WPS6788 Policy Research Working Paper 6788 Customs, Brokers, and Informal Sectors A Cameroon Case Study Thomas Cantens Jonathan Kaminski Gaël Raballand Tchouawou Tchapa The World Bank Africa Region Poverty Reduction and Economic Management Department February 2014 Policy Research Working Paper 6788 Abstract Based on extensive interviews with informal importers informal importers. In a low-governance environment and brokers in Cameroon, this paper explains why with widespread informal practices, blanket policies customs reform aimed at reducing fraud and corruption should be avoided in order to discourage activities of may be difficult to achieve. Informal traders and brokers unprofessional and systematic bribe-taker brokers. It is (without licenses) follow various business models and also essential that customs officials disrupt information practices, which are product-specific. Overall, what asymmetries and better disseminate information to matters first are customs brokers’ practices. Information informal importers on customs processes and official asymmetries mark transactions between brokers and costs. Finally, customs should more strongly sanction importers and are accompanied by misperceptions of some informal brokers in order to reduce collusion with the costs and risks of informal brokers working among some customs officers. This paper is a product of the Poverty Reduction and Economic Management Department, Africa Region. 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://econ.worldbank.org. The author may be contacted at graballand@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 Customs, Brokers, and Informal Sectors: A Cameroon Case Study Thomas Cantens Jonathan Kaminski Gaël Raballand Tchouawou Tchapa 1 Keywords: informal, customs, Cameroon, imports, brokers, information asymmetry. JEL: F14, H26, E26. Sector board: PSM. 1 Thomas Cantens, member of the Research and Strategies Unit of the World Customs Organization, Jonathan Kaminski, consultant, Gaël Raballand, Senior public sector and governance specialist, World Bank, graballand@worldbank.org (corresponding author) and Tchouawou Tchapa, consultant. The authors would like to thank Abel Bove and Jean-François Marteau for their comments and suggestions. Background Based on extensive interviews with informal importers and brokers in Cameroon, this paper explains why customs reform aiming at reducing fraud and corruption may be difficult to achieve. Qualitative information was collected from importers, retailers, brokers, and middlemen in the five sectors in which informal trade prevails in Cameroon: furniture, secondhand clothing, secondhand cars, computer hardware, and electronic and household appliances. As far as possible, the qualitative information in this study is supported by quantitative data provided by the Cameroon Customs information technology (IT) system. In recent years, Cameroon has undertaken major customs reform in three steps: 2  The first step (2008) was implementation of a new automated system, ASYCUDA that allows data crosschecking in order to reduce the scope of corruption.  The second step, undertaken in parallel with the first step, introduced a policy of performance measurement based on a series of quantitative indicators (economic activity, processing times, effectiveness of control and sensitive procedures, and customs channels compliance). The policy was then implemented in 11 customs offices in Douala (Libom, Cantens, and Bilangna 2009). It resulted in a systematic performance assessment of the clearance operations of customs.  The third step involved establishing performance contracts for customs agents in 2010 and for some importers in 2011 in order to provide non-financial incentives for improving compliance and practices; lower risks from a customs standpoint; and lighter controls and fewer visits for importers. The impacts of the reforms were noteworthy for most quantitative indicators (Cantens et al. 2012). There was a significant reduction in the number of bad practices reported by inspectors and brokers, increases in the amount of duties and taxes per declaration, and better transactional and operational performance for importers (e.g., reductions in transactional delays). 2 For more detailed information, see Cantens et al. (2012). 2 However, most of these positive impacts have not applied to importers and brokers classified by customs as high-risk, especially those labeled “informal” by customs agents because of the practice of using other registered company names to declare imports at customs. Informal importers often operate on a small-scale basis. Customs administrations in developing countries face a common dilemma: they must facilitate trade and yet they need to fight fraud and smuggling to increase tax and public revenues. One of the problems facing them is the increased risk of fraud and smuggling (both of which affect the terms of taxation and some administrative procedures) when too much pressure is placed on importers (a classic moral hazard problem). Consequently, customs administrations have developed institutionalized bargaining procedures with the would-be smugglers in order to shift them to the “legal track.” Among these procedures are sector-level agreements on the minimum values declared for various goods. Negotiations on the acceptable minimum taxation level are common and seem to some extent efficient given the actual means and capacities of the customs administration. Contracts with well-identified importers were also inspired by a culture of negotiation that is not common in African countries. For example, exchanging promises of compliance by importers with promises of facilitation by customs is common to all customs administrations. An example is the Authorized Economic Operator (AEO) agreements in developed countries. However, devising contracts and new rules of the game with “informal” importers is difficult because of information asymmetries for which little knowledge is available on practices and constraints. Moreover, identification in the customs IT system is difficult because of bulking (consolidation of loads in containers), the lack of formal registration, and the tendency to work with informal brokers. For analytical purposes, the definition of informality requires clarification. It is a controversial subject because of the variety of existing, and sometimes contradictory, definitions. Here informality refers to any trading activity associated with importers’ business practices (no bank account or no formal accounting) using the customs approach, which is in line with the definition of informal sector by the National Institute of Statistics in Cameroon (even though they may be registered and pay some taxes) . 3 In the import sector, the case of brokers is quite straightforward. Officially, a “licensed customs agent” must process all import-export operations. Importers are not allowed to declare their goods to customs as individuals (except for personal cars). All customs agents obtain their licenses through an administrative procedure, but they are subject to providing customs with a financial guarantee. Informal brokers have no license and no identification (ID) login to access the IT customs system, which is mandatory for lodging a declaration. They thus “hire” an ID from a licensed broker and lodge declarations through the broker’s login (with a payment to the broker). According to our fieldwork, more than 200 informal brokers may be operating in and around the port of Douala (versus 220 official brokers, including official representatives of embassies). Informal brokers are not identified in the customs IT system. As for importers, the picture is even more complicated. Some individuals who are informal importers borrow the identities of registered firms or individuals in order to manage their operations. Customs does indeed allow individuals (not only firms) to import large quantities of goods and handle a significant number of operations. Other importers have informal business models: firms are registered and organized as corporate bodies, known by the customs administration, but they do not have computer-based systems and transit services, and their bank account information does not capture the bulk of their activities, which tend to favor tontines, rotating savings and credit associations (ROSCAs). More generally, customs agents classify some operators as “informal” when these operators do not apply simple internal administrative and management rules, allowing and facilitating the ex post customs control (after the exit of goods through crosschecks between accounting documents, invoices, and so forth—that is, once goods have left the customs area). As a result, importers and customs brokers can both be considered “informal.” Based on our interviews with informal importers, wholesalers, and retailers trading in the specific goods, we learned that (1) the informal sector is extremely heterogeneous (among brokers and importers); and (2) the informal sector comprises businesses ranging from highly professional brokers (but who lack the financial resources to obtain a license) to risk-taking brokers and importers who try from time to time to evade taxes in order to increase profit. This heterogeneity is reflected in the port dwell times for cargo, which range from 2 to 14 days. The 4 increased time is likely due to some brokers taking on additional risks in the clearance process, resulting in a