189) (0.188) Age of firm (logs) 0.269* 0.291** 0.325** (0.139) (0.146) (0.141) Manager experience (logs) -0.269 -0.236 -0.235 (0.177) (0.185) (0.186) Herfindahl index 3.794** -0.394 -0.258 (0.984) (0.970) Exports (proportion of sales) 0.904** 0.644 (0.395) (0.395) Foreign ownership (%) 0.012*** 0.012*** (0.004) (0.004) Women top manager Y:1 N:0 -0.243 -0.277 (0.325) (0.301) Annual employment growth (%) -0.008 -0.008 (0.006) (0.006) New Product Y:1 N:0 0.098 (0.202) Line of Credit Y:1 N:0 0.001 (0.002) Number of power outages (per month) -0.014 (0.019) Foreign Technology Y:1 N:0 0.008*** (0.002) Website Y:1 N:0 0.006*** (0.002) Total population (logs) 10.715* (6.163) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 4,853 4,853 4,853 4,853 45 Panel B: Marginal effect Informal Competition (CY avg.) 0.209** 0.264** 0.243** 0.332*** (0.105) (0.118) (0.121) (0.121) Controls (as above) Yes Yes Yes Yes Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 46 Table 5: Regions other than Latin America All regions excluding Sub-Saharan Africa Eastern Europe and Latin America Central Asia Dependent variable: Quality (1) (2) (3) (4) (5) (6) Certificate Y:1 N:0 Panel A: Log odds ratios Informal Competition (CY avg.) 2.074*** 1.542*** 4.120*** 6.739*** 2.206*** 1.992*** (0.550) (0.565) (1.269) (2.396) (0.478) (0.608) GDP per capita (logs) -1.038 4.139** -0.646 (0.940) (2.030) (1.013) Labor Productivity (logs) 0.190*** 0.100 0.238*** (0.037) (0.089) (0.052) Number of workers (logs) 0.659*** 0.377** 0.663*** (0.066) (0.169) (0.087) Age of firm (logs) 0.166** 0.445** 0.079 (0.079) (0.205) (0.095) Manager experience (logs) -0.143** -0.185 -0.067 (0.069) (0.155) (0.081) Herfindahl index 0.210 -0.369 0.124 (0.496) (0.822) (0.865) Exports (proportion of sales) 0.768*** 1.279** 0.669*** (0.165) (0.525) (0.206) Foreign ownership (%) 0.007*** 0.005 0.007*** (0.002) (0.005) (0.002) Women top manager Y:1 N:0 0.073 -0.149 0.083 (0.105) (0.297) (0.133) Annual employment growth (%) -0.004 -0.001 -0.005 (0.003) (0.007) (0.004) New Product Y:1 N:0 0.161* 0.122 0.171 (0.090) (0.245) (0.114) Line of Credit Y:1 N:0 0.002*** -0.001 0.003*** (0.001) (0.002) (0.001) Number of power outages (per -0.001 -0.012 0.006 month) (0.002) (0.009) (0.019) Foreign Technology Y:1 N:0 0.005*** 0.006** 0.005*** (0.001) (0.003) (0.001) Website Y:1 N:0 0.009*** 0.011*** 0.007*** (0.001) (0.003) (0.001) Total population (logs) 1.264 0.542 3.500* (1.352) (8.915) (1.800) Industry fixed effects Yes Yes Yes Yes Yes Yes 47 Year fixed effects Yes Yes Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Yes Yes Number of observations 31,090 31,090 2,927 2,927 10,355 10,355 Panel B: Marginal effect Informal Competition (CY 0.260*** 0.166*** 0.307*** 0.433*** 0.369*** 0.287*** avg.) (0.069) (0.062) (0.096) (0.004) (0.079) (0.089) Controls (as above) Yes Yes Yes Yes Yes Yes Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 48 Appendix A Table A1: Sample size by country Observations Country Year (Number of firms) Argentina 2009 418 Argentina 2016 372 Barbados 2009 47 Barbados 2022 44 Bolivia 2009 32 Bolivia 2016 59 Colombia 2016 337 Colombia 2022 196 Costa Rica 2009 164 Costa Rica 2022 74 Dominican Republic 2009 54 Dominican Republic 2015 42 Ecuador 2009 67 Ecuador 2016 58 El Salvador 2015 229 El Salvador 2022 189 Guatemala 2009 155 Guatemala 2016 63 Honduras 2009 77 Honduras 2015 56 Mexico 2009 665 Mexico 2022 269 Nicaragua 2009 75 Nicaragua 2015 77 Paraguay 2016 48 Paraguay 2022 55 Peru 2016 302 Peru 2022 311 Suriname 2009 73 Suriname 2017 32 Uruguay 2009 173 Uruguay 2016 40 Total number of firms 4853 49 Table A2: Description of variables Variable Description Quality Certificate Y:1 N:0 Dummy equal to 1 if the firm has an internationally recognized quality certificate when interviewed by WBES and 0 otherwise. