One-stop border posts (OSBPs): An assessment of the economic and social impact

Ayele, Yohannes; Calabrese, Linda; Gharib, Mohamed; Mendez Parra, Maximiliano Research Report One-stop border posts (OSBPs): An assessment of the economic and social impact ODI Report Provided in Cooperation with: ODI Global, London Suggested Citation: Ayele, Yohannes; Calabrese, Linda; Gharib, Mohamed; Mendez Parra, Maximiliano (2023) : One-stop border posts (OSBPs): An assessment of the economic and social impact, ODI Report, Overseas Development Institute (ODI), London This Version is available at: https://hdl.handle.net/10419/289504 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/4.0/ Report One-stop border posts (OSBPs): an assessment of the economic and social impact Yohannes Ayele, Linda Calabrese, Mohamed Gharib and Max Mendez-Parra July 2023 2 Readers are encouraged to reproduce material for their own publications, as long as they are not being sold commercially. ODI requests due acknowledgement and a copy of the publication. For online use, we ask readers to link to the original resource on the ODI website. The views presented in this paper are those of the author(s) and do not necessarily represent the views of ODI or our partners. This work is licensed under CC BY-NC-ND 4.0. How to cite: Ayele, Y., Calabrese, L., Gharib, L.M. and Mendez-Parra, M. (2023) One-stop border posts (OSBPs): an assessment of the economic and social impact. ODI Report. London: ODI 3 Acknowledgements This report has been funded by TMA as part of the ODI–TMA partnership. We thank Kevin Rombo and Catherine Nanzigu from TMA for their valuable comments and contributions. In addition, we are very grateful to the border staff of the Governments of Kenya, Tanzania and Uganda for their assistance in the collection of data and during interviews. We are also grateful to Melaku Desta, Coordinator of the African Trade Policy Centre (ATPC), for his insightful peer review and valuable comments on the final version of this report. The report does not necessarily reflect the views or positions of TMA or ODI. Any errors are the responsibility of the authors. 4 Contents Acknowledgements ................................................................................................. 3 Acronyms ................................................................................................................. 5 List of tables ............................................................................................................ 6 List of figures ........................................................................................................... 6 Executive summary ................................................................................................. 7 1 Introduction ..................................................................................................... 10 2 Methodology ................................................................................................... 12 Transport costs and prices ........................................................................ 12 Identifying products .................................................................................... 13 Business and household surveys .............................................................. 14 3 OSBP impacts on costs and prices ................................................................ 16 Introduction ................................................................................................ 16 Basic information about businesses .......................................................... 17 Impact of OSBPs on costs and prices ....................................................... 17 3.3.1 Assessing the impacts of OSBPs on maize ........................................... 18 3.3.2 Assessing the impacts of OSBPs on rice .............................................. 26 Businesses’ responses on OSBPs, costs and prices ................................ 27 Informal trade and OSBPs ......................................................................... 28 Challenges in trading through OSBPs and potential impact on prices ...... 29 4 OSBP impact on household income ............................................................... 31 Introduction ................................................................................................ 31 Trade costs, consumer prices and household income .............................. 32 Impact of OSBPs on household welfare based on the survey data ........... 34 5 Conclusions and recommendations ............................................................... 37 References ............................................................................................................ 39 Appendix 1 Business questionnaire .................................................................. 41 Appendix 2 Household survey .......................................................................... 44 Appendix 3 Household demographics and other characteristics ...................... 48 5 Acronyms CPI Consumer Price Index EAC East African Community EU European Union OSBP one-stop border post PMA Plan for the Modernisation of Agriculture TMA TradeMark Africa UK United Kingdom UN United Nations USAID United States Agency for International Development WTO World Trade Organization 6 List of tables Table 1 Kenya’s CPI and trade with EAC and rest of world ($ ’000s) ............. 13 Table 2 Typical costs incurred when trading across the border (KES) ........... 21 Table 3 Costs and prices of maize in the market (KES) ................................. 23 Table 5 Impact of OSBPs on rice price (KES) ................................................ 26 Table 6 Impact of OSBPs on consumer prices (rice) ...................................... 27 Table 7 Impact of OSBPs on prices, by different dimensions ......................... 28 Table 8 OSBP impacts on household spending (KES) ................................... 33 Table 9 Incomes since the OSBP has been introduced (%) ........................... 35 Table 10 Perceived effects of the existence of OSBPs (%) .............................. 35 List of figures Figure 1 Surveyed households (%) .................................................................. 15 Figure 2 General value chain for maize and rice ............................................. 19 Figure 3 Trading across borders and main sources of supplies (%) ................ 20 Figure 4 Effect of the OSBP on the prices of traded commodities (%) ............ 27 Figure 5 OSBPs and use of informal cross-border trade (%) ........................... 29 Figure 6 Challenges when trading across the border (OSBPs) (%) ................. 30 Figure 7 Activities households are involved in at OSBPs (%) .......................... 34 Figure 8 OSBPs and their effects (%) .............................................................. 36 Figure 9 Impact of OSBP on local economy (%) .............................................. 36 Figure A4.1 Most commonly consumed goods from across the border (%) ...... 49 Figure A4.2 Reasons household buy commodities from the neighbouring country (%) ................................................................................................. 