longer clearance time. By contrast, short dwell times may reflect the significant network of personal contacts at work, developed by informal brokers among the low-level staff of private and public agencies. Such staff members regulate the port exit and can delay or accelerate procedures. For example, staff members may deliver the receipts or documents required for the successive steps of the clearance process. Several factors seem to play a part in identifying professional informal importers and brokers: the number of containers; the quantities of products imported; sourcing, transit, and marketing strategies; and the customer base. Moreover, based on interviews, recent customs reform seems to have had an interesting result: despite relatively persistent corruption and fraud, it has become more difficult for informal brokers to find corruptible customs agents willing to allow informal brokers to elude increased declaration values (however, the remaining informal brokers are the least risk-averse). The rest of this paper delves further into the determinants and descriptions of such business models and clearance practices. The section that follows presents a literature review, followed by sections devoted to the data collection methodology, some quantitative information on informal practices for car imports and garments, a demonstration of how crucial brokering practices are in understanding informal importers’ practices, the heterogeneity of importers’ sourcing strategies and clearance strategies, and the relationships between informal importers and brokers. The final section presents our conclusions and provides lessons for customs reform in a low-governance environment. 1. Literature Review The performance and practices of informal traders in international trade has not been studied extensively. Despite a lack of material, Akinboade and Kinfack (2012) have demonstrated that tax burdens discourage informal firms from registering their small and medium enterprises (SMEs) in Cameroon, especially in the area of customs regulations. Using a data set of detailed 5 information on firms that have moved from formal to informal status in four African countries (Côte d’Ivoire, Kenya, Nigeria, and Senegal), Gajigo and Hallward-Driemeier (2012) find that productivity and corruption (informal payments by firms to public officials) significantly lead to firms returning to the informal sector. Moreover, they return to the informal sector even after initially registering as a start-up. Relatively limited research and literature are available on tontines ((ROSCAs) 3 and how this system allows for the financing of SMEs and informal firms. According to Rozas and Gauthier (2012), ROSCA participation in Cameroon has had substantial positive effects on entrepreneurial performance. ROSCAs facilitate surmounting financial constraints, restricted access to credit and credit market failures, and selling firms. They are primarily dedicated to financing working capital over investment. Kemayou et al. (2011) explain how ROSCAs make financing easily available for small business formation. These authors also describe the Cameroon savings culture of mobilizing small groups of 10 or more people who share strong social ties. These ties can be based on kin, regional connections, professions (e.g., brokers and small importers), neighborhoods, and friendships. Tontines allow members to obtain loans at the same interest rate as the savings rate thanks to mutuality and membership rules. Loans are readily available with far fewer constraints and are better enforced because of the trust-based nature and functioning of personal relationships. Few papers are as significant to the study of informal traders and brokers as the work of Bilangna and Pagou (2013). They have extensively studied customs and trader relationships using data on fraud cases. Their findings are similar to those derived from our interviews:  Individuals and firms show similar propensities toward fraud.  Negative perceptions of informal trading also shape the perceptions of customs agents, which has led to blanket policies from the Cameroon customs system. 3 A tontine encourages members to save on a regular basis (weekly or monthly), and the accumulated sum is made available as loans to members of from 1 to 100 million CFAF (Communauté Financière Africaine francs) biweekly or monthly. 6  Improving customs skills and capacities would generate better results and tax revenues than increasing tax rates alone. 2. Data Collection and Methodology Little is known about informal sector practices for several goods traded at Douala, the main entry point in Cameroon and the Central African Economic and Monetary Community (CEMAC). Clothing, textiles, tires, spare parts, and appliances are sold in retail markets and by specific resellers who are small importers with suppliers in Europe, China, or Dubai. They can also be tied to local importers. Therefore, it was very important to generate data and obtain information on importers’ business practices and models, constraints, and overall clearance practices. The choice of goods to be studied was dictated by three main criteria:  All goods should be informally traded and widely consumed, and customs agents should consider their market dominated by “informal traders.”  Goods should account for a large share of total revenues.  The global value chain for the selected goods can be empirically explored in the field, and retailers would be accessible to interviewers. The fieldwork consisted of semi-structured interviews with informal importers, wholesalers, and retailers trading in the following goods: electronic and household appliances, furniture, secondhand clothing, and secondhand cars. After petroleum products (HS 2701), personal cars (HS 8703) and secondhand clothing (HS 46309) are the two main revenue sources in Douala. 4 Of the total revenues collected, petroleum products account for 10 percent, personal cars for 7.8 percent, and secondhand clothing for 4 percent. Furniture and electronic and household appliances have lower shares, even if calculation is difficult because of product variety. 4 HS refers to the Harmonized System code. 7 Information was collected on importer and customer relationships by interviewing importers. Interviews were conducted with retailers in markets and stores. Detailed illustrations and examples are provided in the appendix. Table 1 summarizes the number and composition of the pool of interviewees. 5 Table 1. Breakdown of Interviews with Brokers, Importers, and Retailers, Port of Douala, Cameroon Category Number of Comments interviews Brokers 12 (5 informal and 7 Formal and informal brokers were formal–CAD) interviewed equally. Most informal brokers adhered to professional practices and had knowledge of activity. It was difficult to locate occasional bala-bala brokers.a Importers Total: 21 The smallest-scale importers were more SH clothing, 4 difficult to identify (those who do more SH cars, 7 bulking and occasional imports); there Furniture and was an equal proportion of formal to appliances, 7 informal. Most interviews were Computer hardware, 3 scheduled in advance. Retailers Total: 7 Some interviews were scheduled in SH clothing, 3 advance; others were spontaneous visits SH cars, 1 in marketplaces and shops. Furniture and appliances, 1 Computer hardware, 2 Source: World Bank Note: All interviews were carried out in Douala. CAD = commissionnaires agréés en douane (certified customs brokers); SH = secondhand. a. A bala-bala is a nonprofessional broker. The questionnaire for brokers was designed to capture broker typology in terms of clearance practices and the selection of and matching with customers, public administration relations (particularly customs), and types of service and contractual arrangements with importers. A final 5 Interviewees were selected through human networks. Because it was relatively difficult to find individuals willing to talk openly about their business, interviewees were recommended by neighbors, relatives, or business relations. 8 section was added to record the salient issues and inefficiencies experienced by the profession and related sector. The questionnaire for importers was divided into three parts: (1) the personal histories and characteristics of individuals and firms; (2) import and sourcing strategies, as well as clearance practices and relationships with brokers; and (3) marketing strategies with customers, day-to-day management issues, and relationships with public administrative agencies. The questionnaire for retailers and small vendors focused on their relationships with importers, their market and business relationships and practices, and the downstream marketing aspects of their activities, including administrative and regulatory issues. The sections that follow summarize and analyze the data and information collected during the interviews. We specifically demonstrate the importance of broker practices in clearance processes, the heterogeneity of importers’ and brokers’ business models, and the implications of importers’ and brokers’ relationships with customs officials. 