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Informal Competition (CY avg.) The proportion of manufacturing SMEs in a country-year that report competing against informal or unregistered firms. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Informal Competition (service The proportion of SMEs in the services sector in a sectors, CY avg.) country-year that report competing against informal or unregistered firms. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org GDP per capita (logs) Log of Gross Domestic Product (GDP) per capita at constant 2017 International Dollars and adjusted for Purchasing Power Parity (PPP). The GDP figures used are lagged by 1 year from the year the ES was conducted in the country. Source: World Development Indicators, World Bank. Labor Productivity (logs) Log of the ratio of the firm’s total sales (in USD and deflated to 2009 prices) during the last year to the total number of workers working at the firm at the end of the last fiscal year. Total sales are converted to USD and deflated to 2009 prices using exchange rate and GDP deflator values taken from World Development Indicators, World Bank. Total number of workers include all permanent and temporary full-time workers, with temporary workers in a firm adjusted the number of months worked on average by all such workers. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Number of workers (logs) Log of the total number of workers employed at the firm at the end of the last fiscal year. Total workers include all permanent and temporary full-time workers, with temporary workers in a firm adjusted for the number of months worked on average by all such workers. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Age of firm (logs) Log of the age of the firm. Source: World Bank Enterprise Surveys (WBES). 50 www.enterprisesurveys.org Manager experience (logs) Log of the number of years of experience the top manager of the firm has working in the industry. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Herfindahl index Sum of the square of each firm’s share in the total annual sales of the firms in a country-year times industry cell. Industries are at the 2-digit ISIC Rev. 4 level. Annual sales are for the last fiscal year. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Exports (proportion of sales) Annual sales made directly abroad as a proportion of total annual sales of the firm (direct exports). Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Foreign ownership (%) Percentage of the firm’s ownership that is with foreign entities. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Women top manager Y:1 N:0 Dummy variable equal to 1 if the firm has a woman top manager and 0 otherwise. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Annual employment growth (%) Average annual growth rate of the number of workers employed at the firm over the last 3 fiscal years. Workers include all full-time permanent workers. The growth rate is computed as the number of workers at the end of the last fiscal year minus the same 3 fiscal years ago divided by the average value of the workers in the last year and 3 fiscal years ago. The resulting ratio is divided by 3 and multiplied by 100 to convert to the average annual percentage change. For a few countries, sales figures are available for last fiscal year and 2 fiscal years ago. This is appropriately accounted for in computing the employment growth rate. By construction, the variable ranges between plus 100 and minus 100. