Policy Research Working Paper 10109 Estimating Local Agricultural GDP across the World Brian Blankespoor Yating Ru Ulrike Wood-Sichra Timothy S. Thomas Liangzhi You Erwin Kalvelagen Development Economics Development Data Group June 2022 Policy Research Working Paper 10109 Abstract Economic statistics are frequently produced at an admin- administrative statistics of Agricultural GDP into a global istrative level such as the sub-national division. However, gridded dataset at approximately 10 x 10 kilometers using these measures may not adequately capture the local vari- satellite-derived indicators of the components that make up ation in the economic activities that is useful for analyzing agricultural GDP, namely crop, livestock, fishery, hunting local economic development patterns and the exposure to and timber production. The paper examines the exposure natural disasters. Agriculture GDP is a critical indicator of areas with at least one extreme drought during 2000 to for measurement of the primary sector, on which 60 per- 2009 to agricultural GDP, where nearly 1.2 billion people cent of the world’s population depends for their livelihoods. live. The findings show an estimated US$432 billion of Through a data fusion method based on cross-entropy opti- agricultural GDP circa 2010. mization, this paper disaggregates national and subnational This paper is a product of the Development Data Group, Development Economics. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at bblankespoor@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 Estimating Local Agricultural GDP across the World † Brian Blankespoor ‡, Yating Ru §, Ulrike Wood-Sichra ¶ Timothy S. Thomas Liangzhi You ∗∗ Erwin Kalvelagen ¶ Keywords : agriculture, regional economic activity, GDP, cross-entropy JEL Classification Codes : Q10, R12 † The authors would like to thank the following people for discussions and prior reviews: Claudia Berg (World Bank), Gero Carletto (World Bank), Uwe Deichmann (World Bank), Rashmin Gunasekera (World Bank), Barbro Hexeberg (World Bank), Glenn-Marie Lange (World Bank), Michael Lokshin (World Bank), Eric Roland Metreau (World Bank), Jose Pablo Valdes Martinez (World Bank), Tim Robinson (FAO), Steven Rubinyi (World Bank), Juha Siikam¨ aki (IUCN), Ben Stewart (World Bank), Jeffrey R. Vincent (Duke Uni- versity). We appreciate the use of non-timber value data from Juha Siikam¨ aki and ports data from Gilles Hosch. The authors would like to thank the participants of conferences including: American Association of Geographers Annual Meeting 2019 in Washington, DC, 3-7 April 2019; United Nations Economic Com- mission for Europe Workshop on Data Integration: Realising the Potential of Statistical and Geospatial Data in Belgrade, Serbia, 21-23 May 2019; International Institute for Applied Systems Analysis seminar in Laxenburg, Austria, 27 September 2019; IFPRI RISE Workshop in Washington, DC, 19 November 2019; and the Committee for the Coordination of Statistical Activities and United Nations Geospatial Network Joint virtual Workshop on the Integration between Geospatial and Statistical Information, 28 April 2021. We appreciate the support of the World Bank Strategic Research Program on Big Data. The maps displayed in this paper are for reference only. The boundaries, colors, denominations and any other information shown on these maps do not imply, on the part of the World Bank Group any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. ‡ World Bank. Corresponding author can be reached at: bblankespoor@worldbank.org These authors contributed equally to this work. § Cornell University. Email: YR97@cornell.edu ¶ International Food Policy Research Institute International Food Policy Research Institute. Email: T.Thomas@cgiar.org ∗∗ International Food Policy Research Institute. Email: L.YOU@cgiar.org 1 Introduction According to the Food and Agriculture Organization of the United Nations, over 60 percent of the world’s population depends on the agricultural sector for their livelihoods and it provides a key source of income and employment for the poor and vulnerable people (FAO, 2013; FAO, 2019). Yet, economic statistics of the agricultural sector are frequently produced at a national or lower administrative level and may not adequately capture the local variation. Furthermore, a spatial mismatch may exist between the geographic unit of interest like the natural area of a river and the administrative area. Detailed agricultural data are critical to examining a wide range of agricultural issues including technology and land use (e.g. Bella and Irwin, 2002; Luijten, 2003; Samberg et al., 2016; Staal et al., 2002), exposure to natural hazards (e.g. Murthy, Laxman, and Sai, 2015) and patterns of and productivity of economic development (e.g. Elhorst and Strijker, 2003; Gollin, Lagakos, and Waugh, 2014; G. C. Nelson, 2002; Reddy and Dutta, 2018). Carr˜ ao, Naumann, and Barbosa (2016) examine the exposure of people and economic activity to drought using measures of physical elements (e.g. cropland and livestock). Rentschler and Salhab (2020) find that low and middle-income countries have 89% of global flood exposed population and poor people account for almost 600 million, who are directly exposed to the risk of intense flooding. Vesco et al. (2021) examine linkages between climate variability and agricultural production as well as conflict. They find that climate variability contributes to an increase in the spatial concentration of agricultural production within countries. Further- more, in countries with a high share of agricultural employment in the national workforce, they find this combined effect increases the likelihood of conflict onset. To better target rural development strategies for economic growth and poverty reduction, as well as conserve the natural resource base for long-term sustainable development, we need to accurately delin- eate the spatial distribution of agricultural resources and production activities (Wood et al., 1999). One method to partially address spatial mismatch between administrative and other geographic units such as natural hazards uses the gridded (raster) data format by providing an intermediate and consistent unit for disaggregation and aggregation (e.g. UNISDR, 2011). Data-disaggregation methods can use detailed data to inform estimates of aggregated data from large areas at the local level (e.g. see review in Pratesi et al., 2015). Several spatial data products from global models are available to estimate population at a local level (see review in Leyk et al., 2019). Previous evidence-based risk analyses take advantage of global data of hazards to esti- mate exposure of population and economic activity (e.g. Gunasekera et al., 2018; Rashmin 3 Gunasekera et al., 2015; Rentschler and Salhab, 2020; Ward et al., 2020). Gross Domes- tic Product (GDP) is a critical economic indicator in the measurement and monitoring of an economy in a country that is typically only available at national and occasionally sub- national levels. Regional indicators play a key role in the necessary variation to forecast regional GDP (Lehmann and Wohlrabe, 2015) and food security (Andree et al., 2020). Pre- vious efforts to estimate local GDP use high resolution spatial auxiliary information such as luminosity or population data to provide local variation. Methods by Kummu, Taka, and Guillaume (2018), Murakami and Yamagata (2019), Nordhaus (2006), and World Bank and UNEP (2011) took advantage of gridded population data, which is the result of a model disaggregating the most detailed level population data into grids (e.g. see review in Leyk et al., 2019). However, wealth is not evenly distributed among people nor infrastructure (Berg, Blankespoor, and Selod, 2018). In fact, the divide between the rich and poor is even widening in our time (Dabla-Norris et al., 2015). The method used in World Bank and UNEP (2011) stratify the population by rural and urban, yet definition of these geographic areas can vary based on the selection of the population model (Leyk et al., 2019). These measurements matter in application to stylized facts such as the strong negative correlation of the level of urbanization with the size of its agricultural sector (Roberts et al., 2017). Also, the uniform distribution of labor in agriculture is another key concern (Gollin, Lagakos, and Waugh, 2014). Other methods used land cover such as vegetation and built-up indices, however did not incorporate types of agriculture like cropland and livestock (Goldblatt, Heilmann, and Vaizman, 2019; Rashmin Gunasekera et al., 2015). Other methods to estimate GDP at a local level take advantage of the lights at night dataset. Doll, Muller, and Morley (2006) and Elvidge et al. (2009) found nighttime lights to provide a uniform, consistent, and independent estimate for economic activity, and several other studies (e.g. Bundervoet, Maiyo, and Sanghi, 2015; Chen and Nordhaus, 2011; Eberenz et al., 2020; Ghosh et al., 2010; Henderson, Storeygard, and Weil, 2012; Wang, Sutton, and Qi, 2019) utilized this striking correlation between luminosity and economic activities to estimate economic output on the ground. While night light is a good reflection of economic activities in manufacturing and urban areas, night light data may not capture the agricultural activity as it requires areas to emit light. Bundervoet, Maiyo, and Sanghi (2015) suggest that agricultural indicators rather than rural population could improve the estimation of GDP given the