4. Quantitative Assessment of Informal Trade for Secondhand Cars and Clothing In terms of methodology, informality raises an important issue. All import operations, whether from the perspective of an importer or broker, are collected in the IT system, but differentiating them from formal operations is more challenging. The IT customs database houses all information on informal traders, especially in Cameroon where a tax identification number is required for an import operation and a formal broker must intermediate to officially authorize imports for trade. The following sections describe the structure of importers’ and brokers’ markets related to personal cars and secondhand clothing. Both goods are among the main revenue sources of Douala port (and therefore the entire country) and are imported by both formal and informal traders. 9 Car Imports Because trade classification does not distinguish between new and secondhand cars, our study focused on informal traders and secondhand cars. The analysis then proceeded in several differentiation phases to assess the importance of informal trade. During the first phase, we excluded buses from tariff line 8703 because interviewees rejected the hypothesis that buses were being imported by informal traders. During the second phase, we attempted to identify cars sold on the informal market. In 2012 approximately 32,000 vehicles were imported into Douala port for consumption 6 purposes. The breakdown of car imports by value appears in table 2. Table 2. Breakdown of Imported Cars by Value, Douala port, Cameroon Category of vehicles by declared value Percentage of total number of secondhand vehicles ≤1 million CFAF (€1,500) 49.0 >1 million and ≤2 million CFAF (€3,000) 28.0 >2 million and ≤3 million CFAF (€4,500) 10.0 >3 million and ≤4 million CFAF (€6,000) 4.0 >4 million and ≤5 million CFAF (€7,500) 2.0 >5 million and ≤10 million CFAF (€3,000) 3.0 >10 million and ≤15 million CFAF (€3,000) 0.8 >15 million and ≤20 million CFAF (€3,000) 1.6 >20 million CFAF 1.6 Source: ASYCUDA system. What can be observed? The market is clearly segmented for cars under €4,500 (87 percent), which are presumed to be secondhand cars. Because all cars are imported from Europe and car taxation is approximately 53 percent, new car values should not be under €4,500. This gray area of values between 4,500 and €7,500 is the range at which cars are most likely to be secondhand, but not necessarily bought by informal traders. These cars could also be European imports for personal use by individuals who are not resellers. 6 The cars were identified in ASYCUDA as “supplementary units.” 10 In the third phase and continuing in the same class, we tried to separate individuals who import their own cars from the resellers. We assume that, in view of the average personal income, an individual would not import more than one car a year for his or her own use. Thus importing more than one secondhand car per year should be deemed a traded activity. Therefore, we assessed the share of importers for each category of car and the share of importers who had more than one operation a year for each category. This approach offers the lowest ceiling for evaluating the share of informal traders in view of the fact that some informal traders may use numerous tax identification numbers at the same time (the importers classified as “importers for personal use” can be also traders). The aim is to evaluate the secondhand car informal trade in a rank order (see table 3). Table 3. Main Characteristics of Importers and Imports by Car Value, Douala port, Cameroon Category of (declared) value Percentage of imported Percentage of Percentage of cars of total number of cars importers processingimporters processing more than one more than one operation (as share operation (as share of all categories) of own value category) ≤1 million CFAF (€1,500) 48.0 15.00 30 >1 million and ≤2 million 29.0 8.00 28 CFAF (€3,000) >2 million and ≤3 million 12.0 2.00 17 CFAF (€4,500) >3 million and ≤4 million 4.0 0.80 22 CFAF (€6,000) >4 million and ≤5 million 2.0 0.30 14 CFAF (€7,500) >5 million and ≤10 million 3.0 0.40 CFAF (€15,000) >10 million and ≤15 million 0.8 0.10 CFAF (€22,500) >15 million and ≤20 million 0.4 0.08 CFAF (€30,000) >20 million CFAF 0.5 0.10 Source: ASYCUDA system. The figures in table 3 reveal that the share of traders who may also be resellers is non-negligible when the value of the car is less than €7,500. One assumption is that some informal trade may 11 exist for cars valued at between €4,500 and €7,500, but it is marginal compared with cars valued at under €4,500. Finally, the lowest informal trade estimate can be summarized as follows: 4,531 “importers” have imported cars at least twice a year, and the value of each car is less than or equal to 3 million CFAF (€4,500). Among these 4,531 importers, only 148 are identified as firms; the others are identified as individuals. The total amount of taxes and duties paid by these importers is approximately 12 billion CFAF, which is equivalent to that for the sixth most frequently traded product in Douala port, demonstrating that this kind of trade is already non-negligible. We also observed informal traders and their practices. If it is not possible to rely on identification numbers to “identify” traders, then observing the brokers’ market can provide supplementary information. These 4,351 importers work with 109 of the 220 total customs agents operating in Douala port. However, car imports are mainly carried out by individuals (or informal brokers) through the special customs unit for individuals in which the declaration is lodged by the customs official. One plausible assumption is that most importers-resellers declare on their own initiative or make use of “informal brokers,” who represent them. Such operations are quite concentrated because two agents declared more than 500 cars and 10 agents more than 100 cars, whereas half the customs agents in this unit lodged less than 10 operations in 2012. Secondhand Clothing The first step in analyzing the secondhand clothing market was to differentiate traders. When examining the structure of imports, one sees striking patterns emerge. The distinction among importers of secondhand clothing is clearer than that for car traders because the weight per operation is an important criterion for distinguishing formal importers from informal importers. Some traders import containers full of secondhand clothing (the weight per operation is approximately 20 tons or 12 tons for a 20-foot container), whereas others import secondhand clothing with other goods or secondhand clothing in “small” quantities (about 100 or 50 kilograms per operation). Some 1,600 importers are in the second category, and some have imported 2–5 tons a year through several operations. Most are probably informal traders, but they are not counted in the 12 following analysis for two reasons: (1) the 264 importers in the first category represent 98 percent of the imported value; and (2) informal traders in the second category cannot be distinguished in the IT system from traders who import secondhand clothing for personal use (the same problem that applies to importing cars). In the second step, we explore how secondhand clothing is imported by these 264 importers, taking into account that importing secondhand clothing may be a main or a secondary activity. Table 4 describes the characteristics of the secondhand clothing import mix. Table 4. Secondhand Clothing Import Mix Characteristics, Douala port, Cameroon Number of Average share Number of Median, standard importers of declared imports by PM deviation, and (share of value of (legal entities) maximum values total secondhand and PP (real (CFAF) imports of clothes persons) secondhand clothes) Importers who import 96 (37%) 23.0% 11 PM; 75 PP Median: 30 million only 6309 SD: 92 million Max: 432 million Importers who import 9 (3%) 0.7% 1 PM; 8 PP Median: 24 million mainly other products SD: 24 million (more than 60% of Max: 83.7 million total value) Importers who import 10 (4%) 6.0% 3 PM; 7 PP Median: 19 million 50/50 (between 45 and SD: 555 million 55% of total value of Max: 1.787 billion other products) Importers who import 146 (56%) 70.0% 20 PM; 126 PP Median: 67 million mainly 6309 but not SD: 274 million exclusively (between Max:1.6 billion 45 and 1% of total Min: 8 million value of other products) Source: ASYCUDA system. Note: 6309 refers to worn clothing and other worn articles. PM = personne morale or legal entity; PP = personne physique or real person; SD = standard deviation. One main conclusion that can be drawn from these data is that secondhand clothing importers rarely import clothing as a minor commodity in terms of value; nearly all importers who import secondhand clothing do so as a main activity. This sector is dominated by informality. However, albeit informal, this sector provides significant revenues, and the dominant business model 13 imports exclusively or mainly secondhand clothing, which means that this sector is economically robust and well established (but also politically sensitive). A second conclusion is that the 264 importers who imported 98 percent of the value are mainly individuals (216 of 262). A third conclusion is that in terms of content, standardization in declarations is according to the content, value, and weight of the secondhand clothing. This is the result of negotiations between importers and customs to control smuggling. Such an agreement sets a minimal value per kilogram for the weight of a container: for the 264 importers, the median value per kilogram is 498 CFAF, with a standard deviation (SD) of 217 CFAF, and the weight per operation is 23 tons (median), with a standard deviation of 4 tons. This negotiation process ensures that a minimal value is applied to imports and in the process secures important revenue for the port of Douala. And this value is taken seriously. In Cameroon, secondhand clothing importers organized a strike in 2001 to protest a decision by the head of the customs office. According to importers, the office had increased the minimum value of a container to an unacceptable level. Table 5 describes the characteristics of secondhand clothing imports. Table 5. Characteristics of Secondhand Clothing Imports, Douala port, Cameroon Number of Value per kg Number of Number of agents (CFAF) importers per imports by PM (percentage category (legal entities) of total and PP (real persons) imported value) Customs agents’ activity 3 (16%) 458–538 41 22 PM; 78 PP (value treated) mainly in secondhand clothing (more than 50% of total activity) Customs agents’ activity 7 (63%) 474–542 139 16 PM; 123 PP in secondhand clothing (between 20 and 50% of total activity) Customs agents’ activity 10 (13%) 452–554 86 7 PM; 70 PP in secondhand clothing (between 5 and 20% of total activity) Customs agents’ activity 30 (8%) 407–1,156 70 10 PM; 60 PP in second -hand clothing 14 (less than 5% of total activity) Source: ASYCUDA system. The profession is highly concentrated. The 264 importers are accustomed to working with only 50 customs agents. Three major brokers deal with 120 different importers (accounting for 47 percent of the declared value). 