WATER KNOWLEDGE NOTE Pro-Poor Groundwater Development The Case of the Barind Experiment in Bangladesh Partha Sarathi Banerjee1 and Sanjiv De Silva2 The Barind region, a water-stressed area in northwest Bangladesh, had an underdeveloped agricultural economy and high levels of poverty until two projects revitalized the area with enhanced groundwater irrigation. The Barind Integrated Area Development Project in 1985 and Barind Multipurpose Development Authority (BMDA) in 1992 used new water extraction technology and innovative management practices such as deep tubewells (DTWs) fitted with smart card–operated electric pumps to develop drought-resilient irrigation. Both projects have helped the Barind region reduce poverty and achieve self-sufficiency in rice. However, there are concerns about declining groundwater levels in the Barind and nearby regions, resulting in a temporary halt in DTW expansion. Preliminary evidence presented in this case study suggests farmers served by shallow tubewells (STWs) may be losing access to groundwater in some parts of the Barind region, which can have significant development implications because these tubewells remain the predominant source of irrigation. This evidence provides grounds to question whether an irrigation model reliant on DTWs is sustainable and equitable in the long term. Further research is needed to better establish groundwater conditions and understand the risk to STW users to inform future policy on DTW-driven agricultural development. © Sanjiv De Silva / IWMI The information in this case study was prepared as part of the South Asia Water Initiative (SAWI) technical assistance project Managing Groundwater for Drought Resilience in South Asia. SAWI is a multi-donor trust fund supported by the United Kingdom, Australia, and Norway and administered by the World Bank. Background drought-prone region where rain-fed monocropping was the The United Nations’ Millennium Development Goals and norm, the BMDA introduced inverted DTWs and a prepaid the Bangladesh government’s poverty reduction strategy smart metering system that has transformed agriculture into emphasize that “raising land productivity and increasing a “three crop revolution” (Kang 2013). Cropping intensity diversification of agriculture production” are “fundamental has risen from 117 percent to 200 percent, higher than to poverty reduction as well as for food security” in the the national average of 175 percent (Jahan et al. 2010b). country (IMF 2013). In this context, the positive impacts The results have helped reduce poverty in the region of the BMDA on agriculture, especially rice production, has from 57  percent in 2000 to 36 percent in 2010 (Jolliffe been rightly regarded as a success story. This autonomous et al. 2013). The Barind region is characterized by diverse body under the Ministry of Agriculture, presently working agroecological conditions. Aside from the Padma River, all in 16 districts of Bangladesh (map 1), seeks to enhance food other rivers are seasonal and dry up during the dry season security and living standards mainly through agricultural (Jahan et al. 2010a). Precipitation is diverse, erratic, and low development. Its multipronged resource management with an average 1,600 millimeters compared to the national model is a rare example of integrated development. In this average of 2,550 millimeters and sometimes affected single- MAP 1. Districts Covered by the Barind Multipurpose Development Authority Source: Barind Multipurpose Development Authority. WATER GLOBAL PRACTICE CASE STUDY | PRO-POOR GROUNDWATER DEVELOPMENT 2 crop production before irrigation was available. The Rajshahi Agriculture by Marginal and Tenant Farmers in the Eastern district receives low rainfall, whereas Thakurgaon and Gangetic Plains (de Silva and Leder 2017), funded by the Rangpur lie in the 1,500- to 2,000-millimeter and 2,000- Australian Center for International Agricultural Research. to 2,500-millimeter belts, respectively. Runoff is high in Combined, this research covers six districts (Chapai, some areas with limited groundwater storage, which makes Natore, Nowabganj, Rajshahi, Rangpur, and Thakurgaon) it prone to drought (Ahmeduzzaman, Kar, and Asad 2012). and includes discussions with officials from BMDA and the Department of Agriculture, focus group discussions Although this case study profiles the clearly impressive and interviews with farmers from communities with and innovations regarding electric-powered DTWs, it also without a DTW (Rangpur and Thakurgaon), and review poses