importance of agriculture in many of the economies in their sample of Africa. Gibson et al. (2021) find that night time lights data are a poor predictor of economic activity in low density rural areas. In this paper, we present a high resolution gridded Agricultural GDP (henceforth Ag- GDP) dataset that is produced through a spatial allocation model by distributing national 4 and sub-national statistics to 5 arcminute pixels based on satellite-derived information of constituents of AgGDP, including forestry, hunting, and fishing, as well as cultivation of crops and livestock production.1 We make two main contributions. First, we construct a global dataset of gridded AgGDP. This entails a massive effort of data collection and integra- tion. We extend and apply the cross-entropy framework developed in the Spatial Production Allocation Model (SPAM) for crops that pioneered the use of cross-entropy optimization in spatial allocation (You and Wood, 2003; You, Wood, et al., 2014; You, Wood-Sichra, et al., 2018; Yu et al., 2020). We construct and integrate global datasets of the components of agricultural GDP as priors and then reconcile the values with the regional account statistics using cross-entropy optimization. Second, we contribute to efforts assessing the exposure of economic activity to natural hazards with a focus on agricultural GDP. Significant progress has been made to measure physical assets such as built-up area and estimate hazards to quantify its exposure to natural disasters. However, the spatial distribution of agricultural GDP is less known. So, we apply these data to inform efforts quantifying the population and agricultural GDP at risk to drought and water scarcity. The rest of this paper is structured as follows. The next section provides a detailed description of the methodology and data. Then, we present the model results and data. Then, we discuss the results along with validation followed by usage notes from a fitness-for- use perspective. Finally, we provide concluding remarks. 2 Methodology and data Following the composite structure of agricultural GDP, we disaggregate the national and sub-national statistics into a global grid through a cross-entropy allocation model. Given the availability of data and the global scope, our efforts varied on adjusting official statistics and creating priors for different components. Below we discuss the construction of each com- ponent, AgGDP statistics and the allocation model followed by the global natural hazards data. Given the spatial resolution and year of reference of the input data for the crop value of production, we estimate AgGDP for the year 2010 into 5 arc-minute grids (10x10 km) across the world. 1 Agriculture, forestry, and fishing corresponds to ISIC divisions 1-3 and includes forestry, hunting, and fishing, as well as cultivation of crops and livestock production 5 2.1 Construction of components For each pixel, we construct an estimated value of production based on high spatial resolution information of the five components that serve as priors in the modeling process: crop, livestock, forestry, fishing, and hunting. Given the lack of information on the hunting component, we disaggregate the forestry component into two parts: timber and non-timber products of forestry. The non-timber products of forestry includes an even distribution of hunting. The construction of the five components is described below in four subsections: crop, livestock, forestry (timber and non-timber) and fishing. 2.1.1 Crop value of production The crop component in the gridded AgGDP is generated by multiplying the quantity of production from the global SPAM 2010 version 1 dataset 2 (You, Wood-Sichra, et al., 2018) with the producer prices at the country level from FAOSTAT (FAO, 2016) for each crop and then summed together. 3 As mentioned earlier, SPAM is a cross-entropy model, which calculates a plausible allocation of crop areas and production to approximately 10 km pixels, based on agricultural statistics at national and sub-national levels, combined with gridded layers of cropland, irrigated areas, population density and potential crop areas and yields (Yu et al., 2020). SPAM’s output distinguishes between 42 crops (33 individual crops, 9 aggregated crops) that together add up to practically all cultivated crops in a country with four parameters including production, yield, physical area and harvest area. For aggregated SPAM crops (such as other cereals, other pulses, vegetables, fruits, etc.), we computed their prices by taking the weighted average of their components, as follows: Σj pricej prodj P riceJagg = , ∀j ∈ Jagg (1) Σj prodj where Jagg is the aggregated crop group, j is any crop that belongs to Jagg, PriceJagg is the price of the aggregated crop group, pricej is the price of crop j, and prodj is the production of j. For each grid, the value of crop production is thus: Cropvali = Σj prodi,j pricej , ∀j that grow in pixel i (2) where Cropvali is the value of total crop production in pixel i, prodi,j is the production of 2 Available at www.mapSPAM.info 3 As for the producer price, ideally, we need sub-national level figures, but such a dataset is not available globally. Therefore, we use the FAOSTAT’s national producer prices. 6 crop j in pixel i, and pricej is the price of crop j. A map of global gridded crop production value as a prior is shown in Figure 1. Figure 1: This map illustrates the assembled crop production value used as a prior in the cross-entropy model. Sources: Authors’ calculation (2022), FAO (2016), and Yu et al. (2020) 2.1.2 Livestock production Livestock accounts for an estimated 40% of the global value of agriculture output and plays an important role in ensuring the livelihoods and food security for over one-sixth of the world’s population (FAO, 2018). Yet, it is still under rapid expansion as the global demand for animal-sourced products such as meat, milk, eggs, and hides continues to grow (Herrero and Thornton, 2013). While species and quantities of livestock raised vary among regions and husbandry farmers, there are five primary species - cattle, sheep, goats, pigs, and chicken - that prevail worldwide and provide essential products for human consumption. We calculate the component of livestock production in gridded AgGDP based on the distribution maps of the above five primary species from the Gridded Livestock of the World (Gilbert et al., 2018; Robinson et al., 2014) and FAOSTAT’s value of production of live- stock products (including meat, milk, eggs, honey and wool) (FAO, 2020). To facilitate comparison, the animal-specific density numbers are converted to one animal type by using International Livestock Units (Eurostat, 2018), as shown in Table 1. Then the densities 7 Table 1: This table shows the livestock type with the conversion factor. Sources: Eurostat (2018) livestock type conversion factor Cattle 1 Pigs 0.3 Goats 0.1 Sheep 0.1 Chicken 0.01 of the animal equivalent values are multiplied by pixel areas to get the count of animals per grid, which is multiplied by the FAOSTAT’s value of production to obtain the livestock production prior for each pixel. lsnumi lsvali = lsvalx , ∀i ∈ X (3) ΣX lsnumi where lsvali is the total value of livestock production in pixel i; lsvalx is the value of livestock production (meat, milk, eggs, honey and wool) that is reported at the national level; lsnumi is the total number of equivalent animals in pixel i ; and X is a set including all pixels that fall within the boundary of a nation. A map of global gridded livestock production value as a prior is shown in Figure 2. 2.1.3 Forestry production and hunting Forest resources have been utilized by people since the advent of civilization (Hossain, Alam, Miah, et al., 2008). Up until now, over a billion people still rely on forest resources for food security and income generation to some extent (FAO, 2018). In the world’s least developed regions, 34 countries depend on fuelwood to provide more than 70% of energy, among which 13 nations require 90% of energy (FAO, 2018). The contribution of forest production to AgGDP can be classified into two broad types: wood (logging) products and non-wood forest products. Wood (logging) products are the most exploited commodities in the forestry sector. The trees are cut down to be the raw materials for producing timber and pulp, which are further processed and converted into a number of derivatives, such as construction materials and paper products. Non-wood forest products are defined by the Food and Agriculture Organization of the United Nations (FAO). 4 It is estimated that millions of households around the world depend on non-wood 4 These products are “goods of biological origin other than wood derived from forests, other wooded land and trees outside forests”, including foods (nuts, fruits, mushrooms, etc.), food additives (herbs, spices, sweeteners, etc.), fibers (for construction, furniture, clothing, etc.), and plant and animal products with chemical, medical, cosmetic or cultural value. 