7 Their values per kilogram vary from 480 CFAF to 542 CFAF. A difference of 60 CFAF (which is less than 10 percent) in the clothing value has important implications for taxes and duties, because it is approximately 1.5 million CFAF per container or equivalent. The amount 1.5 million CFAF “saved” on the value is approximately 750,000 CFAF in taxes and duties “saved” for the importer or 25 percent of the customs taxes and duties paid per container. Of these three major brokers, two have focused their activity on secondhand clothing (representing 40 percent of the value), and one devotes 21 percent of his activity to secondhand clothing. All three brokers deal with secondhand clothing and one other major product: motorcycles (the minimum value of which is informally set by agreement), cloth (the value of which is set by agreement, but this value is double the minimum value of secondhand clothing), workbooks, or milk. An exclusive relationship between an importer and an agent is rare: we found only seven cases of such a relationship for a total value share of 1.3 percent of secondhand clothing imports. However, we could not confirm whether the same importer used several identification numbers. One assumption was that there are connections among some brokers, informal brokers, and informal traders. To check this assumption, we examined more closely 32 brokers who process approximately 20 tons per transaction (which means one container per operation). The three major official brokers (representing 47 percent of the secondhand clothing market) rely on secondhand clothing for their main product, but the value share treated is less than 50 percent (40 percent, 20 percent, and 10 percent). The shares are more or less similar when calculated in terms of the number of operations. Only two brokers among the 33 have a share of secondhand 7 Of 264 importers, 144 (18 PM and 126 PP) deal with just one agent (and represent 45 percent of the value treated). The maximum is one PP who has 9 agents. Sixteen customs agents (of the 50 who work in secondhand clothing) work with just one secondhand clothing importer. 15 clothing higher than 50 percent. Twenty brokers stated that they usually clear customs in one day (the median), which is good compared with the more than two days usually required for clothing to clear customs in Douala port. Standardizing values and weights through agreements with the administration facilitates the secondhand clothing imports; despite scanning all secondhand containers, the control process does not cause delays. Two observations are important in assessing the importance of the connection between informal and formal brokers, and the impact this connection has on clearance time and value. First, one major broker processes many declarations and importers with a low declared value per kilogram (median = 466 CFAF per kilogram, standard deviation = 224 kilograms). The customs clearance time is limited (median = one day), but the standard deviation is high (five to eight days). One assumption is that this broker shares a login and password on the IT system with informal brokers who personally process, at the same time, fewer importers’ declarations. As a consequence, these informal brokers are efficient in terms of value and delays. They seem to have personal contacts with both the administrative staff of private companies (warehouses, container terminal, SGS, 8 scanning machine) and public agencies (port authority, customs, police), which help to lower the value of goods and speed processing. Second, among the 13 brokers with clearance time exceeding one day (which is the median for all 20 brokers), nine brokers have values exceeding 500 CFAF per kilogram (the highest). The assumption is that these 13 brokers work with less informal brokers. These cases help to explain broker practices when brokers play a major role. 5. The Importance of Broker Practices on Clearance Processes Knowledge of the entire import process, including importer practices and customs administrative procedures, allows better understanding of the key role performed by brokers. Also important is the double informational rent at their disposal to use as leverage with customs and importers. This finding was ascertained through the interviews we conducted. 8 Société Générale de Surveillance, the pre-shipping company in Cameroon for customs revenues. 16 The Blurred Definition of Informal Brokering Approximately 200 official certified customs brokers (commissionnaires agréés en douane, CAD) operate in Cameroon. The CAD status allows them to conduct some import operations on behalf of their clients. In Cameroon, importers must use a customs broker to clear goods, excluding cars. It would be naïve to believe that formal brokers work exclusively with formal importers and therefore exhibit the most professionalism. In reality, many formal brokers partner with informal brokers from whom they secure a rent. For brokers, there is no correlation between being licensed and being professional. The professional practices and performances of both formal and informal brokers 9 vary. Broadly speaking, three broker categories emerged during the study:  Formal brokers (working with informal importers)  “Professional” informal brokers with training, education, and experience in the transit sector and a sound knowledge of customs practices, procedures, and documents  Nonprofessional brokers, otherwise known as bala-balas by their more professional counterparts. Some have an intuitive knowledge of the profession, while others are quite experienced and, in general, are more eager to take on risks with fewer containers. Brokers estimated that between 50 and 80 percent of informal brokers are bala-balas (accounting for less than 30 percent of declarations). Because of their nonprofessional practices and knowledge, the bala-balas have a negative spillover effect for customs. Specific skills, knowledge, and experience are needed for effectively managing the clearance process, properly assessing the risks, and ensuring profitability. Most full-time brokers, whether formal or informal, have some experience in the brokering sector and some specialized training in management, sales, and administration. They must handle complex organizational and management tasks, in addition to the usual administrative and negotiation responsibilities. 9 A significant number of informal brokers work part-time and are appointed by importers on account of social norms (they are recommended by family members or regional networks). In general, they lack the knowledge and competencies to handle the required tasks and pose a high risk for importers. 17 Professional informal brokers have the advantage of being more flexible in their practices and are able to adapt to their customers’ needs and risk exposure. They also may propose better services (such as pre-financing a part or the full cost of operations), but they face increased risks from controls and visits. They tend to cover part of the unexpected costs, unless these costs fall under the responsibility of their customers (e.g., intentional mistakes during the declaration process). They charge less than formal brokers for their services. Customers of bala-balas are, in general, less aware of the risks associated with clearance operations. Although they may end up paying lower fees and taxes, they run higher risks because bala-balas often derive added gains through fraud and double informational rents with less transparent information flows. In what follows and in appendix A are several illustrations of and additional details on the matching of brokers and importers. Nonprofessional brokers tend to underdeclare the value of imports and take more time in order to find the most corruptible officers and agents who they can bribe at each step of the clearance chain. They can save several million CFAF on a 40-foot container. The gain would be split among the bribes needed for corrupt agents, the courtesy reduced fees for the customer, and higher profit margins for the broker, who can earn up to 1 million CFAF, or at least 100,000– 200,000 CFAF, more than professional brokers on the same volume (see table 6). Problems arise when bribery is not possible at a step in the chain. To clear the container, non- professional brokers then have to cover the additional fees and fines after controls and revised declaration values, making everyone, starting with the customer, ultimately worse off. In these situations, some brokers may choose to disappear because they do not have the working capital to finance revised declaration values fees and do not feel they can legitimately claim additional payment from their customers. With informal brokers, the unexpected costs are never covered unless specified. 