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org New Product Y:1 N:0 Dummy variable equal to 1 if the firm introduced a new product or a vastly improved one during the last 3 years and 0 otherwise. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Line of Credit Y:1 N:0 Dummy variable equal to 1 if the firm has a line of credit or a loan from a financial institution. Source: World Bank Enterprise Surveys (WBES). 51 www.enterprisesurveys.org Number of power outages (per Number of power outages experienced in a typical month month) over the last year. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Foreign Technology Y:1 N:0 Dummy variable equal to 1 if the firm uses technology licensed from a foreign company and 0 otherwise. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Website Y:1 N:0 Dummy variable equal to 1 if the firm uses its own website to communicate with clients and suppliers and 0 otherwise. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Total population (logs) Log of the total population of the country lagged by 1 year from the year the WBES was administered. Source: World Development Indicators, World Bank. Rule of Law Rule of Law indicator (point estimates) lagged by 1 year from the year the WBES was administered in the country. Source: Worldwide Governance Indicators, World Bank. Courts: Major obstacle (CY avg.) Country-year level average of the severity level of the functioning of courts as an obstacle to firm’s current operations. The severity levels that firms choose from are no obstacle (0), minor obstacle (1), moderate obstacle (2), major obstacle (3), and a very severe obstacle (4). Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Control of Corruption Control of Corruption indicator (point estimates) lagged by 1 year from the year the WBES was administered in the country. Higher values implies less corruption. Source: Worldwide Governance Indicators, World Bank. Inspections (CY avg.) Country-year level average of the number of meetings that were visited or inspected by tax officials during the last year. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Licenses: Major obstacle (CY Country-year level average of the severity level of avg.) obtaining licenses and permits as an obstacle to firm’s current operations. The severity levels that firms choose from are no obstacle (0), minor obstacle (1), moderate obstacle (2), major obstacle (3), and a very severe obstacle (4). Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Multi establishment firm Y:1 N:0 Dummy variable equal to 1 if the firm is part of a larger organization and 0 otherwise. 52 Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Training Y:1 N:0 Dummy variable equal to 1 if the firm had formal training programs for its permanent full-time employees during the last year and 0 otherwise. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Registered when started Dummy variable equal to 1 if the firm was registered operations Y:1 N:0 when it started operations and 0 otherwise. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Tax rates obstacle severity (0-4 Country-year level average of the severity level of tax rates scale) as an obstacle to firm’s current operations. The severity levels that firms choose from are no obstacle (0), minor obstacle (1), moderate obstacle (2), major obstacle (3), and a very severe obstacle (4). Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Labor regulations obstacle Country-year level average of the severity level of labor severity (0-4 scale) laws as an obstacle to firm’s current operations. The severity levels that firms choose from are no obstacle (0), minor obstacle (1), moderate obstacle (2), major obstacle (3), and a very severe obstacle (4). Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Educated workers obstacle Country-year level average of the severity level of severity (0-4 scale) inadequately educated workers as an obstacle to firm’s current operations. The severity levels that firms choose from are no obstacle (0), minor obstacle (1), moderate obstacle (2), major obstacle (3), and a very severe obstacle (4). Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org GDP per capita growth (annual, Annual growth rate of GDP (%) lagged by 1 year from %) the year the WBES was administered. Source: World Development Indicators, World Bank. Industry fixed effects A set of dummy variables indicating the industry to which the firm belongs. The industry groups are defined at the 2- digit ISIC Rev. 4. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Year fixed effects A set of dummy variables indicating the year the WBES was administered. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org 53 Country fixed effects A set of dummy variables indicating the country where the firm is located. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org Small, Medium, and Large firm Small firms are all firms that have 5 or more but less than 20 permanent full-time workers at the end of the last fiscal year. Medium firms are all firms that have 20 or more but less than 100 full-time permanent. Large firms are those that have 100 or more full-time permanent workers at the end of the last fiscal year. This classification is used by the WBES for sampling stratification purposes. Source: World Bank Enterprise Surveys (WBES). www.enterprisesurveys.org 54 Table A3: Summary statistics Variable Mean Std. Minimum Maximum Observations deviation Quality Certificate Y:1 N:0 0.104 0.305 0 1 4,853 Informal Competition (CY avg.) 0.702 0.164 0.116 0.898 4,853 Informal Competition (service sectors, CY 0.640 0.115 0.263 0.888 4,853 avg.) GDP per capita (logs) 9.411 0.454 8.402 10.051 4,853 Labor Productivity (logs) 12.700 2.836 5.566 25.275 4,853 Number of workers (logs) 2.748 0.837 0 7.356 4,853 Age of firm (logs) 2.959 0.724 0 5.063 4,853 Manager experience (logs) 2.999 0.651 0 4.094 4,853 Herfindahl index 0.075 0.106 0.000 1 4,853 Exports (proportion of sales) 0.045 0.159 0 1 4,853 Foreign ownership (%) 3.816 17.890 0 100 4,853 Women top manager Y:1 N:0 0.196 0.397 0 1 4,853 Annual employment growth (%) 3.558 16.425 -88.976 100 4,853 New Product Y:1 N:0 0.492 0.500 0 1 4,853 Line of Credit Y:1 N:0 52.265 49.954 0 100 4,853 Number of power outages (per month) 2.231 7.581 0 90 4,853 Foreign Technology Y:1 N:0 11.298 31.660 0 100 4,853 Website Y:1 N:0 54.923 49.762 0 100 4,853 Total population (logs) 15.962 1.525 12.520 18.664 4,853 Rule of Law -0.451 0.593 -1.162 1.299 4,853 Courts: Major obstacle (CY avg.) 0.245 0.139 0.015 0.520 4,853 Control of Corruption -0.331 