8 Figure 2: This map illustrates the assembled livestock production value used as a prior in the cross-entropy model. Sources: Eurostat (2018), Gilbert et al. (2018), and Robinson et al. (2014) forest products for their livelihood. Some 80% of people in the developing world use these products in their everyday life (Sorrenti, 2016). For a complete assessment of forest production priors, this study takes both wood and non-wood products into consideration. The gridded non-wood forest products dataset used in this study was jointly developed by Resources for the Future and the World Bank (Siikam¨ aki, ´ Santiago-Avila, and Vail, 2015) through an approach of meta-regression modeling, which integrates over 100 estimates at various locations from a literature review and multifold information on ecological and socioeconomic factors. The value of non-wood forest products is resampled to the 5 arc-minute grid cell size and converted to 2010 USD for consistency with other AgGDP components. As part of non-timber products, we include hunting with an even distribution across units and time given the lack of information. The value of wood products per pixel is calculated based on forest loss from year 2010 to year 2011 excluding loss due to fire, with an assumption that the forests were mainly cut down for timber production. The Moderate Resolution Imaging Spectroradiometer (MODIS) Land Cover map (Friedl et al., 2010) for year 2011 is overlaid on top of that for year 2010 to detect the area that has changed from forest to non-forest. 5 However, forest loss due to fire 5 The measurement is limited to detection of land cover change from satellite and will likely account for selective harvesting or forest degradation. 9 should be removed because it does not result in wood products. Thus, fire information for year 2010 is obtained from the NASA Fire Information for Resource Management System (FIRMS) (NASA, 2018) and areas that experienced forest fire are eliminated. After the identification of forest area change in each pixel, the value of wood production at national level is taken from a FAO lead project (Lebedys and Yanshu, 2014) and proportionally disaggregated to arrive at a pixel-wise value of wood products as follows: f orestlossi W oodvali = (f orestvalx − nonwoodvalx ) , ∀i ∈ X (4) ΣX f orestlossi where Woodvali is the value of wood products in pixel i; forestvalx is the value of forest prod- ucts reported at national level; nonwoodvalx is the value of non-wood products at national level which is derived from Siikam¨ ´ aki, Santiago-Avila, and Vail (2015); forestlossi is the area of forest loss excluding loss to fire in pixel i ; again, X is a set including all pixels that fall within the boundary of a nation. A map of global gridded wood forest production value as a prior is shown in Figure 3. Figure 3: This map illustrates the assembled wood forest production value used as a prior in the cross-entropy model. Sources: Authors’ calculation (2022), Friedl et al. (2010), NASA (2018), and Siikam¨ ´ aki, Santiago-Avila, and Vail (2015) 10 2.1.4 Fishery production Fish makes up approximately 17% of animal-sourced protein in the human diet worldwide (Mathiesen, A. ` M., 2018). The fishery industry supports the livelihood of 12% of world population by creating 200 million jobs along its value chain. In the global trade system, 80 billion USD worth of fish is exported from developing countries and it plays a crucial role in promoting local economic development (Kelleher, Willmann, and Arnason, 2009). We estimate both freshwater inland fisheries and marine production values using the FISHSTAT (FAO, 2009) data with a classification based on the fish production categories. The inland fishery production value is the result of disaggregating corresponding country level statistics in proportion to areas of inland water bodies in pixels. The distribution of inland water bodies is obtained from the ESA-CCI (Lamarche et al., 2017). Thus, the value of inland fishing production in each grid is calculated as follows: waterbodyi f ishvali = f reshvalx , ∀i ∈ X (5) ΣX waterbodyi where fishvali is the value of fishery production in pixel i ; freshvalx is the value of fresh fish production at the national level which is aggregated from FISHSTAT; waterbodyi is the area of water bodies in pixel i ; and X is a set including all pixels i that fall within the boundary of a nation x. The value of marine fisheries production is based on its proximity to fish landing ports weighted by a composite indicator of equal weight from the number of visits and sum of the vessel hold of fishing vessels. We use the port database from the World Port Index (National Geospatial-Intelligence Agency, 2019) and the number of port visits with a vessel hold of fishing vessels from Hosch et al. (2019) to create a composite variable as the prior based on the sum (for each port) of the number of visits (each event in the database) and total vessel hold at the port. The geographic coverage of the ports is calculated for each port using the minimum port distance provided in (Hosch et al., 2019). Any distance greater than 150km is calculated at 150km. The value of marine fishing production in each grid is calculated as follows: portindexi marinevali = marinevalx , ∀i ∈ X (6) ΣX portindexi where marinevali is the value of fishery production in pixel i ; portindexi is an equal weighted composite index of the number of visits in pixel and the total vessel hold in pixel i ; and X is a set including all pixels i that fall within the boundary of a nation x. A map of global gridded fishery production value as a prior is shown in Figure 4. 11 Figure 4: This map illustrates the assembled fishery production value (prior) used as a prior in the cross-entropy model. Sources: Authors’ calculation (2022), FAO (2009), Hosch et al. (2019), Lamarche et al. (2017), and National Geospatial-Intelligence Agency (2019) . 2.2 AgGDP Statistics and Linked Grids Tremendous effort has been made to collect and organize national and sub-national statis- tics from various sources of national ministries or from reports. However, not every country publishes its agricultural GDP figures at the sub-national (regional) level and different meth- ods of regionalization exist including top-down, bottom-up and mixed methods (Eurostat, 2013). 6 Our database has 68 countries that have sub-national agricultural GDP data, ex- pressed in varying domestic currencies and for different years. Table A1 in the Supplementary Information (SI) lists these countries and descriptive statistics including temporal coverage and number of subnational regions at an administrative geographic level including NUTS level. 7 6 Regional Gross Domestic Product (RGDP) can be estimated following the production, income or ex- penditure approaches. However, RGDP is not typically compiled using the expenditure approach due to the scarcity of data such as inter-regional purchases and sales, or regional exports/imports. On the production and income approaches, the estimate of market activities is typically from the production approach, whereas the estimate of non-market industries is from the income approach. 7 The European Union developed a standard for administrative levels: The Classification of Territorial Units for Statistics (NUTS, for the French nomenclature d’unit´ es territoriales statistiques). 12 To overcome discrepancies in temporal coverage and currency terms (constant and cur- rent), and to keep the data consistent and comparable for countries across the world, shares from sub-national statistics are calculated and then applied to a national total to derive a calibrated number at the sub-national level. The national totals are obtained from the World Development Indicators (World Bank, 2019) and averaged over three years around 2010. For a few countries, which do not report their national agricultural GDP in the WDI database, sums of all agricultural GDP components are used as proxies. The calibrated statistics are then linked to grids through a shapefile of the Global Admin- istrative Unit Layers (GAUL) that maintains global geographic layers with a consistent and comprehensively unified coding system (FAO, 2015). Then, we overlay the GAUL adminis- trative boundaries on the grid network to assign the corresponding codes of the administra- tive units to each grid. For areas where sub-national AgGDPs have different administrative areas than GAUL, the GAUL areas are merged or split to match the sub-national AgGDP area. 2.3 Spatial Allocation Model After constructing all the components, we define a spatial allocation model in a cross- entropy framework following (You, Wood, et al., 2014) to allocate administrative statistics to 5 arc-minute pixels.8 National and sub-national AgGDP values are used as a constraint, while the distribution of crop, livestock, fishery, and forestry production (hunting is included in non-timber products of forestry) is used to create priors for estimating pixel-level AgGDP. Measurement units are unified using deflators and exchange rates. 9 The first step is to transform all real-value parameters into corresponding probabilities. Let Si be the share of the total agricultural GDP allocated to pixel i within a country x. AgGDPi,x is the agricultural GDP allocated to pixel i in country x and X is a set including all pixels that fall within the boundary of a nation. Therefore: AgGDPi,x Si = , ∀i ∈ X (7) ΣX AgGDPi,x Let PreAgGDPi be the pre-prior allocation of AgGDP share from our best estimate. The first approximation can be done by summing all five calculated pixel level components of 8 A comprehensive presentation of the cross-entropy method is in Rubinstein and Kroese (2004) 9 The currency varies by source. Crops are in local currency. Livestock are in International USD 2004- 2006. Fish are USD 2009. Non-timber forest products are in USD 2012 and Timber (forest) are in USD 2011. 