10 Formal and some informal professional brokers provide additional services to their customers but not to all, depending on their long-term relationship and proven reliability. They notably offer 10 Specified costs might be port fees when a shipment stays over 11 days at port and the costs of a scanner or additional visits when not expected or not intentionally the result of the importer’s behavior. 18 pre-financing services (reimbursable 60 days after delivery), business services abroad for purchases and sourcing, storage and warehouse facilities, and some other logistical services (e.g., bulking). They can use the importers’ merchandise as collateral, provided those importers are well established. Table 6 presents some figures provided by the brokers interviewed on the customs and other operational costs at the port and related fees. Table 6 Declared Values and Overall Costs for Importer and Brokers’ Fees, Douala port, Cameroon Declared value of Overall cost charged Informal brokers’ fees Formal brokers’ goods (in million to importer fees (official) CFAF) 10 million 6 million 30,000–50,000 for validation 245,000 + 50,000–150,000 honorarium 6.3 million 4 million (under Professional: 80,000 for 230,000 (secondhand clothing): administrative value); validation + 75,000–175,000 20 ft. container 3–3.5 million (when Bala-bala: 80,000 for (reference value) not enforced) validation + 150,000– 300,000 12 million (secondhand 8 million (under Professional: 120,000 for 275,000 clothing): 40 ft. administrative value); validation + 100,000– container (reference 6.5–7 million (when 250,000 value) not enforced) Bala-bala: 120,000 for validation + 300,000– 500,000 Cars (secondhand) Depends on market Professional: 50,000– ? rates 150,000 Bala-bala:10,000–500,000 (including validation fees of about 10% of import value) Source: Interviews. Formal brokers are generally able to cover unexpected expenses (associated with formalities) and overhead (receiving some provisional fees up to 3 percent beforehand from their customers), and exercise more responsibility over the entire clearance process. They tend to be more effective in terms of clearance operations and can deliver containers in three to four days upon arrival at a gate (with no additional control or revised declaration values). Informal brokers provide the same service in 5–10 days on average because of tighter financial constraints (they cannot advance cash funds for customers) and a lack of knowledge about procedures and the relevant documentation. However, some informal brokers perform just as well as formal ones. 19 In a sizable heterogeneous pool of informal brokers, some possess knowledge and practices similar to that of the formal brokers. They offer the same type of services and fare equally well in terms of clearance operational and transactional performance. Formal brokers are, in general, able to clear the import declaration at the port of origin (beforehand) when the customer can provide all the relevant documentation. Professional informal brokers aim to process clearance in less than 11 days with complete documentation in hand, taking into account the potential delays and operational inefficiencies of customs and port agents. Informal brokers cannot handle the entire process on their own; they must find and be tutored by a formal broker. In general, informal brokers must pay formal brokers validation fees. Formal brokers validate the declarations of the informal brokers because of the legal liability associated with the formal broker’s professional license. If their informal brokers create too many problems and deliver low performance for customs, they can be targeted as a high risk by the risk analysis unit of customs and screened and monitored more frequently. Such attention will generate additional costs for them. So, in general, formal brokers also try to screen informal brokers to limit the extent of bad practices. Professional brokers market their services, but rely largely on their own personal and professional networks. Informal brokers and bala-balas find the majority of their customers around the GUCE 11 by soliciting “door to door.” Most importers are aware that these brokers are available if they would like to bypass formal brokers. Many brokers interviewed deplore the lack of professionalism, the way the sector is regulated, and the unequal treatment by customers because of personal networks. Most formal brokers interviewed have a good sense of their informal brokers’ practices and associated risks, tending to select them based on their own personal risk assessment. However, others do not screen informal brokers, knowing (rightly) that they will not be sanctioned. Brokering Practices with Customs Administration Brokering practices with the customs administration are classified into two categories, according to the interpretations of brokers: the transactional/formal route, which is respectful of customs 11 Guichet Unique des Opérations du Commerce Extérieur (GUCE) is a single window for foreign trade transactions in Cameroon. 20 regulations and “good clearance practices,” and the negotiated route where brokers and importers aim to minimize the amount of tax duties and attempt tax evasion, irrespective of the set import value and customs regulations (see appendix C for examples). Most small-scale importers believe the negotiated route is the most cost-effective for them. This belief stems often from a misperception of the risk that may be propagated by their own brokers. One of every five customs agents is estimated to be corrupt, but in most cases this is at the request of agents’ customers and importers. 12 Similar products are sometimes valued differently; the harmonizing process is ongoing or only partly enforced. When litigation cases become too costly, most importers prefer to surrender (because delays entail additional costs) and pay what the customer claims. Many brokers tend to invent operational strategies to avoid the scanner because they deplore the discretionary behavior and manipulation that accompanies it, providing opportunities for agents to request informal payments. Discretionary behavior is also found at GUCE, where the time spent on declaration files may be exaggerated, despite files having been properly documented and processed by brokers and offering informal payments to agents to enhance transaction performance. One of the main criticisms involves Port Authority of Douala (PAD) fees, which are often used to generate additional and unjustified fees and rents. Some brokers lament the lack of operational efficiency caused by visits, which could have been avoided had all containers gone through the scanner first and been documented with photographs as evidence rather than resorting to follow-up and secondary controls. Overall, they complain about the redundancy of operations, the transactions between different authorities, and the ensuing delays. Distrust is prevalent. Some agree, however, that since the introduction of both performance contracts and GUCE, the entire process has improved. In summary, the shipment evaluation process seems to be a key driver of practices, whether based on referenced value per container or on quotes and values per item. Thus the packing list matters, and sometimes the list refers to both criteria such as that for secondhand clothing. In 12 Relationships with other administrative agents also involve corruption, bottlenecks, and inefficiencies. Overall, there seems to be a common structure of practices and learn-by-doing knowledge to improve transactional efficiency for small-scale importers, in addition to sector-specific issues and strategies. 21 other cases (e.g., computer hardware), importers complain about the lack of consistency of accepted values for each item and prefer a shipment-level administrative value (“mercurial value”). Other importers prefer more differentiation. Fraud and corruption practices seem widespread, and monitoring practices could better target and adjust according to the risk profile of importers. Finally, the acquisition of formal broker licenses is inadequately regulated and monitored because licenses can be rented, sold, or ceded to a third party without effectively controlling for financial and management skill. Some formal brokers partner with associates who do not have the necessary skills and experience to operate. Compared with countries in East Africa, in Cameroon the number of official brokers is limited (see table 7), and therefore the barrier to entry seems to be higher (which also could reflect the level of guarantee associated with obtaining a license). However, it is also important to note that licenses are never removed in practice. Therefore, the guarantee level is questionable because it only acts as a barrier to entering the profession and is not used as a sanctioning tool. As a barrier to entry, the guarantee does not separate brokers with good practices from others. The guarantee only screens out those with limited financial resources. Table 7 Comparison of Number of Brokers and Freight Forwarders and Ratio of Imports to Number of Brokers, Cameroon and East Africa Number of customs brokers and Ratio of imports freight (US$, millions) to Country forwarders number of brokers Burundi 43 18.8 Kenya 1,200 12.6 Rwanda 124 12.6 Tanzania 480 21.5 Uganda 339 15.3 Cameroon 200 32.8 Sources: Cameroon: World Bank; East Africa: Arnold et al. 2011. 