0.704 -1.026 1.359 4,853 Inspections (CY avg.) 0.934 0.692 0.019 2.955 4,853 Licenses: Major obstacle (CY avg.) 0.193 0.143 0 0.608 4,853 Multi establishment firm Y:1 N:0 0.084 0.278 0 1 4,853 Training Y:1 N:0 0.380 0.485 0 1 4,844 Registered when started operations Y:1 N:0 0.814 0.389 0 1 4,822 Tax rates obstacle severity (0-4 scale) 1.930 1.263 0 4 4,821 Labor regulations obstacle severity (0-4 1.469 1.184 0 4 4,840 scale) Educated workers obstacle severity (0-4 1.706 1.215 0 4 4,834 scale) GDP per capita growth (annual, %) 0.917 3.948 -7.447 11.162 4,853 55 Table A4: Additional controls Dependent variable: Quality Certificate Y:1 N:0 (1) (2) (3) Panel A: Log odds ratios Informal Competition (CY avg.) 4.718*** 4.489*** 6.062*** (0.992) (0.989) (0.494) GDP per capita (logs) -1.091 -1.137 5.244*** (1.143) (1.182) (1.070) Labor Productivity (logs) 0.277** 0.256** 0.258** (0.114) (0.113) (0.116) Number of workers (logs) 0.877*** 0.899*** 0.904*** (0.183) (0.189) (0.191) Age of firm (logs) 0.473*** 0.482*** 0.483*** (0.149) (0.152) (0.153) Manager experience (logs) -0.276 -0.291 -0.297 (0.181) (0.185) (0.186) Herfindahl index -0.305 -0.420 -0.613 (1.047) (1.024) (1.093) Exports (proportion of sales) 0.757* 0.788** 0.791* (0.394) (0.400) (0.410) Foreign ownership (%) 0.013*** 0.013*** 0.012*** (0.004) (0.004) (0.004) Women top manager Y:1 N:0 -0.353 -0.387 -0.385 (0.295) (0.299) (0.304) Annual employment growth (%) -0.008 -0.008 -0.008 (0.007) (0.007) (0.007) New Product Y:1 N:0 -0.039 0.003 0.057 (0.223) (0.234) (0.232) Line of Credit Y:1 N:0 0.001 0.001 0.002 (0.002) (0.002) (0.002) Number of power outages (per month) -0.018 -0.017 -0.019 (0.021) (0.021) (0.022) Foreign Technology Y:1 N:0 0.007*** 0.007*** 0.007*** (0.002) (0.002) (0.002) Website Y:1 N:0 0.007*** 0.007*** 0.007*** (0.002) (0.002) (0.002) Total population (logs) -0.680 0.638 16.293*** (3.860) (4.077) (2.906) Multi establishment firm Y:1 N:0 0.332 0.351 0.372 (0.335) (0.330) (0.345) Training Y:1 N:0 0.993*** 0.985*** 1.012*** (0.262) (0.272) (0.275) Registered when started operations Y:1 N:0 0.936** 0.942** 0.994** 56 (0.423) (0.423) (0.424) Tax rates obstacle severity (0-4 scale) -0.008 -0.024 (0.090) (0.094) Labor regulations obstacle severity (0-4 scale) -0.024 -0.007 (0.102) (0.105) Educated workers obstacle severity (0-4 scale) -0.135* -0.130* (0.073) (0.071) GDP per capita growth (annual, %) -0.003 (0.035) Rule of Law -4.177*** (0.701) Control of Corruption 1.150 (0.905) Inspections (CY avg.) -1.572*** (0.207) Industry fixed effects Yes Yes Yes Year fixed effects Yes Yes Yes Country fixed effects Yes Yes Yes Number of observations 4,814 4,759 4,759 Panel B: Marginal effects Informal Competition (CY avg.) 0.299*** 0.289*** 0.386*** (0.062) (0.063) (0.033) Controls (as above) Yes Yes Yes Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 57 Table A5: Small firm sample Dependent variable: Quality Certificate Y:1 N:0 (1) (2) (3) (4) Panel A: Log odds ratios Informal Competition (CY avg.) 2.038* 2.442** 2.611** 3.372*** (1.096) (1.114) (1.047) (0.997) GDP per capita (logs) -5.374*** -4.082** -1.013 (1.494) (1.716) (2.192) Labor Productivity (logs) 0.961*** 0.874*** 0.742*** (0.256) (0.248) (0.235) Number of workers (logs) 0.467 0.593* 0.445 (0.369) (0.330) (0.315) Age of firm (logs) 0.298 0.350 0.371 (0.299) (0.286) (0.271) Manager experience (logs) -0.853*** -0.795*** -0.813*** (0.266) (0.256) (0.281) Herfindahl index -0.248 -1.399 (1.715) (1.536) Exports (proportion