50 7 Executive summary This report aims to assess how reduced trade costs resulting from the introduction of one-stop border posts (OSBPs) have been transmitted to the rest of the economy and, in particular, how this has affected consumer prices and household expenditure across value chains. For this purpose, we selected two products based on two criteria. First, the products had to be widely consumed by households in the region. Second, they had to be regularly traded among the countries of the East African Community (EAC) and across OSBPs. The study analysed the Busia (Kenya–Uganda) and Taveta–Holili (Kenya–Tanzania) OSBPs. The assessment of the impact on consumer prices is a significant component to understand the impact of OSBPs on poverty in East Africa. To provide prima facie evidence of the impact of OSBPs on prices and its transmission across actors involved in the supply chain of selected consumer products, we assume that a reduction in cross-border trade costs provides consumers several benefits, through lower prices, higher quality and availability of greater product variety. The change in prices depends on at least two main channels: the competition channel and the cost channel. In the competition channel, the reduction in trade costs may increase the participation of domestic and foreign businesses in importing, exporting and trading activity, which can increase competition. As competitors influence firms’ pricing decisions, the pro-competitive welfare gains of a trade cost reduction can be realised through lower markups by businesses, bringing down firms’ profit margins and affecting the prices consumers pay. In the cost channel, the decline in trading costs may affect prices directly. We assume that the selling prices observed in 2022 reflect both the pro-competitive effects of OSBPs and the actual reduction in the costs of cross-border trade observed because of OSBPs. Therefore, the approach followed to analyse the impact of OSBPs on consumer prices is to project what the selling prices would have been in 2022 had the cost of trading across the border and the profit rate remained the same as in 2015. That is, for the trade costs assumption, we use the share of trade costs in 2015 as a reference point; for the analysis of the procompetitive effects, we use the profit rate in 2015 as a base. For maize, we found the following: • Under Scenario 1 (baseline scenario) (No gain from competition channel and trade cost channel) – that is, the trade cost and the profit rate remain the same as in 2015 – we estimate that consumers saved KES 788, equal to a 12.3% saving to these consumers. The 8 gain from maize in Holili, Tanzania, is 6.8%. This scenario assumes what would have happened if OSBPs had not been there altogether. • Under Scenario 2 (Gain from competition channel but no gains from lower trade cost) – that is, the profit rate in 2022 is as observed in 2022 but the trade cost is as in 2015 – we estimate that consumers saved KES 575 for maize originating from Busia, Uganda. This is equal to 9% of savings to these consumers. Savings for consumers from maize in Holili, Tanzania, equal 6.8%. • Under Scenario 3 (No gain from competition channel but gains from lower trade cost) – that is, the markup rates for 2015 and 2022 are the same but there is a lower trade cost in 2022 – we estimate that consumers in Mombasa, Kenya, saved KES 720 for maize originating from Busia, Uganda, equal to 10% of savings to these consumers. In a similar analysis for maize coming through Taveta– Holili (Tanzania), savings for consumers equal 4.5%. In general, under different scenarios, the gain for consumers in terms of lower prices could range from 9% to 12.3% for maize originating from Busia and from 4.5% to 6.8% for that originating from Taveta–Holili. Using the same approach, we find a similar result but with a lower magnitude of gain for rice. The survey results support these findings. Although OSBPs were introduced several years ago, and thus businesses may not be able to calculate the exact amount of their impact on prices, and traders may not have considered what the price might have been if OSBPs had never existed, they can compare whether the price increased or decreased after their establishment. Close to half of business traders said that OSBPs had reduced the prices of commodities they traded across the border. We also provide some evidence on the impact of OSBPs on household income, based primarily on the assumption that a reduction in consumer prices induced by lower transport costs owing to OSBPs allows us to calculate the monthly food expenditure per adult equivalent households have saved. Using the fall in consumer prices and the Kenya Integrated Household Budget Survey for 2015/16, we find that rural households saved on average KES 68–76 in monthly adult equivalent terms for maize from Busia and KES 34–68 for maize from Taveta–Holili. Similarly, core urban households saved KES 64–87 per month. The household gains from rice are lower than the gains from maize for both rural and urban households. In other words, OSBPs could be saving some rural Kenyan families one month of their total food expenditure on maize every year. To complement these findings related to the impact of OSBPs on household income, we surveyed households living at the border. We found the following: • In aggregate, for 40% of household respondents, they or a member of the household was involved in activities at the OSBP, either working directly at the OSBP (5%), providing services or goods to the 15 households, with 38 male respondents (24%) and 122 (76%) female respondents. Figure 1 Surveyed households (%) Note: The total number of households is 315. Source: Household survey. A. By country B. By border post C. By country and border post 0.0 10.0 20.0 30.0 40.0 50.0 60.0 Kenya Tanzania Uganda 48.0 48.5 49.0 49.5 50.0 50.5 51.0 Busia Taveta-Holili 23.0 23.5 24.0 24.5 25.0 25.5 26.0 Kenya-Busia Kenya-Taveta Tanzania-Holili Uganda-Busia 16 3 OSBP impacts on costs and prices Introduction A reduction in cross-border trade costs, such as through the introduction of OSBPs, has several benefits for consumers, including lower prices, higher quality and more product variety. The impact on consumer prices depends on the prices firms charge, which depend on at least two main channels: the competition and the cost channels. In the competition channel, lower crossborder costs increase the participation of domestic and foreign businesses in trading activity, thus increasing competition. The pro-competitive welfare gains of the trade cost reduction for consumers can be realised through lower markups charged by businesses, bringing down firms’ profit margins and affecting consumer prices. 1 For example, because of OSBPs, more businesses from Uganda may export to the Kenyan market, and more Kenyan businesses may start to be involved in cross-border trading. In the cost channel, the lower trade cost may lead directly to lower consumer prices, although several factors, including market structure, prohibit full pass-through to consumers in terms of lower prices. 2 Precisely quantifying the sources of consumer gains that may arise as a result of reduced cross-border trade costs is very difficult, and these are harder to disentangle from other economic factors. Any effort in this regard requires granular-level data using complicated modelling and econometric analysis. Moreover, the decomposition of prices into costs and markups that is necessary to calculate consumer gains is complex. This section aims to present prima facie evidence on the impact of OSBPs on prices and its transmission across actors involved in the supply chains of selected consumer products using survey data collected from 166 business traders (see Section 2.3). Section 3.2 provides basic information on the business interviewees. Section 3.3 discusses the impacts of OSBPs on the cost of trading. Section 3.4 provides a detailed discussion of the impact of OSBPs on prices. Section 3.5 analyses informal trade and OSBPs. Finally, Section 3.6 1 Decomposing these prices into costs and markups is not straightforward. 