the question of whether DTW-driven rice crop of available secondary data and literature on the Barind intensification is sustainable in parts of the region. A limited region.3 number of interviews in communities using STWs, mainly in the Thakurgaon district, suggests that though farmers Innovations around DTWs within DTW command areas are thriving, farmers outside these areas may be struggling with enhanced groundwater DTWs were introduced by the Bangladesh Water drawdown in the dry season because maximum STW depths Development Board (BWDB) in 1963–66 and were are typically only 15 to 20 meters (de Silva and Leder 2016). originally diesel-powered. In about 1990, the energy With STW-dependent farming communities constituting source powering DTWs changed to electricity because of most of the region’s population, this case study argues that the unreliability of diesel DTWs and resulting crop losses. the viability of DTWs must consider the externalities that The BWDB handed over supervision of DTWs in northern they may induce. Bangladesh to the BMDA in 2001–02. Framing this discussion on sustainability of DTWs Inverted DTWs in certain parts of the Barind region is what could be described as the rice-groundwater nexus in Bangladesh, a The primary aquifer in the Barind region accessed by dominant development paradigm driving investments in DTWs and STWs alike is the Dupi Tila formation the agriculture and irrigation sectors. Bangladesh is now the composed of deltaic sands, silts, and gravels overlain by as world’s fourth-largest rice producer and nearly self-sufficient much as 30 meters of Barind clay. Recharge to the aquifer is in rice (Mainuddin and Kirby 2015). Rice accounts for nearly limited by vertical leakage through a low-permeability clay 80 percent of the country’s gross cropped area (Amarasinghe layer (Ahmed and Burgess 1995). The BMDA engineers et al. 2014), and rice production has almost tripled since designed an inverted DTW because the conventional 1976–77 (Mainuddin et al. 2014). Given the scarcity of DTW technology was not suitable to lift groundwater land, the production increase was driven by yields that rose in the Barind region. The inverted DTWs, described by to almost 4.3 tonnes per hectare from 2.7 tonnes per hectare Asaduzzaman and Rushton (2006), can extract water from in 1995. Bangladesh’s rice area under irrigation has increased relatively “poor” aquifers with limited saturated thickness to 77 percent from about 30 percent in 1995 (Amarasinghe that exist at a depth of approximately 30 to 40 meters. et al. 2014), and the country increased use of modern Of the 25 upazilas (subdistricts) in the Barind region, 10 varieties during the same period. The rapid expansion of the have poor aquifers, where normal DTWs cannot function. dry-season rice crop has especially contributed to enhanced According to the BMDA, 15,813 DTWs currently irrigate production. Today, 80 percent of boro rice is groundwater 496,198 hectares of boro paddy in the Barind region. irrigated (Mainuddin et al. 2014) and accounts for almost DTW command areas vary in size. In Dhondogaon village, all the irrigation-consumptive water use of rice. These Thakurgaon district, it can reach 24 hectares for rice, results have improved the livelihoods of millions of rural though this measurement varies depending on the specific poor (Qureshi, Ahmed, and Krupnik 2014) and reduced water demand of selected crops. rural poverty (Gautam et al. 2016). The Prepaid Metering System Informing this case study is fieldwork undertaken specifically for this study as well as primary data collected The BMDA initially collected irrigation fees from farmers by the International Water Management Institute in 2016 by issuing paper coupons through dealers. This was a under the project Improving Water Use for Dry Season cumbersome process with loopholes for corrupt practices. WATER GLOBAL PRACTICE CASE STUDY | PRO-POOR GROUNDWATER DEVELOPMENT 3 There was scope, for instance, for operators to run a pump Water Savings and Conflict Avoidance without receiving any coupon and to illicitly sell water Using Underground Pipelines for to farmers without the BMDA’s knowledge. In 2005–06, Distribution prepaid metering was introduced, and by 2010, almost all pumps were managed through the new system, in which Underground pipelines are being put in place in all each farmer in the command area buys a smart card from DTW command areas to avoid substantial loss of the BMDA’s upazila office at a nominal cost. The BMDA water during distribution as a