22 In addition to importers and brokers, some customs officials interviewed confirmed our findings, such as the practices of specific small-scale importers. Overall, the perception remains that informal importers are more willing to take risks because they incur lower costs, even when detected. However, they also confirm that what matters the most is the way brokers navigate the clearance process. These importers and brokers also agree that risky behavior correlates with higher profit margins, exposing the risk-profitability trade-off. 6. Heterogeneity of Importers’ Models, Sourcing Strategies and Clearance Strategies Sector-specific organizational models of sourcing and purchasing 13 seem to have developed in Cameroon (see appendix C for examples of models of sourcing). In the secondhand clothing sector, most importers are connected to family members abroad who handle purchases, whereas home-based importers are faced with heavy coordination for marketing and quality demand issues (e.g., unsold quantities—a frequent occurrence). Because the supply chain involves large importers and wholesalers, handling purchases through frequent travels abroad is not possible. Sourcing should be a specialized activity well connected to downstream activities in the chain. In the hardware sector, most sourcing is carried out through email and telephone orders. Long-term and trust-based relationships seem more frequent and widespread than in other sectors. For furniture and secondhand electronic devices and household appliances, sourcing is carried out by means of travel and fewer long-term relationships with providers, and more often through spot market transactions. This also appears to be the case for secondhand vehicles. Purchases and sourcing through family networks are less established than for secondhand clothing, although still significant. Overall, secondhand products are in high demand because they are the most affordable for urban populations. These products are sold at prices competitive with those for new low-quality 13 From our fieldwork, we learned that most small-scale importers originate from the western part of the Cameroon, and thus they may have some network advantages. Regional networks are the basis of ROSCAs (tontines), through which a significant share of set-up and working capital for small-scale importers and wholesalers is channeled. These networks can also provide advantages of scale and scope. 23 products such as ones originating from China (e.g., Amsua electronic devices, but the term Amsua is also applied generally to low-quality Chinese products). The price gap between secondhand and new products is in the range of 200–300 percent for secondhand cars or computers in good condition. Marketing and commercial practices and strategies are also heterogeneous among retailers and wholesalers. In general, fierce competition takes place at the retail level, with its issues regarding the quality and commercial practices that are specific to various import sectors. As expected, the secondhand clothing sector is the most organized, characterized by tremendous variety among the intermediate actors and the ways in which they conduct business. Some market segment specialization helps to increase profits and is associated with specific commercial strategies (e.g., hardware for business and institutional customers). Additional monitoring and repair services can be offered or bundled to satisfy customers. Some marketing networks are clearly visible, although this calls for a more in-depth, larger-scale value chain analysis. We also observed loyal and reciprocal relationships between wholesalers and retailers, and at times with end customers. Personal networks seem to play a significant role in marketing strategies, which can be expected from the way these sectors function. We now turn to business models and estimated margins based on information reported by interviewees (and therefore should be taken with caution). Table 8 presents the figures provided by importers and, when available, the retail margins. Table 8 Estimated Margins by Type of Informal Imports, Douala port, Cameroon (in CFAF (unless otherwise noted)) Sector/product Secondhand Secondhand Furniture Household Computer clothing vehicles appliances hardware FOB cost per 40 ft.: 13 Car: 2 million Roughly 40 ft.: 12–20 20 ft: 16 million container million Truck and kit comparable million (value) + (average) 20 ft.: 6.5 from pieces: 11 to travel and million million household sourcing (1 appliances million) Clearance 40 ft. container: Car (imports Transit: €3,000 Transit: 1.6–2.3 and transit 11.0 million = transit: 14% are mostly Customs: 4.8 million costs 2.6 million Customs: 57 bundled million Broker: 300,000 (transit) + 4.9 million + 5% with Broker + PAD + Customs: 6 million Broker: 10,000– household other taxes: 1.2 million 24 (customs) + 2.5 50,000 appliances) million Other: 800,000 million (VAT) PAD: 21,000 VAT: 2.5 + other (1 Stamp: 10,000 million million) Truck + pieces: 20 ft. container: 7 million overall 6 million Estimated 30–40 million 4.3–4.5 million 25–35 million 30–35 million value of 15–20 million 23 million sales Gross 6 million (20%) Car 3.8–5.5 million 2–7 million = margin after all sold— 600,000 = 14% = 15% (direct 7–20% 4 million with (direct sales); sales) (3–5% for rebates and some do only Other retailers); unsold items 300,000 or operational 10–20% on (13%), other 7% (for costs: 3–4 printers and operational wholesale), million per year Xerox (retail) costs (salesmen 100,000– (employees and + warehouse); 300,000 or 5– taxes) some do only 7% (for retail) Estimated net 500,000—1 Truck + pieces: margins: 2.3–3.5 million (2–4%) 4 million (17%) million (9-10%) Retailers: 2– Retailers: 10% 10% (20% for shoes) Source: Interview. Note: FOB = free on board; VAT = value added tax; PAD = Port Authority of Douala (PAD). The clearance process depends on the imported product category. Targeted importers and brokers with a high-risk profile may design import strategies that avoid lengthy and costly controls. For example, car importers can decide to place the vehicles in containers with other secondhand appliances and electronic devices. 14 The typical strategy followed by high-risk and small-scale importers is to pack containers mixed with different products, intentionally arranging the products to maximize the undervaluation for merchandise declaration and to bypass the evaluation by inspectors (sometimes hiding the highest-value products in places virtually unreachable by agents). This last strategy has further implications for sourcing and provision in the originating countries. Only secondhand clothing importers seem to be specialized and organized differently from other 14 Some brokers and importers always look for one car to put in their container even though their main business is household appliances and furniture. 25 importers. Because secondhand clothing has an administrative price, there is much less gain by mixing it with other products, thereby simplifying the import process. The other sectors are much less concentrated than that for secondhand clothing with many more semi-wholesalers and direct retailers, also owner-importers. Most secondhand importers also specialize in wholesale. 7. Relationships between Brokers and Importers The relationships between brokers and importers appear to be dependent on trust, a strategy most often favored by regional and family networks. However, when considering the medium- and long-term perspectives, regional and family networks seem to matter less and less for importers and brokers who wish to comply with regulations. Moreover, brokers usually have information that importers do not usually have and make good use of this information asymmetry. This is especially true of risk-taking importers. From appendix A, several observations can also be made about importer-broker relationships:  When importing computer hardware, importers select formal brokers who specialize in this trade.  Large-scale importers seem to be efficiency-oriented, favoring cost predictability and short dwell times, but they seem to favor professional informal brokers. Pre-financing resources and professional broker skills also seem to matter.  In a few cases, matching continues to be dictated by personal relationships. Some informal importers expect specific broker performance (e.g., dwell times and delivery date) and a willingness to negotiate litigation cases, avoid administrative annoyances, and use their experience and connections with customs agents. In general, brokers are specialized in the way they navigate the clearance process (e.g., time effectiveness and compliance versus risk and delayed leveling of duties). They also select importers based on contractual arrangements and the services provided. 