2 Recent empirical work examines the impact of trade cost-reducing trade liberalisations on prices charged by domestic firms. De Loecker et al. (2016) find that domestic price reductions are small when taking into account the large reduction in trade costs brought by the liberalisation. Edmond et al. (2015), using Taiwanese data, find that international trade increases competition and reduces markup distortion. 17 presents challenges traders face at OSBPs when trading across the border, and recommendations suggested by these traders. Basic information about businesses For more than half of the businesses surveyed, the primary economic activity is retail (58%); this is followed by wholesale (30%) and farming (9%). Direct importers and brokers constitute 1% and 3% of the sample, respectively. Regarding women-owned businesses, 66% are involved in cross-border trade as retailers, compared with just 37% of men-owned businesses. The main activity of men-owned businesses is wholesaling (48%); this share is only 22% for women-owned businesses. Half of the businesses have sole proprietorship; this is followed by partnerships (29%) and limited partnerships (14%). Only 6% are shareholding companies. Most business respondents were women (71%). Most businesses (79%) are not formally registered, with not much difference across countries (Kenya 87%, Tanzania 77% and Uganda 65%) or border posts (Busia 74% and Taveta–Holili 84%). This is despite the fact that the average and median business ages are 13 and 10.5 years, respectively. The traders have had a long presence in the region: they could provide essential information on the impacts of the OSBPs as they had been operating since before their establishment. Most of the businesses are small, with on average two permanent employees and one temporary worker. The maximum number of permanent employees is 10. Impact of OSBPs on costs and prices Trade theory shows that higher trade costs are associated with higher prices of traded products (WTO, 2005). The direct impact of the introduction of OSBPs is increased efficiency through a reduction in duplicative customs procedures, leading to reduced time to cross borders. This leads to reduced transport costs, arising from a reduction in parking fees, lower costs on accommodation and subsistence for drivers and a significant reduction in labour inputs for transporters. The reduction in the cost of transport could also benefit not only transporters but also forwarders and other trade operators. Mendez-Parra and Calabrese (2021) find that OSBPs have significantly reduced the time needed to process cargo consignments, border crossing times and the associated costs involved in cross-border trade. Under a perfectly competitive market structure, the lower trade costs should be passed through – that is, translated into lower traded commodity prices and producer and consumer prices. This is one of the main channels through which consumer benefits are expected to materialise as a result of trade liberalisation or trade facilitation activities that reduce trade costs. However, several complex factors related to supply chains lead to incomplete pass-through to prices in the form of lower commodity prices. First, the pass-through to import prices is incomplete at the border. Even if the change in costs is passed on in the form of lower consumer prices, this process may be sluggish and take time. Second, distribution services such as local storage, transportation, wholesaling, insurance, retail, etc. increase the local value-added content of the imported good in the final consumer price, which helps dampen the effect of the reduced trade cost on the 18 consumer price. Besides, distributors may also actively adjust their profit margin to absorb some of the reduced trade costs. 3 As a result, the reduced trade costs may be absorbed by other actors involved in the supply chain, such as transporters and distributors, instead of benefiting consumers in the form of lower prices. This is especially the case when there is a lack of competition across the supply chain, for example in the transport, wholesale or retail sectors. Furthermore, the rate of transfer of an increase in trade costs to producers and consumers is faster and more direct than that of a reduction. That is, the cost pass-through to consumer prices is asymmetric: while an increase in the cost of production is passed quickly to consumers, decreases do not usually transmit fully to lower consumer prices. This section looks first at maize, presenting a detailed breakdown of the value chain, costs, prices and the impact of OSBPs on consumer prices and welfare gains. It then assesses the impact of OSBPs on consumer prices of rice. 3.3.1 Assessing the impacts of OSBPs on maize Kenya is a major consumer of maize but domestic production is insufficient. Meanwhile, Uganda produces maize but is not a major consumer. As a result, much of the maize in Kenya comes from Tanzania and Uganda. The major source of maize in Busia, Uganda, is Masindi, while maize coming through Holili in Tanzania into Taveta, Kenya, is sourced from Moshi and other markets near Arusha. The major value chain actors involved in maize production and trade are similar across Kenya, Tanzania and Uganda: farmers, aggregators, wholesalers, importers, retail distributors and transporters. Figure 2 provides a schematic representation of the value chain for maize to help us understand the price transmission from the reduced transport cost as a result of the OSBP to the selected product. The schematic representation is similar across countries studied: the value chains and the actors may differ a little but are broadly similar in Kenya, Tanzania and Uganda. 3 There is a wide literature examining the incomplete pass-through of reduced trade costs, such as lower tariffs or a depreciation or appreciation in exchange rates, to import and consumer prices (see Campa and Goldberg, 2005; Berman et al., 2012; Amiti et al., 2014; Fitzgerald and Haller, 2014; Fontagné et al., 2018). 