13 AgGDP: P reAgGDPi = Cropi + Livestocki + F orestryi + F ishingi + Huntingi (8) where we assume hunting occurs in areas with equal probability. Theoretically, the sum of these components should be close to the official values obtained from the World Development Indicators. We make sure that the official AgGDP values are guaranteed to be no less than the sum of all five components of agricultural GDP. Therefore, we first sum up all prior estimations of AgGDP. AgGDPx = Σi∈x P reAgGDPi (9) Then, we rescale the prior AgGDP to be consistent with the official AgGDP value: P reAgGDPi AgGDPx P riorAgGDPi = (10) Σi P reAgGDPx Then we calculate the prior for Si as a probability by normalizing PriorAgGDP: P riorAgGDPi,x P reAlloci = (11) Σi∈X P riorAgGDPi Finally, we formulate a cross entropy model in the following mathematical optimization framework: M IN CE (Si ) = Σi Si log (Si ) − Σi Si log (P reAlloci ) (12) Subject to the following three conditions: Σi Si = 1 (13) Σi∈k (ΣAgGDP )Si = SubAgGDPk ∀k ) (14) 0 ≤ Si ≤ 1 ∀i (15) where i : i=1,2,3,. . . are pixel identifiers within the allocation unit (e.g. Brazil); and k : k=1,2,3, . . . are identifiers for sub-national geopolitical units (e.g. a state) where AgGDP values (SubAgGDPk ) are available. The objective function is defined as the cross entropy of AgGDP shares and their prior. The first constraint (Equation 13) is the pycnophylactic or volume-preserving constraint (e.g. Tobler, 1979) that ensures the sum of all allocated AgGDP values is equal to the total AgGDP of the country. The next equation (14) sets the sum of all allocated AgGDP within those subnational units with available data to be 14 equal to the corresponding sub-national AgGDP values. The last equation (15) is a natural constraint for the percentage of AgGDP, which is also the probability in the cross-entropy model. The modeling framework is flexible in that more constraints can be added if more data are available and/or more reasonable assumptions on how AgGDP should be spatially disaggregated are discovered. 10 Last but not least, we multiply the total regional agricultural GDP by the probability in the cross-entropy model to derive the final pixel level agricultural GDP: AgGDPi = Σi AgGDPx Si (16) 2.4 Natural hazards We use measures of two natural hazards to gain insight into the spatial distribution of agricultural activity with regards to drought and water scarcity. We calculate the Standard- ized Precipitation-Evapotranspiration Index (SPEI) (Sergio M Vicente-Serrano, Beguerıa, and L´ opez-Moreno, 2010), which measures the difference between observed precipitation and estimated potential evapotranspiration with a 3 month interval using the base climatol- ogy of 1980 to 2019, which is implemented in R (Beguer´ ıa and Sergio M. Vicente-Serrano, 2017) using climate data from Harris et al. (2020). Extra dry years are defined as the number of years that are less than or equal to -2.0 during the period from 2000 to 2009. Figure 5 shows the results of the SPEI. The Water Crowding Index (WCI) is a measure of water scarcity considering the lo- cal population as the annual water availability per capita (Falkenmark, 1986; Falkenmark, 2013). Veldkamp et al. (2015) model global water crowding index with return periods. We take the mean of any pixels of the ensemble WCI with a 10 year return period within an agricultural GDP pixel. Following the literature (e.g. Alcamo, Fl¨ orke, and M¨ arker, 2007; Arnell, 2003; Kummu, Ward, et al., 2010; Veldkamp et al., 2015), we categorize the WCI into four categories: Absolute is less than 500 m3 /capita per year; severe is less than or equal to 1000 m3 /capita per year; moderate is less than or equal to 1,700 m3 /capita per year; and low is the remainder (Figure 6). Then, we evaluate water shortage events using a threshold of 1,700 m3 /capita per year with a return period of 10 years. 10 For instance, market access may play a role in determining the spatial distribution or spatial structure of AgGDP and can be included as a constraint in the model. However, we provide a parsimonious model without market access. 15 Figure 5: This map illustrates the number of years with at least one extreme drought from 2000 to 2009 measured by a 3 month SPEI. Sources: Authors’ calculation (2022), Beguer´ ıa and Sergio M. Vicente-Serrano (2017), and Harris et al. (2020) . 2.5 Night time lights Night time lights data are commonly used in the estimation of local human development and economic activity (e.g. Bruederle and Hodler, 2018; Bundervoet, Maiyo, and Sanghi, 2015; Ghosh et al., 2010; Henderson, Storeygard, and Weil, 2012; Kummu, Taka, and Guil- laume, 2018). We use the radiance calibrated data for 2010 from the F16 satellite to quantify the correlation between agricultural GDP and night time lights by geographic regions of the world defined by the World Bank. 11 3 Results and Discussion Figure 7 illustrates the result of the cross-entropy model in a global map of gridded agri- cultural GDP. The global gridded AgGDP for the year 2010 in 2010 US dollars is in gridded 11 Specifically, we use the version 4 product from the F16 satellite (20100111 - 20101209) available at: https://ngdc.noaa.gov/eog/dmsp/download radcal.html 16 Figure 6: This map illustrates the Water Scarcity Index categories with a return period of 10 years. Sources: Authors’ calculation (2022) and Veldkamp et al. (2015) . (raster) format at a resolution of 5 arc-minute, which approximates to 10 km. 12 The spatial extent and quantity distribution of AgGDP over the world are in agreement with general knowledge of agricultural technology adoption and suitability, with well-known agricultural nations, such as India, China and the United States standing out as regions with high Ag- GDP. A number of European countries also exhibit high agricultural GDP values, which is likely due to the benefit of adopting mechanized farming and technological facilitation, considering that the shares of agricultural land and agrarian population are relatively low in these well-developed places. Countries in Sub-Saharan Africa remain low in agricultural production, as indicated by low-value pixels sparsely spreading over the continent. Within the continent, agricultural production activities primarily take place in geographic areas with suitability. The correlation of AgGDP with night light varies across world regions as it requires areas to emit light (Table 2). Most World Bank regions have similar patterns of correlation with night time lights across the measures of AgGDP, GDP and population. The relationship is strongest with the correlation between GDP and night light compared with AgGDP or population. Likewise, World Bank income groups show similar patterns across the measures 12 The coordinate system is the standard WGS84 and saved in GeoTIFF format. The data will be publicly and freely available through the World Bank Development Data Hub website. 17 Figure 7: This map illustrates the global gridded Agricultural GDP circa 2010 from the Cross-Entropy model in 2010 USD. Source: Authors’ calculation (2022) . with lower middle and upper middle income groups having higher correlations than low and high income groups. Following previous global studies (e.g. Blankespoor, Dasgupta, and Lange, 2017), we present an application of exposure to a natural hazard with the AgGDP dataset and popu- lation. A common drought measure is the Standard Precipitation Evapotranspiration Index (SPEI) (Sergio M Vicente-Serrano, Beguerıa, and L´ opez-Moreno, 2010). The global popu- lation estimates for the year 2010 are from WorldPop and Center for International Earth Science Information Network (CIESIN), Columbia University, 2018. 13 The exposure to drought is not uniform across the world. Across the world, the group of high income countries have less population and agricultural GDP exposed to drought in each number of years with extremely dry compared to the countries in other income categories (Figure 8). The top ten countries in agricultural GDP exposed to an extreme drought from 2000 to 2009 include the large economies in the agriculture sector such as China, India, the United States and Russian Federation (Table 3). However, other countries have a high share of their agricultural GDP exposed to an extreme drought (Table 4). The top 10 countries in 2010 population exposed to dry areas include countries with the largest economies in the 13 They use a Random Forest-based dasymetric redistribution method. 