of sales) 1.003 0.710 (0.712) (0.727) Foreign ownership (%) 0.017*** 0.017** (0.006) (0.007) Women top manager Y:1 N:0 0.387 0.461 (0.338) (0.318) New Product Y:1 N:0 -0.347 (0.297) Line of Credit Y:1 N:0 0.003 (0.003) Number of power outages (per month) 0.035 (0.028) Foreign Technology Y:1 N:0 0.008* (0.004) Website Y:1 N:0 0.014*** (0.004) Total population (logs) -7.947 (6.316) Annual employment growth (%) -0.024** -0.027*** (0.010) (0.010) Industry fixed effects Yes Yes Yes Year fixed effects Yes Yes Yes Country fixed effects Yes Yes Yes 58 Number of observations 2,226 2,226 2,226 2,226 Panel B: Marginal effect Informal Competition (CY avg.) 0.083* 0.088** 0.090*** 0.111*** (0.045) (0.039) (0.035) (0.031) Controls (as above) Yes Yes Yes Yes Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 59 Table A6: Medium firm sample Dependent variable: Quality Certificate Y:1 N:0 (1) (2) (3) (4) Panel A: Log odds ratios Informal Competition (CY avg.) 2.945*** 2.768*** 2.821*** 2.492*** (0.766) (0.782) (0.845) (0.879) GDP per capita (logs) 0.432 0.101 -0.179 (1.230) (1.369) (1.613) Labor Productivity (logs) 0.282*** 0.217** 0.184 (0.089) (0.108) (0.113) Number of workers (logs) 1.376*** 1.224*** 1.131*** (0.215) (0.203) (0.197) Age of firm (logs) 0.285* 0.344** 0.373** (0.146) (0.154) (0.156) Manager experience (logs) -0.091 -0.075 -0.085 (0.236) (0.261) (0.260) Herfindahl index -1.603 -1.533 (1.458) (1.525) Exports (proportion of sales) 0.830* 0.627 (0.494) (0.485) Foreign ownership (%) 0.016*** 0.014*** (0.006) (0.005) Women top manager Y:1 N:0 -0.778** -0.901** (0.390) (0.379) New Product Y:1 N:0 0.369 (0.291) Line of Credit Y:1 N:0 -0.000 (0.002) Number of power outages (per month) -0.062* (0.035) Foreign Technology Y:1 N:0 0.007*** (0.002) Website Y:1 N:0 0.004 (0.003) Total population (logs) 2.349 (5.354) Annual employment growth (%) -0.000 -0.001 (0.007) (0.007) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 2,350 2,350 2,350 2,350 60 Panel B: Marginal effect Informal Competition (CY avg.) 0.411*** 0.350*** 0.341*** 0.294*** (0.104) (0.096) (0.100) (0.102) Controls (as above) Yes Yes Yes Yes Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 61 Table A7: Interaction with Courts quality (log odds ratios) Dependent variable: Quality Certificate Y:1 N:0 (1) (2) (3) (4) Informal Competition (CY avg.)*Courts: Major 17.749*** 14.156*** 12.964*** 14.679*** obstacle (CY avg.) (3.445) (4.487) (4.472) (4.817) Informal Competition (CY avg.) -1.800** -0.025 0.375 -27.278** (0.890) (1.116) (1.044) (13.202) Courts: Major obstacle (CY avg.) -10.148*** -7.087** -5.815* -6.295* (2.455) (3.500) (3.488) (3.385) Informal Competition (CY avg.)*GDP per capita 2.436* (logs) (1.473) Informal Competition (CY avg.)*Number of workers 1.090 (logs) (1.152) GDP per capita (logs) 1.171 1.094 -0.666 (1.194) (1.155) (1.844) Labor Productivity (logs) 0.480*** 0.410*** 0.404*** (0.118) (0.133) (0.128) Number of workers (logs) 1.174*** 0.992*** 0.231 (0.187) (0.186) (0.900) Age of firm (logs) 0.283** 0.341** 0.333** (0.140) (0.140) (0.137) Manager experience (logs) -0.294* -0.256 -0.262 (0.173) (0.184) (0.184) Herfindahl index -0.289 -0.473 (1.042) (1.105) Exports (proportion of sales) 0.619 0.654 (0.395) (0.403) Foreign ownership (%) 0.013*** 0.013*** (0.004) (0.004) Women top manager Y:1 N:0 -0.276 -0.262 (0.311) (0.310) Annual employment growth (%) -0.008 -0.008 (0.006) (0.006) New Product Y:1 N:0 0.086 0.079 (0.207) (0.205) Line of Credit Y:1 N:0 