26 Total clearance costs (which include duties, taxes, brokerage fees, and port charges) matter most. Most informal importers are not interested in the cost components of what their brokers pay for clearing containers. The informal importer’s only concern is acquiring products and selling them in a timely manner and at an affordable cost via trust and reciprocal relationships. Thus they tend to favor informal broker connections recommended by relatives or friends rather than develop relations with a formal broker, who may require too much information sharing and who may not be trustworthy. In table 8, informal and formal costs and the charges of brokers do not differ much (except for bala-bala). Therefore, it is necessary to focus more on the process and practices selected by informal and small-scale importers and the ways in which these choices materialize in relationships with brokers and subsequent business arrangements. Informal importers usually use informal brokers because they perceive that formal brokers would not be interested by their business and would be too expensive. On the supply side, because of barriers to entry, including licensing costs and certain qualifications, the number of formal brokers in operation is limited. Informal importers are either too risky or not profitable enough for most formal brokers. However, some brokers lacking a large company customer base and with less professional connections are often keen to work with informal and small-scale importers. These importers mainly work with informal brokers because they draw up implicit contracts that are more likely to be enforceable in the context of shared high risk, together with recurring interactions. Our interviews with brokers reveal a striking heterogeneity in broker and importer business practices. Some brokers are mainly performance-oriented and start by evaluating customers’ import values. They will not try to negotiate declared values, and they rarely engage in litigation unless they have the means to support their claims and are almost certain to win. These brokers often have an informative approach toward their customers, explaining the procedures, risks, benefits, and costs of the declaration and clearance processes. They ensure that additional risks 27 are assumed by the customer in order to secure fees without having to commit to costs and undergo delays. 15 On the demand side, many informal importers aim to save more on customs duties and taxes, preferring to take higher risks. We observed that informal brokers operating with small-scale, informal importers often specialize with this type of importer because they have difficulties developing business relationships with larger-scale firms and importers because of a bad reputation or image (sometimes incorrect), including possible problems with the tax administration. Most of these brokers are aware of the risks and costs faced by importers making use of their services, but who in general have no alternative. Small-scale and informal importers may represent 70–80 percent of an informal broker’s annual volume, if not more. Table 9 presents rough estimates (with intervals) of transit activity outcome by category of broker. Table 9 Comparison of the Main Characteristics of Brokers (Formal and Informal) and Their Perceptions and Reality from Informal Importers’ Points of View, Douala port, Cameroon Type of broker Formal (CAD) Professional informal Bala-bala Number of containers, 25–300, 15–100 + secondhand 1–20 + secondhandcars, 5– import value per year 100–5,000 cars,100–1,500 250 (CFAF, millions) Gross estimated income 6–250 2–20 6–25 (million CFAF) Average dwell time 2–6 4–10 1–10 (days) Perceptions of informal Too expensive, Cheapest brokers and importers do not wish to accept flat budget work with informal importers 15 This can also happen with formal brokers who do not comply with official regulations, negotiate their own fees, and try to respond to their customers’ demands for the clearance process. 28 Reality May work partially with Use information informal importers when asymmetry against customer base limited importer and flat budget to overcharge and are not cheaper than professional informal broker Source: Interviews. Note: The income of formal brokers includes wages for company’s employees and the validation fees paid by informal brokers. CAD = commissionnaires agréés en douane (certified customs brokers). The business models of informal and small-scale importers are characterized by a high degree of heterogeneity according to their financial resources, import volumes, and profits. But they also are characterized by trade-off risks (being caught by customs when goods are undervalued) and profits. This affects whether informal and small-scale importers decide to work with brokers based on the expected services and the level of risk in the clearance process. However, in this context social capital (family, kinship, ethnicity, neighborhood) is an important method for matching brokers with importers. 8. Conclusions and Lessons for Customs Reform in a Low-Governance Environment Based on extensive interviews in the entire Cameroon customs sector, it appears that the same kinds of business practices are found in both the informal sector and formal sector. However, in the informal sector more extreme business practices emerged because of the increased difficulty for customs to sanction transactions when unable to trace operations along each step of the customs process, increasing the scope for moral hazard. Informal brokering is quite important in Cameroon because it is perceived to fulfill demands from informal importers who do not dare to operate with formal brokers, fearing they are too expensive and would not wish to work with them (which is probably a misperception for most of them). Moreover, it seems that the number of formal brokers is excessively low because of the high barriers to entry. 29 Brokerage practices matter enormously for customs because brokers primarily shape the business environment for customs. Bad practices from some brokers have a significant negative spillover effect for the entire administration. In order to address this issue, a three-pronged strategy should be pursued by customs:  Disseminate more information on import requirements and procedures and the detailed costs of reducing the information asymmetry. (The most professional informal brokers and importers have called for better treatment by customs officials and more differentiation from unprofessional informal brokers.)  Reduce barriers to entry for a broker’s license in order to create more competition in this area and include the professional informal brokers.  And finally, to make the "player behavior" of the bala-balas (high stakes, high reward) better sanctioned. Based on the current ratio between the number of brokers per U.S. million dollars in imports, Cameroon customs should consider developing a comprehensive package aimed at increasing the traceability of operations and rewarding the most professional informal brokers by reducing the information asymmetry between importers and brokers. With this end in view, customs should also strive to avoid applying blanket policies to the IT system for some sectors because systematic controls have become an incentive for importers to evade taxes and duties. But this would not be an easy task because brokers are sometimes politically powerful, and some past initiatives have not led to the expected impacts. For example, the Trade Facilitation Committee, GUCE, and customs wanted to establish a certification program for brokers and freight forwarders, but could not achieve their plans because some “professionals” did not meet the required criteria and sabotaged the whole initiative. A differentiated approach (through contractual incentives) could also be envisioned to address that matter. 30 Shaping brokers’ market structure and practices is therefore the required next step to changing informal trade practices in an incentive-compatible approach. However, for political economy reasons, a coalition between customs and some brokers and importers should be encouraged. 31 References Akinboade, O. A., and E. Kinfack. 2012. “Regulation, Awareness, Compliance and SME Performance in Cameroon’s Manufacturing and Retail Sectors.” International Journal of Socio- Economics 39 (12): 933–50. Arnold, J., Mathenge, N. Dihel, and N. Strychacz. 2011. “The Role of Clearing and Forwarding Agents in Reforming the EAC Logistics Sector.” Africa Trade Policy Note 17, World Bank, Washington, DC. Bilangna, S., and L. Pagou. 2013. Représentations, pratiques officielles et officieuses des administrations frontalières en rapport avec l’informel: une étude de cas inspirée des douanes du Cameroun. Paper presented at WCO Conference on Research on Informal Business, International Trade, and Customs, Brussels, June 3–4. Cantens, T., G. Raballand, S. Bilangna, and M. Djeuwo. 2012. "Comment la contractualisation dans les administrations fiscales peut-elle limiter la corruption et la fraude ? Le cas des douanes camerounaises." Revue d’économie du développement 26 (3) : 35–66. Gajigo, O., and M. Hallward-Driemeier. 2012. “Why Do Some Firms Abandon Formality for Informality? Evidence from African Countries.” Working Paper Series No. 159, African Development Bank, Tunis, Tunisia. Kemayou et al. 2011. “Tontine et Banque en contexte camerounais.” La revue des sciences de gestion 249–50: 163–70. Libom, M., T. Cantens, and S. Bilangna. 