result of seepage and selects a dealer for each upazila to recharge the smart cards, evaporation and to minimize the need for farmers to who is provided a mobile vending unit (MVU) at a cost of have water channels on their land. Maintenance of Tk 30,000 to 60,000 (US$360 to $720). The BMDA pays concrete channels has been costly because of frequent the dealer a commission of 2.5 percent on the amount of breakage. Construction included overhead water tanks recharge. Farmers recharge their respective smart cards with at DTWs for irrigation and supply lines to distribute a dealer and approach the DTW operators to buy water. drinking water when supplies of the latter are scarce in When a farmer requires irrigation, the farmer-elected DTW the summer months. operator enters the farmer’s smart card in a prepaid meter attached to the DTW, and water is supplied for irrigation. DTW-Enabled Transformation of The pump stops automatically after the scheduled duration Agriculture in the Barind Area requested by the farmer, and the prepaid meter deducts the necessary amount from the farmer’s smart card (photo 1). Electric-powered DTWs have enabled farmers in the Irrigation charges are fixed depending on the discharge rate Barind region to largely overcome the risks of crop failure of the pumps and vary between Tk 75 and 115 (US$0.9 to and drought-related crises. DTWs help farmers stabilize US$1.40) crop production during midseason dry spells and carry over groundwater storage during multiyear monsoon failures, This system collects 100 percent of irrigation charges for ensuring minimum crop losses. Farmers interviewed during the BMDA, making it a surplus-generating, self-sufficient the field study confirmed that since construction of DTWs organization. About 60 percent of the money collected from in their area, they have not faced the severity of drought as farmers pays for electricity consumption, 10 percent goes to in earlier years. DTW operators, 2.5 percent to the smart card dealers, 15 to 20 percent for maintenance, and the remainder pays for DTWs substantially increase cropping intensity and other costs such as BMDA staff salaries. Field operations are lowered costs for farmers. Purchasing water from private overseen by the BMDA’s upazila offices, which are managed STWs has been expensive, costing between Tk 3,000 and by Assistant Engineers, who hire mechanics and other staff to 3,500 (approximately US$36 to $42) for irrigating one maintain the DTWs and collect data from the MVUs. The bigha (approximately 0.1 hectare) of boro (summer) paddy. BMDA is working to introduce telemetering systems that Irrigation from electric DTWs costs one-third to one-half would monitor DTW functioning from BMDA headquarters. less. The changes have boosted farmer income, reducing In this system, the data from each pump will automatically be migration in search of jobs and improving children’s uploaded to a server at the BMDA’s head office. education. PHOTO 1. Components of Prepaid Irrigation Fee Setup a. Mobile vending unit b. Prepaid smart card c. Prepaid meter Source: Authors. WATER GLOBAL PRACTICE CASE STUDY | PRO-POOR GROUNDWATER DEVELOPMENT 4 Emerging Concerns over Sustainability supplementary irrigation for aman rice. The same issue arises in the case of dry-season boro rice when STW functionality Recent studies have expressed apprehension about is particularly impacted, with a lack of irrigation also for overutilization of groundwater resources (Adhikary rabi crops such as potato, wheat, and maize. The overall et  al.  2013; Ahmeduzzaman, Kar, and Asad 2012; Islam impact is lower yields across these crops and seasons and et al. 2010; Rahamatulla et al. 2013). According to Dey et al. increased irrigation costs that now constitute more than (2013), some parts of the Barind region reveal a declining 50 percent of total production costs. Farmers in these areas trend in groundwater levels between 1981 and 2011. Aziz estimate that they therefore earn Tk 30,000 to 40,000 et al. (2015) show a direct relationship between the increase (US$360 – US$480) a year less than farmers serviced by a in boro paddy production and groundwater depletion in the DTW. They expect this gap to grow, believing production Rajshahi district (figure 1). costs will increase, particularly as a result of rising fuel costs. Farmer observations in Rajshahi and Thakurgoan reflect this Domestic hand pumps in such villages also struggle during concern (de Silva and Leder 2017). Farmers in the Tanore March and April each year. The division between