0.001 0.001 (0.002) (0.002) Number of power outages (per month) -0.019 -0.017 (0.018) (0.018) Foreign Technology Y:1 N:0 0.008*** 0.008*** (0.002) (0.002) 62 Website Y:1 N:0 0.006*** 0.006*** (0.002) (0.002) Total population (logs) 4.717 6.031* (3.041) (3.364) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 4,853 4,853 4,853 4,853 Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 63 Table A8: Interaction with Obtaining licenses obstacle (log odds ratios) Dependent variable: Quality Certificate Y:1 (1) (2) (3) (4) N:0 Informal Competition (CY avg.)*Licenses: 29.275*** 28.506*** 36.599*** 38.742*** Major obstacle (CY avg.) (4.472) (4.676) (6.825) (7.361) Informal Competition (CY avg.) -3.372*** -0.810 -1.981 -16.969 (1.024) (1.305) (1.311) (14.933) Licenses: Major obstacle (CY avg.) -21.855*** -24.582*** -31.554*** -33.123*** (3.597) (3.580) (5.308) (5.738) Informal Competition (CY avg.)*GDP per 1.079 capita (logs) (1.565) Informal Competition (CY avg.)*Number of 1.215 workers (logs) (1.115) GDP per capita (logs) 0.674 -0.274 -1.181 (0.853) (0.776) (1.365) Labor Productivity (logs) 0.480*** 0.416*** 0.408*** (0.118) (0.133) (0.128) Number of workers (logs) 1.186*** 1.010*** 0.160 (0.189) (0.186) (0.870) Age of firm (logs) 0.290** 0.346** 0.338** (0.140) (0.141) (0.137) Manager experience (logs) -0.302* -0.255 -0.258 (0.173) (0.184) (0.184) Herfindahl index -0.325 -0.473 (1.032) (1.111) Exports (proportion of sales) 0.633 0.672* (0.395) (0.405) Foreign ownership (%) 0.012*** 0.012*** (0.004) (0.004) Women top manager Y:1 N:0 -0.265 -0.242 (0.312) (0.310) Annual employment growth (%) -0.007 -0.007 (0.006) (0.006) New Product Y:1 N:0 0.149 0.144 (0.211) (0.210) Line of Credit Y:1 N:0 0.001 0.001 (0.002) (0.002) Number of power outages (per month) -0.016 -0.015 (0.019) (0.019) Foreign Technology Y:1 N:0 0.008*** 0.008*** (0.002) (0.002) 64 Website Y:1 N:0 0.006*** 0.006*** (0.002) (0.002) Total population (logs) 10.591*** 11.395*** (2.737) (2.531) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 4,853 4,853 4,853 4,853 Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 65 Table A9: Interactions with Rule of Law and Inspections simultaneously (log odds ratios) Dependent variable: Quality Certificate Y:1 N:0 (1) (2) (3) (4) Informal Competition (CY avg.) *Rule of Law -4.180*** -3.154*** -3.808*** -3.697*** (0.534) (0.884) (0.803) (1.163) Informal Competition (CY avg.) *Inspections (CY 1.087 3.286*** 2.472** 3.509*** avg.) (1.045) (0.989) (1.040) (1.352) Informal Competition (CY avg.) 0.551 2.328*** 2.526*** -11.429 (0.679) (0.821) (0.697) (11.095) Rule of Law 3.420*** 1.873* 0.217 0.513 (0.661) (1.073) (1.001) (1.095) Inspections (CY avg.) -1.445 -3.868*** -3.446*** -4.280*** (0.880) (0.879) (0.976) (1.174) Informal Competition (CY avg.)*GDP per capita 1.039 (logs) (1.161) Informal Competition (CY avg.)*Number of workers 1.085 (logs) (1.201) GDP per capita (logs) 3.061** 4.732*** 3.897*** (1.314) (1.094) (1.307) Labor Productivity (logs) 0.478*** 0.416*** 0.411*** (0.119) (0.135) (0.129) Number of workers (logs) 1.193*** 1.007*** 0.246 (0.190) (0.187) (0.936) Age of firm (logs) 0.287** 0.353** 0.345** (0.140) (0.140) (0.137) Manager experience (logs) -0.312* -0.266 -0.266 (0.175) (0.183) (0.184) Herfindahl index -0.244 -0.344 (1.050) (1.120) Exports (proportion of sales) 0.629 0.664 (0.400) (0.408) Foreign ownership (%) 0.013*** 0.013*** (0.004) (0.004) Women top manager Y:1 N:0 -0.270 -0.250 (0.311) (0.310) Annual employment