18 Table 2: This table shows the Spearman correlation with night time lights across the measures of AgGDP, GDP and population grouped by World Bank Region where AFR is Sub Saharan Africa; EAP is East Asia and Pacific; ECA is Eastern Europe and Central Asia; LCR is Latin America; MENA is Middle East and North Africa; SOA is South Asia and Other is the category for the remaining countries. Sources: Authors’ calculation (2022), NOAA (2011), and World Bank (2019). World Bank Regions AgGDP and NTL GDP and NTL POP and NTL AFR 0.665 0.874 0.694 EAP 0.951 0.953 0.947 ECA 0.830 0.894 0.790 LCR 0.941 0.961 0.925 MENA 0.779 0.893 0.756 Other 0.528 0.549 0.529 SOA 0.929 0.952 0.929 agriculture sector as noted above, but the list includes countries such as the Democratic Republic of Congo, Tanzania and Uganda (Table 5). Figure 8: The total exposure of agricultural GDP [A] and population [B] aggregated from areas with at least one extreme drought from 2000 to 2009 measured by a 3 month SPEI. Sources: Authors’ calculation (2022), World Bank (2019), and WorldPop and Center for International Earth Science Information Network (CIESIN), Columbia University (2018). Across the world, high income countries have less population and agricultural GDP in areas of absolute or severe categories of the Water Crowding Index compared to countries in other income categories (Figure 9). The top ten countries of agricultural GDP exposed to 19 the Water Crowding Index include large economies in the agriculture sector such as China, India, Pakistan, Indonesia and Nigeria (Table 6). However, several countries have a high share of their agricultural GDP exposed to the Water Crowding Index (Table 7). The top 10 countries in 2010 population exposed to dry areas include countries with the largest economies in the agriculture sector as noted above, but the list includes countries such as Bangladesh, the Arab Republic of Egypt and Mexico (Table 8). Figure 9: Total Agricultural GDP [A] and population [B] by mean Water Crowding Index, where Absolute is less than 500 m3 /capita per year, severe is less than or equal to 1000 m3 /capita per year, moderate is less than or equal to 1700 m3 /capita per year and low is the remainder. Sources: Authors’ calculation (2022), Veldkamp et al. (2015), and World Bank (2019). 3.1 Validation A true validation of the predictive accuracy of this model involves data collection and construction of agricultural gross regional product in different pixels and testing those inde- pendent observations against the predicted values. The regional product data are generally constructed at the administrative level rather than the pixels, so validation would have to be done on an aggregation of model predictions. Few countries provide the required data to assess the prediction accuracy to examine the internal validation of the disaggregation efficiency and the data collection would be extremely costly and time-consuming. For the case of Brazil, Thomas et al. (2019) examine the predictive accuracy of three models to disaggregate agricultural GDP spatially including: cross-entropy, rural population and spa- tial regression. The cross-entropy and spatial regression models outperform a naive rural 20 population AgGDP model as measured by the Mean Absolute Deviation (MAD) and Root Mean Square Error (RMSE). 14 While the spatial regression performs the best, global data requirements that allow high enough degrees of freedom is a challenge. Given these data requirements and challenges, we compare the cross-entropy model to another spatial allocation model based on rural population at the country level. Then we extend a comparison of both models at the global level by mapping the correlation. One advantage of the cross-entropy is the volume preserving pycnophylactic property, which ensures the sum of the gridded data is the original value and allows the possibility to include all information that is available from mixed levels of data (e.g. You, Wood, et al., 2014). However, this presents a challenge in terms of an assessment of a global model. Previous work on gridded data products includes evaluations of accuracy. Typically, studies evaluate the internal accuracy of the model exploiting multiple geographic levels of data (as mentioned above in Thomas et al., 2019). Similarly, Van Boeckel et al. (2011), who examine duck data in Thailand, conclude that input levels do matter, especially the importance of the presence of administrative level 1 data. Robinson et al. (2014) evaluate the livestock model in Brazil and find a positive association between the model accuracy and the administrative level of the training data used in the model. They also illustrate this inverse relationship of prediction accuracy and level of intensification in the case of chickens in Europe (Robinson et al., 2014). At the local cell level, previous models of land or population have compared results to independent local data (Siebert, D¨ oll, and Hoogeveen, 2002) or identified errors of omission in a gridded population model using the locations from household surveys (e.g. Tiecke et al., 2017). Since we can not perform an evaluation of prediction accuracy for all countries, we com- pare the global cross-entropy model with another allocation model, which is similar to the global assessment of maize and rice production in You, Wood, et al. (2014). For the compar- ison, we construct a proportional allocation model using rural population density following the method in Thomas et al. (2019) for the case of Brazil. 15 Then, we can test the similarity of the two maps. Following Levine et al. (2009), we assume a normal distribution over the 2 million land pixels and perform a pairwise student t test to test the null hypothesis that both maps were identical. This test allows us to examine whether the mean difference in 14 Specifically, the MAD and RMSE for each model are respectively: the rural population density model (28,744 and 25,397), Cross-entropy spatial allocation (8,249 and 18,347) and Spatial disaggregation from a regression on agricultural production (7,214 and 16,673). 15 We use the 2010 Gridded Population of the World version 4 from Center for International Earth Sci- ence Information Network - CIESIN - Columbia University (2017) adjusted to the United Nation’s World Population Prospects followed by including the rural area defined by the Global Human Settlement grid for 2015: namely, “Rural cluster”, “Low Density Rural grid cell”, or “Very low density rural grid cell” (Pesaresi and Freire, 2019). 21 the corresponding pixel value from one map to another was greater than would be expected by chance alone. The t test statistic tell us that we can not reject the null hypothesis which provides some evidence of similarity between the two models using all the global pixels. Figure 10 displays three global maps: the two models and their Spearman correlation. 16 We exclude areas from the analysis with values that are less than 200,000. The correlation shows both areas of high and low correlation as the input of the models draws from the relationship of agriculture from productions values or a (rural) population perspective. The cross-entropy model can also propagate errors from the ancillary data that are inputs to the components. For the SPAM model, the CGIAR network held expert consultation and validation workshops according to each crop and subsequently incorporated their feedback with modifications of the priors used in the model (You, Wood, et al., 2014). The authors of Gridded Livestock Of the World (GLW) note regional differences in accuracy (i.e. RMSE values) are the result of the variation of production intensity and thus dependence on the initial conditions of the land upon which the prediction variables are mainly drawn (spatial agro-ecological variables)(Robinson et al., 2014). Lastly, the models integrate higher spatial resolution data to inform the spatial disaggregation procedures, which is subject to the MAUP (Openshaw, 1981). 3.2 Usage Notes We provide descriptive statistics of the data and modeling from a fitness-for-use per- spective (e.g. Leyk et al., 2019). The data are most appropriate for applications at global, continental and regional scales (You and Wood, 2006). Decisions regarding the use of this version over smaller spatial extents should be carefully considered in relation to the underly- ing assumptions and characteristics of a particular area. However, as the spatial refinement of ancillary data advances along with greater currency, coverage and representativeness, we expect validation possibilities to increase and inform a better understanding of the uncer- tainty and the associated fitness-for-use. Also, we intend to improve spatial and temporal coverage when it is feasible. The data disaggregation model from source to target level does impose spatial relation- ships and is subject to error (T. Li et al., 2007). The measurement of GDP is also challenging (Angrist, Goldberg, and Jolliffe, 2021), especially agricultural production (Carletto, Jolliffe, and Banerjee, 2015). The level of uncertainty associated with these results includes the the- matic, spatial and temporal accuracies. Below, we discuss these data and modeling issues in relation to two aspects: regional accounts and the regional components of AgGDP (mainly 16 The raster correlation in R performs a simple moving window correlation between two grids with a 3x3 pixel window. 22 Figure 10: A panel map of gridded Agricultural GDP circa 2010 from the Cross-Entropy model and from the rural population model (A) and its Spearman correlation in areas of AgGDP above or equal to 200,000 in the Cross-Entropy model (B). Source: Authors’ calculation (2022). 23 crop, fishery, forest, and livestock production values) that are priors in the cross-entropy model, and the outcome of the cross-entropy model. 3.2.1 Regional accounts We collect regional accounts by sector from various sources into a global database. The data are not balanced over time nor at the geographic level. The variation in the reference year of the regional accounts data influences the temporal balance of the database. This mismatch can influence the regional distribution of the agricultural GDP that may be dif- ferent than the target reference year of 2010. Given climate 17 and specifically rainfall is important input to crop and livestock production and may contribute to variation across years (Stanimirova et al., 2019; Y. Zhang et al., 2020), we attempt to reduce this source of error by averaging over multiple years when data are available similar to You, Wood, et al. (2014). However, this does not eliminate this mismatch. The availability of data varies when grouped by World Bank income (low or lower middle, upper middle and high income). The average absolute temporal difference (ATD) defined as the mean difference in years between the reference regional accounts and the target year (2010) is higher in the low and lower middle income group. Likewise, the mean deviation of the share of AgGDP by country over the year(s) is larger in low or lower middle compared to high income. The global regional account database includes national and subnational units at various administrative levels. 