2009. “Se Regarder dans le Miroir.” Discussion Paper 08, Sub-Saharan Africa Transport Policy Program, World Bank, http://siteresources.worldbank.org/EXTAFRSUBSAHTRA/Resources/DP08-Cameroon-Full.pdf. Rozas, S., and B. Gauthier. 2012. “Les tontines favorisent-elles la performance des enterprises au Cameroun. RED 26: 5–39. 32 Appendix A. Broker-Importer Matching Characteristics Type of Wholesaler– Informal semi- Retailer– Formal Formal semi- broker- informal wholesaler– informal retailer–formal wholesaler– importer broker informal broker broker informal broker broker association Product Secondhand Secondhand Electronic Appliances Appliances and clothing cars and devices + cars and furniture furniture appliances Annual import 85 million 55 million 10 million 4 million 30 million (2 value FOB (bulking, other (bulking) containers, 40 ft.) (CFAF) retail activities) Dwell time ? 11 days 5–6 days for 7–8 days 2 days cars Selection ? Broker fees: Broker fees: Full integrated Broker fees: criteria and 10,000/car 30,000/car service (3 100,000/container practices feet); 800,000 Based on trust for transit, and performance taxes, and broker fees Type of Formal semi- Formal semi- Retailer– Formal semi- Transit service broker- wholesaler– wholesale–- without broker wholesaler– provider and importer formal broker informal broker formal broker retailer– informal association broker Product Computer Appliances Containerized Computer Organize hardware cars, hardware and bulking + consumables, electronic appliances and new clothing, devices furniture + appliances cars Annual 30 million (2 35–40 million 20–25 million 50 million (4 10 million import value containers, 40 (4 containers, (2 containers, containers, 40 FOB (CFAF) ft.) 40 ft.) 20 ft.) ft.) Dwell time 1–2 weeks 3–7 days 10 days 7 days Selection 300,000– Personal Does everything Satisfaction for Personal criteria and 330,000 relationship on her own; just services, relationship practices broker fees pays validation accountability, Overall 4.5 per container fees to formal and pre- million broker (50,000) financing of clearance cost operations per container Broker fees are Does not know regulated ones details 33 Type of Wholesaler– Wholesaler–informal Wholesaler– Retailer–without broker- informal broker broker informal broker importer broker and formal broker association Product Computer hardware Secondhand clothing Computer Secondhand cars + hardware and devices screen Annual 200 million (8 150–300 million (5–10 90 million (6 100–300 million import value containers, 40 ft.) containers, 40 ft.) containers, 40 (50–100 cars ) FOB (CFAF) ft.) Dwell time 2–3 days 3 days 10–15 days 12–17 days Selection Work with several Compare formal broker With a formal Prefer to handle criteria and different brokers: with informal broker: 9.5 broker for new on own because of practices 9,450,000 CFAF million CFAF per hardware and unpredictable per container for all container for formal informal delays in the clearance costs; broker, 8.5 million CFAF broker for does not know for informal broker with secondhand clearance process details (but stated good performance + items; requires (SGS, by brokers in the experience and good much disembarking, bill) connections with customs bargaining formalities)— agents + prefinancing (not with inspectors declaration based offered by formal brokers) over scanner on argus (blue book) Source: World Bank. Note: FOB = free on board; SGS = Société Générale de Surveillance. 34 Appendix B. Sourcing Strategies from Interviews with Importers Sector/product Secondhand Secondhand cars Computer Furniture Household clothing hardware appliances Origin Belgium Germany France Germany Belgium Purchases Handled by Travel: vehicle Handled by Travel: Handled by one family parks—known one family collecting one family member based location, member places where member based abroad depends on based abroad stocks are abroad quality and price + travel adequacy Sourcing Quality Identify By Internet Secondhand Quality concerns available cars by and shops marketplaces concerns Internet in Hamburg Packing Composition Purchased cars Only With Furniture with and calibration with appliances hardware, no appliances, household and furniture consumables and sometimes appliances and (not cars consumables containerized) (bulking) Relationships Familial Spot market No specific No trust- Familial with providers networks relationships relationships, based, some networks (cash), quality payment by regular orders checks, random transfer (cash), but trustworthiness have to travel for quality checks Upstream Competitive, Competitive, Competitive, Oligopolistic, Competitive, market quality quality good quality quality issues quality structure heterogeneity heterogeneity heterogeneity Importer’s Mid-scale: 50 Small-scale (10– Mid-scale: Mid-scale: 30- Small-scale (3 category million CFCF 15 cars three 32 million 40 million 20 feet import value times a year import value import value containers 4 times a year) Sector/product Secondhand Secondhand Computer Furniture Household appliances clothing cars hardware 35 Origin US and Europe Europe and Germany Canada US Purchases Associated Travel: by By email— Travel: collecting family member building secondhand places where stocks based in US + networks in prices are are + orders by email interaction known higher under and telephone within with marketplace pressure of familial networks Cameroon- s and on the demand from based family to spot North Africa, better serve Asia, and East demand Europe Sourcing Handled by Targeting Depending on Secondhand familial links Toyota’s customers’ marketplaces of (from NGOs cars needs and Hamburg or handled and collecting irrespective obsolescence by family relatives places) of age (Pentium IV currently) Packing According to Cars + New and With furniture the market’s devices secondhand informal loaded standards Relationships Familial Spot market Usual ones Trust with family; with providers links— transactionsunder mutual otherwise only spot marketing —rapid understanding, market transactions strategies to quality expected better serve checks delivered market quality, demand monitoring and trust- based relationship Upstream Competitive, Competitive Competitive, Competitive and market structure quality, and quality, and quality issues composition knowledge (calibration) issues issues Importer’s 5–6 up to 20 50–100 cars 6 40 ft. 4 40 ft. containers a category 40 ft. a year containers a year containers a year year Source: World Bank. 36 Appendix C. Examples of Clearance Practices and Relationships with Customs Administration Sector/product Secondhand Secondhand Household Computer hardware cars clothing appliances and furniture Identified issues Poor Corruption; Some import Damages and functioning of errors in regulations; thieves encountered GUCE; port processing; customs duties during control visits; police claims customs duties (too high); contestation of new- and fines; (too high); contestation of tax looking secondhand thieves and disruption of values by customs products to damages on the scanner; for the items for reclassify in higher way to the connection which there are tax categories; slow vehicle park; problems to flat-rate processing; lack of long SGS ASYCUDA agreements; “DIT knowledge of tax procedure; all mafia”; number of categories and vehicle-related checkpoints in values by customs taxes; problems port; intentionally with products slow processing loaded in the by inspectors vehicles (for evaluation purposes); all annoyances at checkpoints (police, environmental services) Solutions Bribe police Declaration Fraud— Use of packaging and other based on underestimation of list for declarations; administration packing list; import value; mercurial values and security avoid mistakes bribes would be helpful; officers to save in simplification of the time and documentation whole process; resources; better adapt monitoring to the risk profile of importers (number of visits, scanner) and reduce in the long term if the process is better mastered by importers Strategies Make Working with Load consumables Solicit post- declaration on a an informal in appliances and clearance visits at personal basis; broker is old TVs home; be assigned 37 collective associated with (forbidden) to the same agreement more bribes for inspector; better would not be customs agents functioning of helpful and can help GUCE; declaration save costs and based on receipts avoid additional when possible; and costly bribing when monitoring; influencing the post- avoid scanner scanner report made by customs agents Practices Official quotes Cost and time of Is close to the Complicated are respected, clearance ones for negotiations and a but not always process and secondhand risky process; fiscal occurrence of clothing, except GUCE is not categories; scanner and that there are operational; many brokers should visits depend on systematic control bottlenecks are help just in the practices of visits and more based on personal case of the broker; difficulties in mood and litigation— taxes levied monitoring, which relationship issues; otherwise according to generate occurrence of everything can already agreed- additional costs control visits be handled on rates and for every actor depends on the through GUCE values overall value of if better containers functioning Source: World Bank. Note: GUCE = Guichet Unique des Opérations du Commerce Extérieur; SGS = Société Générale de 38