19 Figure 2 General value chain for maize and rice Source: Adapted from PMA (2009); World Bank (2009); USAID (2010); Ahmed (2012). In the survey, most businesses in Busia, Uganda, source maize supplies from farmers (68%), wholesalers (58%) and brokers (56%), while those in Busia, Kenya, source mainly from direct importers (50%) and wholesale traders (33%) (see Figure 3). In Taveta–Holili, 58% of businesses trade maize and 49% trade rice. In Moshi, 44% of businesses trade maize, 44% rice and 11% oil. In Taveta, Kenya, most suppliers obtain their maize and rice from wholesalers, while those in Holili obtain most of their maize and rice from farmers and wholesalers. In general, at all border posts, maize is the dominant traded commodity across the border, followed by rice. Traders follow the maize supply chain to take advantage of supply and demand in various markets to buy maize at lower prices, consolidate the product and eventually cross the border into Kenya. Specifically, rural agents buy maize from numerous farmers. After aggregating a large enough quantity, they sell this to urban traders and processors (Daly et al., 2017). Next, the maize flows to regional towns, urban markets, major buying centres and export markets. For example, traders from Kenya can purchase from the wholesale market or go to markets on the outskirts of Moshi. Large buyers of maize use semi-trailers (300 bag carrying capacity) to buy maize from various markets. Wholesale markets normally open twice a week. 20 Figure 3 Trading across borders and main sources of supplies (%) Note: Authors’ calculations. Brokers also play a role in determining the prices of products in the wholesale market. Most times, brokers fix the price. They purchase maize from trucks and sometimes actually own maize trucks. Most traders can store maize for two to three months. Wholesalers, however, store maize for only six weeks maximum, given the volatile nature of prices and the arrival of new maize, which reduces demand for old maize. It is risky for wholesalers in Busia, Uganda, to store maize for long periods. Traders from Busia, Uganda, also resort to smuggling maize across the border, especially when it is not dried enough and may fail the moisture content test at the border. Others cross the border by bicycle, disguised as small traders moving one bag at a time across the border to the local market, thus avoiding phytosanitary requirements. Maize that is rejected at the border for failing phytosanitary, moisture content and pest tests is usually sold through illegal channels at a lower-than-market rate. Before the maize reaches Mombasa, several costs are involved. Businesses reported that the typical costs incurred during cross-border trading were transport costs (68%), customs duties (66%) and local municipal council charges (49%); the rest (7%) comprised other kinds of costs. Table 2 shows typical costs incurred during cross-border trade by businesses. 020 40 60 80 100 Arusha (#=13) Moshi (#=18) Taveta-Holili (#=45) Busia (#=90) Trade across the border Maize Oil Rice Sugar 020 40 60 80 Arusha (#=13) Moshi (#=18) Taveta-Holili (#=45) Busia (#=90) Main sources of supplies Importers Retailers Wholesalers Middlemen Farmers 21 Table 2 Typical costs incurred when trading across the border (KES) Border post Country Number of businesses Customs duties Local municipal council charges Transport costs Other costs Busia Kenya 40 48 98 100 0 Uganda 50 72 14 60 14 Taveta– Holili Kenya 41 71 46 44 2 Tanzania 31 81 55 81 13 Source: Business survey Table 3 below presents a detailed breakdown of the costs from the border to the final destination, comparing the prices and costs of maize between 2015 and 2022 in Mombasa, Kenya. The first two columns report on maize originating through Busia while columns 3 and 4 show maize originating through Taveta–Holili. Generally, the prices of maize in the Busia and Taveta–Holili markets are determined by demand and supply on the Kenyan side. For example, maize prices decrease if there is a bumper harvest in Kenya. Transporters: Transport costs include vehicle operating costs (direct costs to operate a given vehicle, notably maintenance, tires, fuel, labour and capital costs) and indirect costs such as licences, insurance and road toll and roadblock payments. The relationship between transport prices and transport costs depends partly on the transport sector market structure: the less competitive the transport sector, the more likely there will be a divergence between transport prices and costs. The price of transporting maize is affected by demand and supply. For example, during the high season (December for Busia, Uganda), transport providers flock to Busia town; therefore, transporting bags of maize to Kenyan cities becomes cheaper because of the oversupply of transport options. This, in turn, affects the commodity’s price in the cities. At the time of the survey (2022), transporting one bag of maize from Busia to Mombasa cost KES 400, although the price may come down to KES 350 per 90 kg bag during the high season. However, the transport price (KES 400) is what the trader in Mombasa pays, which in most cases is different from the price the transporter charges. This is because brokers negotiate the transportation of the goods, and they charge a small fee, typically between KES 20 and 50 per bag or more, depending on the truck size and transport availability. In 2015, the transport cost was KES 240. Therefore, transport prices in 2022 were up by 66.7% from 2015. Note that the transport prices shown in the table include only charges from Busia to Mombasa. 4 In 4 Transport prices would be a great deal higher and be the significant cost in determining the price of maize if we aggregated transport charges, starting from obtaining maize from farmers. This is because it passes through several transportation stages before reaching the border, also requiring loading and offloading at each stop. 22 aggregate, the share of transport in the wholesale price at Mombasa declined from 9.87% in 2015 to 6.87% in 2022. In Taveta, Kenya, transporters cited fuel prices as the determining factor in transport costs. The cost of transporting maize from Taveta to Mombasa at the time of the study (2022) was KES 300 per 90 kg bag, while it was KES 200 in 2015. 5 For this maize, the share of transport in the wholesale price at Mombasa declined from 8.1% in 2015 to 5.5% in 2022. Labour costs: The labour offloading cost constituted 0.5% of the wholesale price in Mombasa in 2022, down from 0.8% in 2015. Local tax costs: One of the main local taxes for maize (and other agricultural produce) is a council/local cess, charged by the municipal council for every product accessing a wholesale/retail market. Table 3 shows that the local cess constituted 0.5% of costs in 2022, up from 0.4% in 2015. OSBP costs: In 2022, the phytosanitary and clearance charges at the border were KES 140 per bag (2.4%). In 2015, there were no phytosanitary charges, and the fumigation charges were KES 300 per truck (a truck carries 300 bags). Thus, the OSBP equivalent cost was KES 1 per bag. Other costs: In Taveta, Kenya, maize of different qualities is sometimes mixed and sold at a higher price than it should. If the Kenya Bureau of Standards notes the inconsistency, the maize must undergo grading, incurring additional costs and time. Table 3 shows the wholesale and selling price of maize, which includes costs and profit margin. In 2022, a bag of maize from Busia would arrive at the wholesale market in Mombasa at KES 5,820 per bag with a selling price of KES 6,390. That is, maize from Busia would sell in Mombasa at KES 71 per kg in 2022. In 2015, the wholesale price of maize from Busia was KES 27 per kg, with a selling price of KES 33 per kg. Between 2022 and 2015, then, the price of maize from Busia increased by 142%, while the wholesale and retail prices in Mombasa increased by 139% and 115%, respectively. Similarly, the price of maize from Taveta–Holili had increased by 120% in 2022 relative to 2015. In addition, the wholesale and retail prices for this maize in Mombasa had increased by 119.5% and 109.6%, respectively. 6 5 If a transporter spends more time at the border, the charges for transportation per bag remain the same. 6 The Kongowea wholesale market in Mombasa had maize from Mpeketoni in Lamu county (north coast). Maize arrived at a cost of KES 5,300 per bag, or KES 58 per kg, to the trader, and was sold at KES 65–70 per kg. The maize from Busia, at KES 5,820, was, therefore, too expensive to sell in Kongowea market. 