STW- upazila of the Rajshahi district reported that the water table and DTW-dependent farmers in Thakurgaon is captured in late February had declined from about 9 meters in the in their differing perceptions of water security. Farmers in mid 2000s to about 23 meters now, whereas monsoonal DTW command areas expect another 15 to 20 years of recharge now causes it to rise to about 15 meters, compared good irrigation and feel it less imperative to change their to 5 or 6 meters in the past. Similarly, interviews with crop choices and cultivation practices. STW users are, not DTW operators in Rajshahi indicate DTW depths had to surprisingly, more anxious. In Rajshahi, the need to deepen be increased from time to time as groundwater levels drop DTWs and pump failure early in the boro season suggests well beyond 30 meters in the boro season. DTW operators there, too, groundwater-based irrigation is becoming more in some locations in Rajshahi also noted DTW pump challenging. breakage in the early boro season as a result of intensive use (de Silva and Leder 2017). The full implications of the divergence in water security views are still emerging, and villages dependent on STWs Although farming communities served by DTWs have significantly outnumber those served by a DTW (table 1). clearly benefited, communities without access to DTWs STWs, in fact, still account for most farmers’ served and report conditions have become more challenging, and this cultivated areas. Although this is not an immediate concern has potentially significant developmental implications (de in Rangpur, where water tables appear more stable, in Silva and Leder 2017). In some parts of the Thakurgaon Thakurgaon, DTWs irrigate only an estimated 30 percent district, STW users can no longer provide timely of cultivable land (de Silva and Leder 2017). With STW FIGURE 1. Production of Paddy in Metric Tons Shown Together with Minimum Groundwater Depth 2,000,000 7 6 groundwater (meters) Production of paddy Minimum depth to 1,500,000 5 (metric tons) 4 1,000,000 3 500,000 2 1 0 0 4 1 03 5 6 07 8 9 10 1 2 –0 –1 0 0 –0 0 0 0 9– 10 6– 2– 1– 4– 5– 8– 3– 0 07 0 0 20 0 0 0 0 0 0 0 20 20 20 20 20 20 20 20 20 20 Year Production of paddy (metric tons) Minimum groundwater depth (meters) Source: Aziz et al. 2015. WATER GLOBAL PRACTICE CASE STUDY | PRO-POOR GROUNDWATER DEVELOPMENT 5 TABLE 1. Deep Tubewells and Shallow Tubewells in Rajshahi and Rangpur Districts DTWs (electric and diesel) STWs (electric and diesel) Farmers Farmers District Wells Area irrigated served Wells Area irrigated served Rajshahi 16,495 553,095 1,339,311 328,101 706,905 2,168,594 Rangpur 7,445 212,979 365,335 396,938 779,339 2,519,088 Source: Minor Irrigation Survey 2015–16, Ministry of Agriculture, Government of Bangladesh. Note: DTWs = deep tubewells; STWs = shallow tubewells. irrigation (fuel) costs rising and rainfall and temperature water security issues. The future of STW users appears less fluctuations likely to increase irrigation demand, the certain across the board, though even some DTWs appear plight of STW-dependent farmers is a fundamental policy to struggle to deliver adequate water due to depletion, as in question. For these communities, the absence of DTWs is a Rajshahi. dual problem because hand pumps used to supply domestic water also cease to function. The development conditions The negative impacts associated with DTWs, namely where STWs no longer work may further regress if those of declining groundwater tables, speak to the households become mainly reliant on rain-fed cultivation. policy question of what should constitute “success” or In other words, they risk returning to a pre-irrigation era. “sustainability” with respect to this groundwater-based rice production model. Are changes in groundwater conditions In addition to lower earnings and food security, a scenario bringing to the surface an overdependence on irrigated where STWs become wholly or seasonally redundant also agriculture and the need to consider supply-side measures? means fewer crop combination options to respond to Is there a need to reimagine what sustainable and equitable climate change and other stresses. The capacity for greater rural development means and what indicators should be agrarian diversification was in fact highlighted by Gautam used to track future progress? The solution does not seem et al. (2016) as a key condition if multiple developmental to be to add more DTWs, and the government’s precaution goals are to be met, including reducing rural poverty and of suspending