18 Following Robinson et al. (2014) in their assessment of Gridded Livestock Of the World (GLW) 2.0, we summarize the average spatial resolution (ASR) of the input regional data, which is the square root of the land area divided by the number of administrative units. We find that on average the ASR decreases from high to low income groups. 3.2.2 Components Another source of uncertainty is indirect temporal inaccuracy propagated from the input datasets of the components, which are modeled. We discuss all five components of agricul- tural GDP: crop, livestock, forest, fish and hunting. The SPAM model (You, Wood, et al., 2014) is a result of several gridded modeled datasets including rural population density from Global Rural-Urban Mapping Project (GRUMP) Alpha version (Balk et al., 2006). Like- wise, the Gridded Livestock of the World v2.0 includes rural population density in 2006 (GRUMP) along with other predictors such as precipitation (Hijmans et al., 2005) and a 17 For a discussion on climate yield factors see Block et al. (2008). 18 This also includes cases where administrative units at the same level are merged to match the geography of the regional accounts data. 24 modeled travel time to places with 50,000 inhabitants circa 2000 (A. Nelson, 2008). (An- derson et al., 2015) find variation in their examination of global data products of cropping systems models. For livestock, we transform the 5 major livestock into international values from livestock products (namely, meat, milk, eggs, honey and wool). The forest (non-wood products, wood-products) components relies on a remote sensing model to estimate forest loss. With regards to the non-timber values, limitation from the sources present two chal- lenges. The estimates use simple averages from the literature that accordingly assume a property of uniformity in the value of a hectare of forest as similar across the world and the aki, sample of forests with literature drawn for the study is representative of the world (Siikam¨ ´ Santiago-Avila, and Vail, 2015). The fishing model relies on proximity and association with ports or water bodies. 19 Finally, since we do not incorporate any information on hunting, the result is an even distribution across units and time. Another source of uncertainty is the geographic distribution of the components. Ideally, we would use subnational prices, however it was not feasible, and the results do not reflect this occurrence, including administrative units with higher variation of prices due to the heterogeneity of distinct urban and rural areas. 4 Conclusion Natural disasters impact both lives and livelihoods and a higher frequency and severity of disasters will likely increase in a changing climate. Socio-economic estimates at the local level inform disaster preparations of the exposure of physical assets and production to natural disasters and have implications for food security. Significant advancement in the spatial allocation of indicators has occurred in the past 10 years such as population (e.g Leyk et al., 2019). The advantages of gridded data as a common spatial unit of integration and the cross-entropy models are clear. These common units allow us to examine within-country characteristics, especially in the case of spatial data that do not conform with each other such as administrative boundaries and natural hazards to inform analyses with local estimates. We present a novel data set that disaggregates the national and regional accounts of the agriculture sector across areas as a result of a model where we use ancillary data including satellite data. This allows us to estimate especially in countries that have a relatively higher share of agricultural activity in the entire economy. Then, we examine the exposure of areas with at least one extreme drought during 2000 to 2009 to agricultural GDP, where nearly 1.2 billion people live, and find an estimated US$432 billion of agricultural GDP in 2010. 19 The freshwater case does not account for any variation, whereas the marine port locations incorporate variation on vessel holds. 25 These data are the result of data collection and collaboration across multiple entities to ensure the most current, widest coverage and creativity of the users to apply it. How- ever, persistent challenges to data collection remain, including limited geographic levels and temporal lag at low frequencies. Also, the reference year and spatial resolution of the local AgGDP estimates are limited to the contemporaneous availability of the economic statistics and components such as the crop production model. We often have to consider the fitness- for-use while considering the accuracy; the model has lower average spatial resolution in areas where we have little data, however these same areas may benefit from the availability of these estimates to inform policy. Predictions are dependent on the availability and quality of the training data on which the model is based and the modeling process is flexible to update individual countries as the data are available. In the near future, we hope to increase the currency and number of countries with subnational data as updates of the regional account data and models of the components upon which the model relies become available. 26 References Alcamo, Joseph, Martina Fl¨ orke, and Michael M¨ arker (2007). “Future long-term changes in global water resources driven by socio-economic and climatic changes”. In: Hydrological Sciences Journal 52.2, pages 247–275. Anderson, Weston, Liangzhi You, Stanley Wood, Ulrike Wood-Sichra, and Wenbin Wu (2015). “An analysis of methodological and spatial differences in global cropping sys- tems models and maps”. In: Global Ecology and Biogeography 24.2, pages 180–191. Andree, Bo Pieter Johannes, Andres Chamorro, Aart Kraay, Phoebe Spencer, and Dieter Wang (Sept. 2020). Predicting food crises. World Bank Policy Research Working Paper 9412. Washington, DC: World Bank. Angrist, Noam, Pinelopi Koujianou Goldberg, and Dean Jolliffe (2021). “Why Is Growth in Developing Countries So Hard to Measure?” In: Journal of Economic Perspectives 35.3, pages 215–42. Arnell, Nigel W (2003). “Effects of IPCC SRES* emissions scenarios on river runoff: a global perspective”. In: Hydrology and Earth System Sciences 7.5, pages 619–641. Authors’ calculation (2022). Balk, Deborah L, Uwe Deichmann, Greg Yetman, Francesca Pozzi, Simon I Hay, and Andy Nelson (2006). “Determining global population distribution: methods, applications and data”. In: Advances in parasitology 62, pages 119–156. Beguer´ıa, Santiago and Sergio M. Vicente-Serrano (2017). SPEI: Calculation of the Stan- dardised Precipitation-Evapotranspiration Index. R package version 1.7. url: https : //CRAN.R-project.org/package=SPEI. Bella, Kathleen P and Elena G Irwin (2002). “Spatially explicit micro-level modelling of land use change at the rural–urban interface”. In: Agricultural Economics 27.3, pages 217–232. Berg, Claudia N, Brian Blankespoor, and Harris Selod (2018). “Roads and rural development in Sub-Saharan Africa”. In: The Journal of Development Studies 54.5, pages 856–874. Blankespoor, Brian, Susmita Dasgupta, and Glenn-Marie Lange (2017). “Mangroves as a protection from storm surges in a changing climate”. In: Ambio 46.4, pages 478–491. Block, Paul J, Kenneth Strzepek, Mark W Rosegrant, and Xinshen Diao (2008). “Impacts of considering climate variability on investment decisions in Ethiopia”. In: Agricultural Economics 39.2, pages 171–181. Bruederle, Anna and Roland Hodler (2018). “Nighttime lights as a proxy for human devel- opment at the local level”. In: PloS one 13.9, e0202231. Bundervoet, Tom, Laban Maiyo, and Apurva Sanghi (Oct. 2015). Measuring National and Subnational Economic Growth in Africa from Outer Space, with an Application to Kenya 27 and Rwanda. World Bank Policy Research Working Paper 7461. Washington, DC: World Bank. Carletto, Calogero, Dean Jolliffe, and Raka Banerjee (2015). “From tragedy to renaissance: improving agricultural data for better policies”. In: The Journal of Development Studies 51.2, pages 133–148. Carr˜ao, Hugo, Gustavo Naumann, and Paulo Barbosa (2016). “Mapping global patterns of drought risk: An empirical framework based on sub-national estimates of hazard, exposure and vulnerability”. In: Global Environmental Change 39, pages 108–124. Center for International Earth Science Information Network - CIESIN - Columbia Univer- sity (2017). Gridded Population of the World, Version 4 (GPWv4): Population Density Adjusted to Match 2015 Revision UN WPP Country Totals, Revision 11. NASA Socioe- conomic Data and Applications Center (SEDAC). Chen, Xi and William D Nordhaus (2011). “Using luminosity data as a proxy for economic statistics”. In: Proceedings of the National Academy of Sciences 108.21, pages 8589–8594. Dabla-Norris, Ms Era, Ms Kalpana Kochhar, Mrs Nujin Suphaphiphat, Mr Frantisek Ricka, and Evridiki Tsounta (2015). Causes and consequences of income inequality: A global perspective. International monetary fund. Doll, Christopher NH, Jan-Peter Muller, and Jeremy G Morley (2006). “Mapping regional economic activity from night-time light satellite imagery”. In: Ecological Economics 57.1, pages 75–92. Eberenz, Samuel, Dario Stocker, Thomas R¨ osli, and David N Bresch (2020). “Exposure o¨ data for global physical risk assessment”. In: Earth System Science Data Discussions. Elhorst, J Paul and Dirk Strijker (2003). “Spatial developments of EU agriculture in the post-war period: The case of wheat and tobacco”. In: Agricultural Economics Review 4.389-2016-23407, pages 63–72. Elvidge, Christopher D, Edward H Erwin, Kimberly E Baugh, Daniel Ziskin, Benjamin T Tuttle, Tilottama Ghosh, and Paul C Sutton (2009). “Overview of DMSP nightime lights and future possibilities”. In: 2009 Joint Urban Remote Sensing Event. IEEE, pages 1–5. Eurostat (2013). Manual on Regional Accounts. Luxembourg. — (2018). Glossary:Livestock unit (LSU) - EUROSTAT Statistics Explained. url: https:// ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:Livestock_ unit_(LSU). Falkenmark, Malin (1986). “Fresh water: Time for a modified approach”. In: Ambio, pages 192– 200. 