23 Table 3 Costs and prices of maize in the market (KES) Busia Taveta–Holili Kampala 2022 2015 2022 2015 2022 2015 Maize price per kg 58 24 55 25 64 25.6 90 kg bag 5,220 2,160 4,950 2,250 6,400 2,560 OSBP 140 1 140 1 0 0 Transport Mombasa 400 240 300 200 320 96 Labour (offloading) 30 20 30 20 32 16 Market cess 30 10 30 10 32 Total amount per bag 5,820 2,431 5,450 2,481 6,784 2,672 Current cost price in Mombasa 5,820 2,431 5,450 2,481 6,784 2,672 Cost price per kg in Mombasa 65 27 60.6 27.6 67.84 26.72 Selling price per kg 71 33 65 31 70.4 30.4 Selling price per bag 6,390 2,970 5,850 2,790 6,336 2,736 Note: Authors’ calculations. Assessing the impact of OSBPs on maize prices A reduction in cross-border trade costs, such as through the introduction of OSBPs, provides consumers with several benefits, including lower prices, higher quality and more product variety. Specifically, as we have seen, the impact on consumer welfare depends on the prices charged by domestic and foreign exporting firms. This depends on at least two main channels: the competition channel and the cost channel. Ideally, to examine whether a reduction in trade costs associated with the introduction of OSBPs is passed through to producers and consumers in the form of lower prices needs to control for other factors that may have affected the prices of the products in both the domestic and the international market. For example, an increase in fuel prices could affect the price of transport, and an increase in fertiliser prices could raise the cost of production of maize. These and other factors make it harder to pin down exactly how much the efficiency gain actually results in lower prices to consumers and welfare gains. This may need a carefully crafted econometrics analysis. Nevertheless, we show prima facie evidence of the impact of OSBPs on prices and its transmission across actors involved in the maize supply chain using the survey data. 7 To do this, we assume that the selling prices observed in 2022 reflect both the pro-competitive effects of OSBPs and the actual reduction in the costs of cross-border trade observed because of OSBPs. Therefore, the approach followed to analyse the impact of OSBPs on consumer prices is to project what the selling prices would have been in 2022 had the cost of trading across the border and the profit rate remained the same as in 2015. That is, for the trade costs assumption, we use the share of trade costs in 2015 as a reference point; for the analysis of the pro- 7 We should note that several factors affect the price of the products and other costs involved in cross-border trade, some domestic and some international, such as fuel prices. 24 competitive effects, we use the profit rate in 2015 as a base. Based on the approach mentioned above, we present three scenarios to quantify the impact of OSBPs on consumer prices of maize. The baseline scenario is essentially the prices and costs observed in 2022 – that is, the selling price reflects reductions in both the trade cost and the pro-competitive effects (see Scenario 3: No gain from competition channel but gains from lower trade cost. In the third scenario, we assume the introduction of OSBPs has reduced cross-border trade costs but has not resulted in a pro-competitive induced reduction in consumer prices, so that the profit rate for business traders in 2022 remains the same as in 2015. This means that the profit rate in 2022 is 22.2% when maize is sold to consumers, instead of the observed 9.8% in 2022. In other words, we assume that the reduction in the profit rate is because of the introduction of OSBPs, while several factors affect this profit rate. In this case, the result we find would be the upper-bound estimates of the impact of OSBPs on consumer prices. We find that, if the profit rate for 2022 had stayed at the same rate as in 2015, but the cost reduction benefits of OSBPs had been maintained, then the selling prices to consumers in 2022 would have been KES 7,110.4. If we compare this with the 2022 selling prices (KES 6,390), the consumer would have saved KES 720 for maize originating from Busia, Uganda. This is equal to a saving of 10% for consumers. This is, of course, an upper-bound estimate, and we are attributing all of the fall in the profit rate or markup to OSBPs. In a similar analysis, we find savings for consumers from maize coming through Taveta–Holili of 4.5%. Table 4 for a summary of the results). Scenario 1 (baseline scenario): No gain from competition channel and trade cost channel. In the baseline scenario, we consider what the selling prices of maize would have been had there not been an OSBP. This means the trade cost and profit rate remain the same as in 2015. In this case, in 2022, the selling price of maize would be KES 7,178. If we compare this with the actual selling price of 2022, then consumers saved KES 788. This is equal to a 12.3% saving to consumers. The gain from Holili would have been 6.8%. For maize going from Taveta–Holili to Mombasa, when we compare prices in 2015 and those in 2022, the costs involved cover 9.2% of the selling price in 2022; the figure for 2015 is 6.4%. If we assume that this owes simply to the trade facilitation efforts of the OSBP, then we can assume that it resulted in a 0.2 percentage drop in the prices of goods transferred. This gain emanates from the transport sector’s unchanged prices over the time period, probably as a result of the efficiency gained at the OSBP. Scenario 2: Gain from competition channel but no gains from lower trade cost. In the second scenario, there is a gain from the competition effect but none from a cost reduction. That is, the profit rate in 2022 is as observed in 2022 but the trade cost is as in 2015. In this case, we estimate what the selling prices of maize to consumers in 2022 would have been had the share of the cost of cross-border trade remained the same as before (2015) but the profit rate is at 9.8%. In this scenario, we find that the selling price to consumers in 2022 would have been KES 6,450. If we compare this with 31 4 OSBP impact on household income Introduction This section aims to provide an overview of the impact of the reduction in transport costs observed as a result of the introduction of OSBPs on household welfare. Theoretical and empirical evidence shows that trade can influence household income through several channels, including relative prices, spurring high economic growth, providing macroeconomic stability and increasing government revenue, which can be spent on poverty reduction (Winters et al., 2004). Trade can benefit the poor by reducing the prices of the goods they consume and also by creating overseas market opportunities for the products they produce. 