further investments in DTWs in this region malnourishment. The emerging concerns over future should be applauded. Although this study, based on data groundwater irrigation as a key means of production must from limited case studies, cannot make conclusions on the be taken seriously as a driver of developmental inequity that overall groundwater situation in the study districts, nor the may potentially reverse hard-won gains fighting poverty. prevalence of STW failure, the differing farmer irrigation experiences suggest future DTW investments should Conclusions be based on a more nuanced consideration of localized groundwater conditions and other users. The Barind approach augmenting groundwater-driven irrigation with about 16,000 DTWs has been an example The current suspension of new DTW installations also of adaptive resource management in the face of adverse provides a pause to reflect on the important emerging agroecological conditions. But field interviews suggest that trade-offs between food security policy defined through changing groundwater conditions are driving an increasing rice production and water security and crop diversification. heterogeneity in the outcomes arising from this irrigation- This break may present an opportunity to promote based rice-intensification model of rural development in more crop diversification that could reduce pressure on the Barind region. At the center of this emerging policy groundwater resources while addressing the need for issue is an apparent divergence in farmers’ developmental agrarian diversification, as emphasized by Gautam et  al. trajectories based on their geographical location and (2016). Farmer interviews suggest that the innovations whether they are served by DTWs or STWs. In Rangpur, around DTWs may, ironically, have stifled crop farmers linked to both DTWs and STWs appear to diversification as they attempt to maintain water security remain water secure. But in Thakurgaon, farmers in DTW in areas where groundwater levels may be in decline. Such command areas gain better crop yields, saving time and scenarios provide little incentive for farmers to shift from money, whereas villages relying on STWs are already facing their comfort zone of rice cultivation. In other words, WATER GLOBAL PRACTICE CASE STUDY | PRO-POOR GROUNDWATER DEVELOPMENT 6 diversification to less-water-demanding crops and other Asaduzzaman, M., and K. R. Rushton. 2006. “Improved water-saving methods is unlikely to occur voluntarily until Yield from Aquifers of Limited Saturated Thickness Using a water insecurity threshold is experienced, by which time Inverted Wells.” Journal of Hydrology 326 (1–4): 311–24. diversification would be reactive rather than a proactive adaptive strategy. Aziz, M. A., Md. A. K. Majumder, Md. S. Kabir, Md. I.  Hossain, Md. N. Rahaman, F. Rahman, and S. Hosen. The analysis in this case study purposefully ends on 2015. “Groundwater Depletion with Expansion of Irrigation cautionary footing. Although not seeking to detract in the Barind Tract: A Case Study of Rajshahi District of from the significant positive gains made through Bangladesh.” International Journal of Geology, Agriculture investments in DTW technology to date, it does seek to and Environmental Sciences 3 (1): 32–8. make the point that a resource base cannot sustain long- term overextraction, so the balance between benefits and de Silva, S., and S. Leder. 2017. “The Barind Multipurpose externalities is likely to shift. The evidence presented Development Agency, Electric Deep Tube Wells and here suggests a need for more-comprehensive research Agricultural Transformation in Northern Bangladesh: to verify groundwater conditions, risks to STW users, Lessons from Success and Emerging Challenges in and developmental implications to inform future policy Groundwater-Driven Responses to Water and Climatic on farm development driven by DTWs in the Barind Stress.” Working Paper under the project Improving Water region. Use for Dry Season Agriculture by Marginal and Tenant Farmers in the Eastern Gangetic Plains https://dsi4mtf.usq​ NOTES .edu.au/wp-content/uploads/2017/11/No-17-Agricultural​ -Innovations-for-water-security-Bangladesh-DeSilva-Leder​ 1. The researcher, Kolkata, India. -Final.pdf. CGIAR. 2. International Water Management Institute (IWMI), Colombo, Sri Lanka. Dey, N. C., S. K. Bala, A. K. M. S. Islam, and Md. A. 3. A. H. Chowdhury, personal communication, November 5, Rashid. 