growth (%) -0.006 -0.006 (0.006) (0.006) New Product Y:1 N:0 0.120 0.117 (0.207) (0.206) Line of Credit Y:1 N:0 0.001 0.001 (0.002) (0.002) Number of power outages (per month) -0.018 -0.017 66 (0.019) (0.019) Foreign Technology Y:1 N:0 0.008*** 0.008*** (0.002) (0.002) Website Y:1 N:0 0.006*** 0.006*** (0.002) (0.002) Total population (logs) 16.304*** 15.036*** (2.839) (3.341) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 4,853 4,853 4,853 4,853 Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 67 Table A10: Own firm excluded from the cell average Dependent variable: Quality (1) (2) (3) (4) Certificate Y:1 N:0 Panel A: Log odds ratios Informal Competition (CY avg., own 2.220*** 3.021*** 3.237*** 3.325*** firm excluded) (0.810) (0.764) (0.690) (0.721) GDP per capita (logs) -0.053 0.295 0.479 (0.923) (0.891) (1.267) Labor Productivity (logs) 0.483*** 0.436*** 0.409*** (0.117) (0.127) (0.132) Number of workers (logs) 1.180*** 1.131*** 1.005*** (0.185) (0.185) (0.185) Age of firm (logs) 0.297** 0.325** 0.350** (0.140) (0.146) (0.141) Manager experience (logs) -0.280 -0.250 -0.252 (0.177) (0.186) (0.185) Herfindahl index -0.297 -0.357 (1.027) (0.994) Exports (proportion of sales) 0.858** 0.613 (0.393) (0.391) Foreign ownership (%) 0.013*** 0.013*** (0.004) (0.004) Women top manager Y:1 N:0 -0.238 -0.279 (0.330) (0.307) Annual employment growth (%) -0.008 -0.008 (0.007) (0.006) New Product Y:1 N:0 0.063 (0.204) Line of Credit Y:1 N:0 0.001 (0.002) Number of power outages (per -0.016 month) (0.019) Foreign Technology Y:1 N:0 0.008*** (0.002) Website Y:1 N:0 0.006*** (0.002) Total population (logs) 0.718 (4.572) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes 68 Number of observations 4,853 4,853 4,853 4,853 Panel B: Marginal effect Informal Competition (CY avg., own 0.184*** 0.210*** 0.219*** 0.220*** firm excluded) (0.067) (0.052) (0.047) (0.047) Controls (as above) Yes Yes Yes Yes Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 69 Table A11: Falsification test using the sample of large firms Dependent variable: Quality Certificate (1) (2) (3) (4) Y:1 N:0 Panel A: Log odds ratios Informal Competition (CY avg.) 0.434 0.370 0.429 -0.718 (0.898) (0.652) (0.666) (0.830) GDP per capita (logs) 4.222** 4.117** 2.421 (1.832) (1.936) (1.738) Labor Productivity (logs) 0.394*** 0.380*** 0.326*** (0.100) (0.096) (0.094) Number of workers (logs) 0.818*** 0.748*** 0.658*** (0.144) (0.157) (0.154) Age of firm (logs) 0.313** 0.347** 0.338** (0.132) (0.149) (0.148) Manager experience (logs) -0.185* -0.159 -0.117 (0.105) (0.107) (0.094) Herfindahl index 2.321** 2.448* (1.113) (1.264) Exports (proportion of sales) 0.509 0.740** (0.310) (0.296) Foreign ownership (%) 0.002 0.002 (0.003) (0.003) Women top manager Y:1 N:0 0.488 0.427 (0.422) (0.421) Annual employment growth (%) -0.000 -0.001 (0.007) (0.007) New Product Y:1 N:0 0.735** (0.296) Line of Credit Y:1 N:0 0.001 (0.002) Number of power outages (per month) 0.011 (0.023) Foreign Technology Y:1 N:0 0.003 (0.002) Website Y:1 N:0 0.009* (0.005) Total population (logs) 11.278** (4.414) Industry fixed effects Yes Yes Yes Yes Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Number of observations 1,886 1,886 1,886 1,886 70 Panel B: Marginal effect Informal Competition (CY avg.) 0.090 0.067 0.076 -0.122 (0.187) (0.118) (0.119) (0.141) Controls (as above) Yes Yes Yes Yes Standard errors clustered on country-year in brackets. Significance is denoted by *** (1%), ** (5%), * (10%) 71