11 However, lack of competition in the market and high transportation and logistics costs could prevent the gains of trade from being passed to poor households and consumers. Trade facilitation efforts such as OSBPs reduce trade costs. Precisely quantifying the impact of OSBPs on households and welfare requires complicated econometric modelling and extensive data, as there are several factors other than OSBPs that have simultaneously affected poverty since the inception of the border posts. This section uses Kenya’s household survey to provide evidence of OSBP impacts on household income based primarily on the assumption that a reduction in consumer prices induced by lower transport costs and trading costs through reduced clearance time and simplification of border procedures at OSBPs allows us to calculate the monthly food expenditure per adult equivalent households have saved. The introduction of OSBPs could affect household income and poverty through the impact on trade costs, especially through lower transportation costs. This channel needs the transmission of lower transport costs to lower transport prices and finally to lower consumer prices. We showed this channel in Section 3. This approach provides a rough and upper-bound estimate, and the results should be interpreted carefully. OSBPs could also affect the welfare of households living at the border through direct employment or by providing goods and services to OSBP users. To complement the first assessment based on consumer prices and to measure OSBPs’ direct impact in border towns, data were collected from 315 households in the two border towns of Taveta–Holili and Busia. The survey covered the most common goods coming from the neighbouring 11 If rural households are net producers of the goods that saw a fall in prices as a result of trade liberalisation they could see a fall in income. 32 country across the border, the reasons households buy commodities, whether they or their household members are involved in OSBP activities directly or indirectly and finally to what extent the existence of OSBPs has affected households in terms of jobs, incomes, new business, rent and prices. Section 4.2 provides estimates of OSBP transport cost reduction impacts on household income. Section 4.3 presents results from the survey to assess the impact of the OSBPs on households. Trade costs, consumer prices and household income The approach followed is that the reduction in consumer prices induced by lower transport costs through OSBPs allows us to calculate the monthly food expenditure per adult equivalent households have saved. We use the Kenya Integrated Household Budget Survey for 2015/16. For rural households, food and non-food expenditure per adult equivalent were KES 3,447 and KES 1,879 in 2015/16, respectively, while for core urban households they were KES 5,550 and KES 6,349. In other words, rural households spend 64.7% of their income on food, significantly more than the share of households in core urban areas, which spend 46.6%. Table 8 shows the results. For example, it shows various maize products, with the share of each in the household expenditure basket. By multiplying the product’s weight in the basket by the average monthly food expenditure per adult equivalent household member, we can obtain the exact amount the household is spending on the product. For example, rural consumers spend KES 565 on loose maize flour (i.e., 0.164*3447 KES=565). Table 8 shows the amount households have saved under different scenarios since the fall in consumer prices owing to lower transportation costs. The three scenarios are from Section 3.3. For example, in scenario 1, where we assumed the total absence of OSBPs (i.e., no gain from the competition channel and the trade cost channel), we find that the typical rural household might have saved KES 70 per month. If we aggregate all kinds of maize and maize products, Table 8 shows that rural households saved KES 68–93 in monthly adult equivalent terms from Busia and KES 34–68 from Taveta–Holili. Core urban households saved KES 64–87 per month. The household gain from rice is lower than that from maize for both rural and urban households. This indicates a potentially large impact of the OSBPs on Kenyan households. Consider that maize and rice are only two out of over 200 products consumed by Kenyan households, albeit very important ones, standing at 25% of the total consumption basket for rural households and 22% for urban households. So, if we consider that the typical rural household saved around KES 80 per month (an average of the most and least impactful scenarios) on 25% of its food basket, we can hypothesise that the total monthly saving will be of around KES 320. In a year, this translates to a total saving of KES 3,840 per rural household, higher than the cost of food for one month. In other words, the introduction of OSBPs 33 could potentially be saving some rural Kenyan families one month of food expenditure on maize every year. This applies only to food items, but similar considerations could be applied to non-food items, making the savings generated by OSBPs potentially even higher. Table 8 OSBP impacts on household spending (KES) Food item Disaggregated food item Household survey Busia Taveta–Holili Share in basket Consumer expenditure (food) S1 S2 S3 S1 S2 S3 Rural Maize Loose maize flour 0.164 565 51 70 57 51 28 25 Loose maize grain 0.038 131 12 16 13 12 6 6 Green maize 0.009 31 3 4 3 3 2 1 Loose green maize 0.008 28 2 3 3 2 1 1 0.219 755 68 93 76 68 37 34 Rice Non-aromatic white rice 0.029 100 4 4 4 Broken white rice 0.017 59 2 2 2 0.046 159 14 20 16 Urban Maize Loose maize flour 0.064 355 32 44 36 24 17 16 Loose maize grain 0.022 122 11 15 12 8 6 5 Fortified maize flour 0.021 117 10 14 12 8 6 5 Sifted maize flour 0.021 117 10 14 12 8 6 5 0.128 710 64 87 72 48 35 32 Rice Non-aromatic white rice 0.049 272 11 11 10 Broken white rice 0.024 133 12 16 13 Aromatic white rice 0.008 44 4 5 4 Brown rice 0.008 44 4 5 4 0.089 494 Note: S1 refers to scenario 1: No gain from competition channel and trade cost channel; S2 refers to scenario 2: Gain from competition channel but no gains from lower trade cost; S3 refers to scenario 3: No gain from competition channel but gains from lower trade cost. As Section 3 showed, though, we should interpret these results carefully. These are upper-bound estimates that make several critical assumptions. Actual savings by households from OSBPs could be lower. Moreover, we are generalising our findings from only two products to the full food consumption basket of Kenyan households: the price reductions for other products may be different (higher or lower) than those found for maize and rice. Furthermore, the impact of lower consumer prices as a result of lower trade costs owing to OSBPs depends on whether the household is a net consumer or a producer. This is especially important when considering rural households’ welfare. 34 Impact of OSBPs on household welfare based on the survey data Section 4.2 estimated how much households had saved as a result of OSBPs based on the reduction in consumer prices. To complement this assessment, and also to consider the welfare of households living on the border through direct employment in the project or by providing goods and services to the users of OSBPs, data were collected from 315 households in the border towns of Taveta–Holili and Busia (see Section 2.3 and Appendix 3). OSBPs create direct and indirect employment. In aggregate, 40% of households responded that either they or a member of their household was involved in OSBP activities, either working directly at the OSBP (5%) or providing services or goods to OSBPs (10.5%) or as users of OSBPs and truck drivers (27%). In Busia, the majority of households provide goods or services to users of the OSBP as well as truck drivers (see Figure 7). In Taveta, Kenya, however, only 10% of households are working on OSBP activities; the share is higher on the Tanzanian side (Holili): close to 70% of respondents here said household members were involved in OSBP activities (Holili is close to the OSBP, whereas Taveta, on the Kenyan side, is far from it). In Holili, 40% of respondents provide goods or services to the OSBP or truck drivers. Figure 7 Activities households are involved in at OSBPs (%) Source: Household survey With respect to incomes from activities associated with OSBPs, at both border posts most households (61%) said that incomes had increased since the OSBP had been introduced, while 21% responded that incomes had decreased (see Error! Not a valid bookmark self-reference.). Among those who pointed to an increase, 97% said that income had increased by more than 50%, while the rest 3% said it had been by less than 50%. 