2013. “Sustainability of Groundwater Use for 2018. Irrigation in North-West Bangladesh.” Research Grant Report, Food Planning and Monitoring Unit, Government REFERENCES of People’s Republic of Bangladesh, Dhaka. http://fpmu​ .gov.bd/agridrupal/sites/default/files/ToR-2.pdf Adhikary, S. K., S. K. Das, G. C. Saha, and T. Chaki. 2013. “Groundwater Drought Assessment for Barind Irrigation Gautam, Madhur, Rashidur R. Faruqee, Md Mansur Projects in North-Western Bangladesh,” 20th International Ahmed, Forhad J. Shilpi, Shahidur R. Khandker, Congress on Modelling and Simulation, Adelaide, Australia, S.  Amer, Ahmed, Patrick Verissimo, Anuja Kar, and December 1–6. Gnanaraj Chellaraj. 2016. “Dynamics of Rural Growth in Bangladesh: Sustaining Poverty Reduction.” World Bank Ahmed, K. M., and W. G. Burgess. 1995. “Bils and the Group, Washington, DC. http://documents.worldbank​ Barind Aquifer, Bangladesh.” In Geomorphology and .org/curated/en/951091468198235153/Dynamics​ Groundwater, edited by A. G. Brown, 143–55. New York. -of-rural-growth-in-Bangladesh-sustaining-poverty​ John Wiley and Sons Ltd. -reduction. Ahmeduzzaman, M., S. Kar, and A. Asad. 2012. “A Study IMF (International Monetary Fund). 2013. “Sixth Five Year on Groundwater Fluctuation at Barind Area, Rajshahi.” Plan FY2011–FY2015: Accelerating Growth and Reducing International Journal of Engineering Research and Applications Poverty.” IMF, Washington, DC. 2 (6): 1465–70. Islam, M. B., M. Y. Ali, M. Amin, and S. M. Zaman. 2010. Amarasinghe, U. A., B. R. Sharma, L. Muthuwatta, and “Climatic Variations: Farming Systems and Livelihoods in Z. H. Khan. 2014. “Water for Food in Bangladesh: Outlook the High Barind Tracts and Coastal Areas of Bangladesh.” to 2030.” IWMI Research Report 158, International In Climate Change and Food Security in South Asia, edited Water Management Institute, Colombo, Sri Lanka. by R. Lal, M. Sivakumar, R. A. Mustafizur, and K. Islam, doi:10.5337/2014.213. 477–97. Dordrecht, The Netherlands: Springer. WATER GLOBAL PRACTICE CASE STUDY | PRO-POOR GROUNDWATER DEVELOPMENT 7 Jahan, C. S., Q. H. Majumder, N. Akter, M. I. Adham, Mainuddin, M., and M. Kirby. 2015. “National Food and M. A. Zaman. 2010a. “Hydrological Environment and Security of Bangladesh to 2050.” Food Security 7 (3): 633. Groundwater Occurrences in the Plio-Pleistocene Aquifer https://doi.org/10.1007/s12571-015-0465-6. in Barind Region, Northwest Bangladesh.” Bangladesh Geological Journal 16: 23–37. Mainuddin, M., M. Kirby, R. A. R. Chowdhury, L.  Sanjida, M. H. Sarker, and S. M. Shah-Newaz. 2014. Jahan, C. S., Q. H. Mazumder, A. T. M. M. Islam, and M. I. “Bangladesh Integrated Water Resources Assessment Adham, 2010b. “Impact of Irrigation in Barind Area, NW Supplementary Report: Land Use, Crop Production, and Bangladesh—An Evaluation Based on the Meteorological Irrigation Demand.” Water for a Healthy Country Flagship Parameters and Fluctuation Trend in Groundwater Table.” Report,  CSIRO, Australia. https://publications.csiro.au/rpr​ Journal of the Geological Society of India 76 (2): 134–42. /download?pid=csiro:EP141373&dsid=DS6. Jolliffe, Dean, Iffath Sharif, Lea Gimenez, and Faizuddin Qureshi, A. S., A. Ahmed, and T. J. Krupnik. 2014. Ahmed. 2013. “Bangladesh - Poverty Assessment: Assessing “Groundwater Management in Bangladesh: An Analysis a Decade of Progress in Reducing Poverty, 2000-2010.” of Problems and Opportunities.” Cereal Systems Initiative Bangladesh development series; paper no. 31. World for South Asia Mechanization and Irrigation (CSISA- Bank, Washington DC. http://documents.worldbank.org​ MI) Project, Research Report No. 2, CIMMYT, Dhaka, /curated​/en/109051468203350011/Bangladesh-Poverty​ Bangladesh. -assessment-assessing-a-decade-of-progress-in-reducing​ -poverty-2000-2010. Rahmatullah Imon, A.H.M. and Ahmed, M. 2013. “Water Level Trend in the Barind Area.” Malaysian Journal of Kang, A. 2013. “Barind’s Three Crop Revolution.” Down to Mathematical Sciences 7(1): 1-15. Earth, October 5. Connect with the Water Global Practice www.worldbank.org/water ­ worldbankwater@worldbank.org ­ @worldbankwater ­ logs.worldbank.org/water b © 2019 International Bank for Reconstruction and Development / The World Bank. Some rights reserved. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. This work is subject to a CC BY 3.0 IGO license (https://creativecommons.org/licenses/by/3.0/igo). The World Bank does not necessarily own each component of the content. It is your responsibility to determine whether permission is needed for reuse and to obtain permission from the copyright owner. If you have questions, email pubrights@worldbank.org. SKU W19066