28 Falkenmark, Malin (2013). “Growing water scarcity in agriculture: future challenge to global water security”. In: Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 371.2002, page 20120410. FAO (2009). Fishery and aquaculture statistics 2009. url: https://www.fao.org/publications/ card/en/c/1030779b-3733-5f5d-b3e4-0e779b94e498/. — (2013). World Food and Agriculture Statistical yearbook. Rome. — (2015). Global Administrative Unit Layers (GAUL) Dataset. url: %7Bhttp://www.fao. org/geonetwork/srv/en/metadata.show%5C%3Fid=12691%7D. — (2016). FAOSTAT Database. Rome. url: http : / / faostat . fao . org / site / 291 / default.aspx. — (2018). Forests and poverty reduction. Rome. url: %7Bhttp://www.fao.org/forestry/ livelihoods/en/%7D. — (2019). World Food and Agriculture Statistical pocketbook. Rome. — (2020). FAOSTAT Database. Rome. url: http://www.fao.org/faostat. Friedl, Mark A, Damien Sulla-Menashe, Bin Tan, Annemarie Schneider, Navin Ramankutty, Adam Sibley, and Xiaoman Huang (2010). “MODIS Collection 5 global land cover: Al- gorithm refinements and characterization of new datasets”. In: Remote sensing of Envi- ronment 114.1, pages 168–182. Ghosh, Tilottama, Rebecca L Powell, Christopher D Elvidge, Kimberly E Baugh, Paul C Sutton, and Sharolyn Anderson (2010). “Shedding light on the global distribution of economic activity”. In: The Open Geography Journal 3.1. Gibson, John, Susan Olivia, Geua Boe-Gibson, and Chao Li (2021). “Which night lights data should we use in economics, and where?” In: Journal of Development Economics 149, page 102602. Gilbert, Marius, Ga¨ elle Nicolas, Giusepina Cinardi, Thomas P Van Boeckel, Sophie O Van- wambeke, GR William Wint, and Timothy P Robinson (2018). “Global distribution data for cattle, buffaloes, horses, sheep, goats, pigs, chickens and ducks in 2010”. In: Scientific data 5.1, pages 1–11. Goldblatt, Ran, Kilian Heilmann, and Yonatan Vaizman (Dec. 2019). Can medium-resolution satellite imagery measure economic activity at small geographies? Evidence from Landsat in Vietnam. World Bank Policy Research Working Paper 9088. Washington, DC: World Bank. Gollin, Douglas, David Lagakos, and Michael E Waugh (2014). “The agricultural productiv- ity gap”. In: The Quarterly Journal of Economics 129.2, pages 939–993. Gunasekera, R, J Daniell, A Pomonis, RAD Arias, O Ishizawa, and H Stone (2018). “Method- ology Note on the Global RApid post-disaster Damage Estimation (GRADE) approach”. 29 In: World Bank and GFDRR Technical Report, World Bank and GFDRR, Washington, DC. Gunasekera, Rashmin, Oscar Ishizawa, Christoph Aubrecht, Brian Blankespoor, Siobhan Murray, Antonios Pomonis, and James Daniell (2015). “Developing an adaptive global exposure model to support the generation of country disaster risk profiles”. In: Earth- Science Reviews 150, pages 594–608. Harris, Ian, Timothy J Osborn, Phil Jones, and David Lister (2020). “Version 4 of the CRU TS monthly high-resolution gridded multivariate climate dataset”. In: Scientific data 7.1, pages 1–18. Henderson, J Vernon, Adam Storeygard, and David N Weil (2012). “Measuring economic growth from outer space”. In: American economic review 102.2, pages 994–1028. Herrero, Mario and Philip K Thornton (2013). “Livestock and global change: emerging issues for sustainable food systems”. In: Proceedings of the National Academy of Sciences 110.52, pages 20878–20881. Hijmans, Robert J, Susan E Cameron, Juan L Parra, Peter G Jones, and Andy Jarvis (2005). “Very high resolution interpolated climate surfaces for global land areas”. In: In- ternational Journal of Climatology: A Journal of the Royal Meteorological Society 25.15, pages 1965–1978. Hosch, Gilles, Bradley Soule, Max Schofield, Trevor Thomas, Charles Kilgour, and Tim Huntington (2019). “Any Port in a Storm: Vessel Activity and the Risk of IUU-Caught Fish Passing through the World’s Most Important Fishing Ports”. In: Journal of Ocean and Coastal Economics 6.1, page 1. Hossain, Mohammed Kamal, M Khairul Alam, Md Danesh Miah, et al. (2008). “Forest restoration and rehabilitation in Bangladesh”. In: Keep Asia Green 3, pages 21–66. Kelleher, Kieran, Rolf Willmann, and Ragnar Arnason (2009). The sunken billions: the eco- nomic justification for fisheries reform. The World Bank and FAO. Kummu, Matti, Maija Taka, and Joseph HA Guillaume (2018). “Gridded global datasets for gross domestic product and Human Development Index over 1990–2015”. In: Scientific data 5, page 180004. Kummu, Matti, Philip J Ward, Hans de Moel, and Olli Varis (2010). “Is physical water scarcity a new phenomenon? Global assessment of water shortage over the last two mil- lennia”. In: Environmental Research Letters 5.3, page 034006. Lamarche, C´ eline, Maurizio Santoro, Sophie Bontemps, Rapha¨ el d’Andrimont, Julien Radoux, Laura Giustarini, Carsten Brockmann, Jan Wevers, Pierre Defourny, and Olivier Arino (2017). “Compilation and validation of SAR and optical data products for a complete 30 and global map of inland/ocean water tailored to the climate modeling community”. In: Remote Sensing 9.1, page 36. Lebedys, Arvydas and Li Yanshu (2014). “Contribution of the forestry sector to national economies, 1990-2011”. In: Forest Finance Working Paper (FAO) eng no. 09. Lehmann, Robert and Klaus Wohlrabe (2015). “Forecasting GDP at the regional level with many predictors”. In: German Economic Review 16.2, pages 226–254. Levine, Rebecca S, Krista L Yorita, Matthew C Walsh, and Mary G Reynolds (2009). “A method for statistically comparing spatial distribution maps”. In: International Journal of Health Geographics 8.1, pages 1–7. Leyk, Stefan, Andrea E Gaughan, Susana B Adamo, Alex de Sherbinin, Deborah Balk, Sergio Freire, Amy Rose, Forrest R Stevens, Brian Blankespoor, Charlie Frye, Joshua Comenetz, Alessandro Sorichetta, Kytt MacManus, Linda Pistolesi, Marc Levy, Andrew J. Tatem, and Martino Pesaresi (2019). “The spatial allocation of population: A review of large-scale gridded population data products and their fitness for use”. In: Earth System Science Data 11.3. Li, Tiebei, David Pullar, Jonathan Corcoran, and Robert Stimson (2007). “A comparison of spatial disaggregation techniques as applied to population estimation for South East Queensland (SEQ), Australia”. In: Applied GIS 3.9, pages 1–16. Luijten, JC (2003). “A systematic method for generating land use patterns using stochastic rules and basic landscape characteristics: results for a Colombian hillside watershed”. In: Agriculture, ecosystems & environment 95.2-3, pages 427–441. Mathiesen, A.` M. (2018). “Fisheries: feeding humanity in 2030. Conference presentation at Our Ocean 2018”. In: url: %7Bhttp : / / www . fao . org / fi / static - media / ADG / MathiesenOurOceanConference2018.pdf%7D. Murakami, Daisuke and Yoshiki Yamagata (2019). “Estimation of gridded population and GDP scenarios with spatially explicit statistical downscaling”. In: Sustainability 11.7, page 2106. Murthy, CS, B Laxman, and MVR Sesha Sai (2015). “Geospatial analysis of agricultural drought vulnerability using a composite index based on exposure, sensitivity and adaptive capacity”. In: International journal of disaster risk reduction 12, pages 163–171. NASA (2018). MODIS Collection 6 NRT Hotspot / Active Fire Detections MCD14DL. doi: DOI:10.5067/FIRMS/MODIS/MCD14DL.NRT.006. url: https://earthdata.nasa.gov/ firms. National Geospatial-Intelligence Agency (2019). World Port Index. Nelson, Andrew (2008). “Travel time to major cities: A global map of Accessibility”. In: Ispra: European Commission. 