050 100 Other Provide services or goods to lorry drivers Provides services or goods to the OSBP Work directly at the OSBP Provide services or goods to users of the… None Busia Uganda Kenya 050 100 Other Provide services or goods to lorry drivers Provides services or goods to the OSBP Work directly at the OSBP Provide services or goods to users of the OSBP None Taveta-Holili Tanzania Kenya 35 Twenty-one percent responded that incomes had remained the same or that they could not tell. Looking at households’ responses disaggregated by border post/country, there is a substantial difference between Kenya– Uganda at the Busia border post and Kenya–Tanzania at the Taveta–Holili border post. In Busia, 84% of Kenyan households said that incomes had increased since the OSBP, while only 56% of Ugandans said the same – a 28 percentage point difference. The same is observed at Taveta–Holili: 69% of Kenyans said incomes had increased while only 44% of Tanzanians said the same. Table 9 Incomes since the OSBP has been introduced (%) How has income associated with these activities evolved since the OSBP was introduced? Busia Taveta–Holili Aggregate Kenya Uganda Total Kenya Tanzania Total Decreased 2.3 17.3 11.8 26.3 28.8 27.5 20.8 I cannot tell 11.4 17.3 15.1 2.5 16.3 9.4 11.8 Increased 84.1 56 66.4 68.8 43.8 56.3 60.6 Remained the same 2.3 9.3 6.7 2.5 11.3 6.9 6.8 Source: Household survey Households were asked what had changed after the establishment of the OSBPs. Table 10 reports the response in aggregate. A quarter said that there were more jobs and new businesses while a third reported that incomes had increased. In the same period, prices and rent have also increased, as has vehicle traffic. Table 10 Perceived effects of the existence of OSBPs (%) Increased Reduced No change I cannot tell Jobs 24.8 10.5 1.0 1.3 Incomes 33.0 17.5 1.0 1.6 New businesses 24.4 8.9 0.0 2.5 Prices 38.1 4.8 0.0 2.5 Vehicle traffic 28.6 0.3 0.6 0.3 Rent 36.2 0.3 0.3 1.0 Source: Household survey Figure 8 reports on the same question, disaggregated by country/border post. Most households in Kenya indicated that rent, the price of goods and services and vehicle traffic had increased; fewer households responded thus in Tanzania. In aggregate, at both border posts 87% of households reported that the impact of the OSBP on the economy of the area had been positive. Only 6% reported a negative impact, while 7% said there had been no impact or they were not sure. Figure 9 shows responses by border post. The majority 36 of households in Busia (90%) and Taveta–Holili (83%) indicated that the OSBP had had a positive impact on the economy of the border town. However, in Taveta–Holili, more Kenyans (96%) than Tanzanians (70%) said that the impact of the OSBP had been broadly positive. Figure 8 OSBPs and their effects (%) Source: Household survey Figure 9 Impact of OSBP on local economy (%) Source: Household survey 050 100 Jobs Incomes New Businesses Prices Vehicle Traffic Rent Busia Kenya Decreased Increased 050 100 New Businesses Vehicle Traffic Jobs Rent Incomes Prices Busia Uganda Decreased Increased 050 100 Jobs Incomes Vehicle Traffic New Businesses Rent Prices Taveta-Kenya Decreased Increased 050 100 New Businesses Vehicle Traffic Rent Jobs Prices Incomes Holili-Tanzania Decreased Increased 020 40 60 80 100 Negative No impact Not sure Positive Busia Uganda Kenya 050 100 Negative No impact Not sure Positive Taveta-Holili Tanzania Kenya 37 5 Conclusions and recommendations This report has examined how the impact of trade cost reductions through the introduction of OSBPs has been transmitted to the rest of the economy. For this purpose, we selected two products based on two criteria: their weight in consumer baskets and their tradability across the EAC. For maize, we found the following under different scenarios: the gain for consumers in terms of lower prices could range from 9% to 12.3% for maize originating from Busia and from 4.5% to 6.8% for maize from Taveta–Holili. Using the same approach, we find similar results for rice but with a lower gain magnitude. These findings are partially supported by the survey results: half of business traders responded that OSBPs had reduced the prices of commodities they traded across the border. We also provide some evidence on the impact of OSBPs on household income, using the fall in consumer prices and the Kenya Integrated Household Budget Survey for 2015/16. We find that, for maize, rural households have saved between KES 68 and KES 76 in monthly adult equivalent terms for maize from Busia and between KES 34 and KES 68 for maize from Taveta–Holili. Core urban households have saved KES 64–87 per month on rice and maize. The household gain from rice is lower for both rural and urban households compared with maize. Generally, the introduction of OSBPs could potentially be saving some rural Kenyan families one month of food expenditure every year. To complement the above assessment of the OSBPs and household income, we surveyed households living at the border. A total of 40% of households said that they or a member of their household was involved in OSBP activities and, at both border posts, most households (61%) said incomes from activities associated with the OSBPs had increased. In aggregate, a third of the respondents said that, since the OSBP had been introduced, income had increased, and a quarter said there were more jobs and new businesses. In aggregate, at both border posts, 87% of households reported that the impact of the OSBP on the economy of the area had been positive. Only 6% reported a negative impact, while 7% said no impact or were unsure. 38 Finally, we asked business traders if they had any recommendations on the OSBPs. These are listed below: • Allow small traders with small luggage to pass through the customs yard: traders said that they had to pass through the informal route and there was a lot to pay there (this recommendation was very common). • Consider motorcycle and bicycle users and reserve them a route that is safe from traffic. • Address traffic jams. • Reduce bribery. • Reduce custom duties. • Improve services. • Reduce taxes (this recommendation was very common). 39 References Ahmed, M. (2012) ‘Analysis of incentives and disincentives for maize in Uganda’. MAFAP Technical Note. Rome: FAO. Amiti, M., Itskhoki, O. and Konings, J. (2014) ‘Importers, exporters, and exchange rate disconnect’ American Economic Review 104(7): 1942–1978. Anderson, J.E. and Van Wincoop, E. (2004) ‘Trade costs’ Journal of Economic Literature 42(3): 691–751. Balat, J., Brambilla, I. and Porto, G. (2009) ‘Realizing the gains from trade: Export crops, marketing costs, and poverty’ Journal of International Economics 78(1): 21–31. Berman, N., Martin, P. and Mayer, T. (2012) ‘How do different exporters react to exchange rate changes?’ The Quarterly Journal of Economics 127(1): 437–492. Bhagwati, J.N. and Srinivasan, T.N. (1980) ‘Revenue seeking: A generalization of the theory of tariffs’ Journal of Political Economy 88(6): 1069–1087. Breinlich, H., Dhingra, S., Sampson, T. and Van Reenan, J. (2016) ‘Who bears the pain? How the costs of Brexit would be distributed across income groups’. Brexit Analysis Paper 7. London: CEP, LSE. Campa, J.M. and Goldberg, L.S. (2005) ‘Exchange rate pass-through into import prices’ Review of Economics and Statistics 87(4): 679–690. Daly, J., Hamrick, D., Gereffi, G. and Guinn, A. (2017) ‘Maize value chains in East Africa’. Policy Brief, February. London: IGC. De Loecker, J., Goldberg, P.K., Khandelwal, A.K. and Pavcnik, N. (2016) ‘Prices, markups, and trade reform’ Econometrica 84(2): 445–510. Edmond, C., Midrigan, V. and Xu, D.Y. (2015) ‘Competition, markups, and the gains from international trade’ American Economic Review 105(10): 3183–3221. Fitzgerald, D. and Haller, S. (2014) ‘Pricing-to-market: evidence from plant-level prices’ Review of Economic Studies 81(2): 761–786. Fontagné, L., Martin, P. and Orefice, G. (2018) ‘The international elasticity puzzle is worse than you think’ Journal of International Economics 115: 115–129. Mendez-Parra, M. and Calabrese, L. (2023) ‘One-stop border posts in East Africa: Impact on transport costs and issues for further analysis’. London: ODI. PMA – Plan for the Modernisation of Agriculture – Secretariat (2009). ‘Maize value chain study in Busoga subregion’. Kampala: PMA Secretariat. Porto, G.G. (2005) ‘Informal export barriers and poverty’ Journal of international Economics 66(2): 447–470. Siu, J. (2021) ‘Essays on informality and economic development’. Doctoral dissertation, University of Birmingham. 40 USAID – United States Agency for International Development (2010) ‘Market assessment and baseline study of staple foods’. Country Report Uganda. Kampala: USAID. Winters, L.A., McCulloch, N. and McKay, A. (2004) ‘Trade liberalization and poverty: The evidence so far’. Journal of Economic Literature 42(1): 72–115. World Bank (2009) ‘Eastern Africa: A study of the regional maize market and marketing costs’. Report 49831–AFR. Washington, DC: Agriculture and Rural Development Unit, World Bank. WTO – World Trade Organization (2005). World trade report 2005: Exploring the links between trade, standards and the WTO. Geneva: WTO. 47 11. What has been the impact of this OSBP broadly on the economy of this area? Impact Put mark Positive Negative No impact 12. Are you or any member of the household involved in the following activities? Work directly at the OSBP (e.g. customs/migration officer) Provide services or goods to the OSBP (e.g. cleaner) Provide services or goods to users of the OSBP (e.g. custom clearing agent/forwarder) Provide services or goods to lorry drivers (e.g. selling food) Other 13. How has the income associated with these activities evolved since the OSBP was introduced? Increased Decreased Remained unchanged I cannot tell Greater than 50% Less than 50% Greater than 50% Less than 50% 14. To what extent has the existence of the OSBP affected the following aspects in this area? Increased Reduced No change I cannot tell Jobs Incomes New businesses Prices Vehicle traffic Rent 15. Any other comments you may have on OSBPs? 48 Appendix 3 Household demographics and other characteristics In aggregate, 46% of respondents in our survey have primary school as the highest level of education, 17% college or university and 31% secondary school; only 5% are not educated or attended informal education. The majority of respondents in Busia (63%) and Taveta–Holili (64%) are selfemployed. In Busia, 68% of Kenyans are self-employed compared with 57% of Ugandan households. In Taveta–Holili, 69% of Kenyans are selfemployed compared with 59% of Tanzanians. At both borders, only 9% of respondents are employed full time; 2% are not employed. The average number of people currently living in the same house or homestead is six; this is the same in Busia and Taveta–Holili and across the borders. Figure A4.1 shows the most common goods households consume that come from the neighbouring country across the border. In aggregate, horticulture (fruits, vegetables), cereals and grains, and new clothes are the top three items. At both border posts, cereals and grains, and horticulture are the top two items; this is followed by new clothing for Busia and secondhand clothing for Taveta–Holili. While there is a difference at the border posts, there is a significant overlap in the top five items, albeit with a different order. The lower half of the figure shows the most commonly consumed goods by households from across the border disaggregated by border post and country. For example, at the Busia border post, the top four most common items Kenyans buy from Uganda are cereals and grains, new clothes, horticulture, and fish and fish products. At the same border post, the top four most common items for Ugandans are household consumables (flour, cooking oil, salt), second-hand clothes, others and electronics. At Taveta–Holili, for Kenyans, cereal and grains, horticulture, second-hand clothes, and poultry and poultry products are the top items they buy from Tanzania; for Tanzanians, these items are new clothing, horticulture, household consumables and second-hand clothes. Furthermore, we find that more Kenyans from Taveta buy items from Holili in Tanzania than Tanzanians who buy from the Kenyan side. In Busia, 99% of households reported that a member of their household crossed the border to buy goods from the other side of the border. In Taveta–Holili, the figure is 93%. Figure A4.2 shows the reasons households buy from the neighbouring country. In aggregate, affordable prices and availability in the vicinity are the top reasons. By border post, in Busia, Kenyan and Ugandan households buy goods from across the border mainly 49 because of the affordable prices. Good quality products are the second most important reason for Kenyans, while availability in the vicinity is the second reason for Ugandans. In Taveta–Holili, Kenyan and Tanzanian households buy goods across the border for reasons to do with affordability and stable supply. In general, affordable prices are one the main reasons at both border posts, for households from the three countries. Figure A4.1 Most commonly consumed goods from across the border (%) Source: Household survey 010 20 30 40 50 60 Basic farm equipment Household furniture Meat products Electronics Others Milk & dairy products Poultry & Poultry products Fish & fish products Household consumables Second hand clothes New cloths Cereals & grains Horticulture A. Aggregate 050 100 Basic farm equipment Household furniture Meat products Electronics Others Milk & dairy products Poultry & Poultry… Fish & fish products Household… Second hand clothes New cloths Cereals & grains Horticulture B. By border post Taveta-Holili Busia 50 Figure A4.2 Reasons household buy commodities from the neighbouring country (%) Source: Household survey 020 40 60 80 100 Other Stable supply Good quality Available near this vicinity Affordable prices Aggregate 050 100 150 Other Stable supply Good quality Available near this… Affordable prices By border post Taveta-Holili Busia 050 100 150 Affordable prices Good quality Stable supply Available near this vicinity Other Busia Uganda Kenya 0100 200 Affordable prices Good quality Stable supply Available near this… Other Taveta–Holili Uganda Kenya