31 Nelson, Gerald C (2002). “Introduction to the special issue on spatial analysis for agricultural economists”. In: Agricultural Economics 27.3, pages 197–200. NOAA (2011). Version 4 DMSP-OLS Global Radiance Calibrated Nighttime Lights. url: %7Bhttps://ngdc.noaa.gov/eog/dmsp/download/radcal.html%7D. Nordhaus, William D (2006). “Geography and macroeconomics: New data and new findings”. In: Proceedings of the National Academy of Sciences 103.10, pages 3510–3517. Openshaw, Stan (1981). “The modifiable areal unit problem”. In: Quantitative geography: A British view, pages 60–69. Pesaresi, Martino and Sergio Freire (2019). GHS-SMOD-GHS settlement grid. Pratesi, Monica, Nicola Salvati, Caterina Giusti, and Stefano Marchetti (2015). Spatial disag- gregation and small-area estimation methods for agricultural surveys: solutions and per- spectives. Technical Report Series GO-07-2015. FAO Global Office of the Global Strategy. Reddy, Tamma Koti and Madhubanti Dutta (2018). “Impact of Agricultural Inputs on Agri- cultural GDP in Indian Economy”. In: Theoretical Economics Letters 8.10, pages 1840– 1853. Rentschler, Jun and Melda Salhab (Oct. 2020). People in harm’s way: Flood exposure and poverty in 189 countries. World Bank Policy Research Working Paper 9447. Washington, DC: World Bank. Roberts, Mark, Brian Blankespoor, Chandan Deuskar, and Benjamin Stewart (Mar. 2017). Urbanization and development: Is latin america and the caribbean different from the rest of the world? World Bank Policy Research Working Paper 8019. Washington, DC: World Bank. Robinson, Timothy P, GR William Wint, Giulia Conchedda, Thomas P Van Boeckel, Valentina Ercoli, Elisa Palamara, Giuseppina Cinardi, Laura D’Aietti, Simon I Hay, and Marius Gilbert (2014). “Mapping the global distribution of livestock”. In: PloS one 9.5, e96084. Rubinstein, Reuven Y and Dirk P Kroese (2004). The cross-entropy method: a unified ap- proach to combinatorial optimization, Monte-Carlo simulation, and machine learning. Volume 133. Springer. Samberg, Leah H, James S Gerber, Navin Ramankutty, Mario Herrero, and Paul C West (2016). “Subnational distribution of average farm size and smallholder contributions to global food production”. In: Environmental Research Letters 11.12, page 124010. Siebert, Stefan, Petra D¨ oll, and J Hoogeveen (2002). “Global map of irrigated areas ver- sion 2.1”. In: Center for Environmental Systems Research, University of Kassel, Ger- many/Food and Agriculture Organization of the United Nations, Rome, Italy. Siikam¨ ´ aki, J., F. J. Santiago-Avila, and P. Vail (2015). Global Assessment of Non’Wood Forest Ecosystem Services. Working Paper. Washington, DC.: Resources for the Future. 32 Sorrenti, Simona (2016). “Non-wood forest products in international statistical systems”. In: Staal, Steven J, Isabelle Baltenweck, MM Waithaka, T DeWolff, and L Njoroge (2002). “Location and uptake: integrated household and GIS analysis of technology adoption and land use, with application to smallholder dairy farms in Kenya”. In: Agricultural Economics 27.3, pages 295–315. Stanimirova, Radost, Paulo Ar´ evalo, Robert K Kaufmann, Victor Maus, Myroslava Lesiv, Petr Havlık, and Mark A Friedl (2019). “Sensitivity of global pasturelands to climate variation”. In: Earth’s Future 7.12, pages 1353–1366. Thomas, Timothy S, Liangzhi You, Ulrike Wood-Sichra, Yating Ru, Brian Blankespoor, and Erwin Kalvelagen (Aug. 2019). Generating Gridded Agricultural Gross Domestic Product for Brazil: A Comparison of Methodologies. World Bank Policy Research Working Paper WPS 8985. Wasington, D.C.: World Bank. Tiecke, Tobias G, Xianming Liu, Amy Zhang, Andreas Gros, Nan Li, Gregory Yetman, Talip Kilic, Siobhan Murray, Brian Blankespoor, Espen B Prydz, et al. (2017). “Mapping the world population one building at a time”. In: arXiv preprint arXiv:1712.05839. Tobler, Waldo R (1979). “Smooth pycnophylactic interpolation for geographical regions”. In: Journal of the American Statistical Association 74.367, pages 519–530. UNISDR (2011). Global Assessment Report on Disaster Risk Reduction 2011: Revealing Risk, Redefining Development. Geneva, Switzerland: United Nations International Strategy for Disaster Reduction. Van Boeckel, Thomas P, Diann Prosser, Gianluca Franceschini, Chandra Biradar, William Wint, Tim Robinson, and Marius Gilbert (2011). “Modelling the distribution of domestic ducks in Monsoon Asia”. In: Agriculture, ecosystems & environment 141.3-4, pages 373– 380. Veldkamp, Ted I.E., Stephanie Eisner, Yoshihide Wada, Jeroen C.J.H. Aerts, and Philip J. Ward (2015). “Sensitivity of water scarcity events to ENSO-driven climate variability at the global scale”. In: Hydrology and Earth System Sciences 19.10, pages 4081–4098. Vesco, Paola, Matija Kovacic, Malcolm Mistry, and Mihai Croicu (2021). “Climate vari- ability, crop and conflict: Exploring the impacts of spatial concentration in agricultural production”. In: Journal of Peace Research 58.1, pages 98–113. Vicente-Serrano, Sergio M, Santiago Beguerıa, and Juan I L´ opez-Moreno (2010). “A multi- scalar drought index sensitive to global warming: the standardized precipitation evapo- transpiration index”. In: Journal of climate 23.7, pages 1696–1718. Wang, Xuantong, Paul C Sutton, and Bingxin Qi (2019). “Global mapping of GDP at 1 km2 using VIIRS nighttime satellite imagery”. In: ISPRS International Journal of Geo- Information 8.12, page 580. 33 Ward, Philip J, Veit Blauhut, Nadia Bloemendaal, James E Daniell, Marleen C de Ruiter, Melanie J Duncan, Robert Emberson, Susanna F Jenkins, Dalia Kirschbaum, Michael Kunz, et al. (2020). “Natural hazard risk assessments at the global scale”. In: Natural Hazards and Earth System Sciences 20.4, pages 1069–1096. Wood, Stanley, Kate Sebastian, Freddy Nachtergaele, Daniel Nielsen, and Aiguo Dai (1999). Spatial aspects of the design and targeting of agricultural development strategies. Environ- ment and Production Technology Division Discussion Paper No. 44. Washington, DC.: International Food Policy Research Institute. World Bank (2019). World Development Indicators Database (World Bank). World Bank and UNEP (2011). Gross Domestic Product 2010. Washington, D.C. and Geneva. url: https://preview.grid.unep.ch/. WorldPop and Center for International Earth Science Information Network (CIESIN), Columbia University (2018). Global High Resolution Population Denominators Project - Funded by The Bill and Melinda Gates Foundation (OPP1134076). doi: https://dx.doi.org/10. 5258/SOTON/WP00647. You, Liangzhi and Stanley Wood (2003). Spatial allocation of agricultural production using a cross-entropy approach. Environment and Production Technology Division Discussion Paper No. 126. Washington, DC.: International Food Policy Research Institute. — (2006). “An entropy approach to spatial disaggregation of agricultural production”. In: Agricultural Systems 90.1-3, pages 329–347. You, Liangzhi, Stanley Wood, Ulrike Wood-Sichra, and Wenbin Wu (2014). “Generating global crop distribution maps: From census to grid”. In: Agricultural Systems 127, pages 53– 60. You, Liangzhi, U. Wood-Sichra, S. Fritz, Z. Guo, L. See, and J. Koo (2018). “Spatial pro- duction allocation model (SPAM) 2010 Version 1.0”. In: IFPRI, Washington DC. url: http://MapSPAM.info.. Yu, Qiangyi, Liangzhi You, Ulrike Wood-Sichra, Yating Ru, Alison KB Joglekar, Steffen Fritz, Wei Xiong, Miao Lu, Wenbin Wu, and Peng Yang (2020). “A cultivated planet in 2010–Part 2: the global gridded agricultural-production maps”. In: Earth System Science Data 12.4, pages 3545–3572. Zhang, Ying, Liangzhi You, Donghoon Lee, and Paul Block (2020). “Integrating climate pre- diction and regionalization into an agro-economic model to guide agricultural planning”. In: Climatic Change 158.3, pages 435–451. 34 5 Tables Table 3: Top 10 countries of largest total Agricultural GDP (millions of USD) exposed to dry areas with share of Agricultural GDP and Population (thousands) Rank Country Ag GDP Share of Ag GDP Pop (2010) 1 China 146,000 0.26 323,000 2 India 60,600 0.22 255,000 3 United States 21,800 0.14 69,100 4 Russian Federation 14,300 0.26 27,100 5 Iran, Islamic Rep. 13,400 0.44 40,600 6 Brazil 12,600 0.14 9,230 6 Pakistan 12,600 0.28 42,600 7 Australia 10,900 0.44 6,130 8 Italy 6,560 0.17 7,120 9 Canada 5,540 0.25 5,000 Table 4: Top 10 countries of largest share of Agricultural GDP exposed to dry areas with Agricultural GDP (millions of USD) and Population (thousands) Rank Country Share of Ag GDP Ag GDP Pop (2010) 1 Rwanda 1.00 1670 9850 1 Saint Vincent and the Grenadines 1.00 11.6 29.9 1 Micronesia, Federated States of 1.00 <1 <1 2 Burundi 0.97 732 8320 3 Brunei Darussalam 0.91 99.3 92.8 4 West Bank and Gaza 0.85 543 2770 5 Gambia, The 0.81 170 1420 6 Finland 0.79 4400 3950 7 Belize 0.79 126 208 8 Jordan 0.73 733 5400 35 Table 5: Top 10 countries of 2010 population (thousands) exposed to dry areas with Agricultural GDP (millions of USD) and share of Agricultural GDP Rank Country Pop (2010) Ag GDP Share of Ag GDP 1 China 323,000 146,000 0.26 2 India 255,000 60,600 0.22 3 United States 69,100 21,800 0.14 4 Congo, Dem. Rep. 45,100 2,780 0.59 5 Pakistan 42,600 12,600 0.28 6 Iran, Islamic Rep. 40,600 13,400 0.44 7 Russian Federation 27,100 14,300 0.26 8 Tanzania 23,200 4,140 0.55 9 Uganda 18,700 2,990 0.66 10 Thailand 17,400 4,930 0.15 Table 6: Top 10 countries of largest total Agricultural GDP (millions of USD) with Population (thousands) and share of Ag GDP exposed to WCI areas with Agricultural GDP (million of USD) and Population (thousands) Rank Country Ag GDP Share of Ag GDP Pop (2010) 1 China 400,000 0.70 950,000 2 India 233,000 0.87 990,000 3 Pakistan 44,400 1.00 170,000 4 Indonesia 42,000 0.46 140,000 5 Nigeria 32,300 0.38 71,000 6 Egypt, Arab Rep. 25,800 0.94 76,000 7 Italy 17,800 0.47 24,000 8 Korea, Rep. 17,400 0.95 37,000 9 Bangladesh 17,000 0.91 120,000 10 Iran, Islamic Rep. 16,300 0.54 50,000 36 Table 7: Top 10 countries of largest share of Agricultural GDP (millions of USD) by total Agricultural GDP and Population (thousands) exposed to WCI areas.20 Rank Country Share of Ag GDP Ag GDP Pop (2010) 1 Sri Lanka 1.00 4,760 15,000 1 Israel 1.00 3,520 6,800 1 Dominican Republic 1.00 2,990 7,100 1 Yemen, Rep. 1.00 2,550 22,000 1 United Arab Emirates 1.00 1,730 4,700 1 Lebanon 1.00 1,390 4,700 1 Haiti 1.00 1,380 9,000 1 Jordan 1.00 996 5,800 1 Oman 1.00 715 2,700 1 Puerto Rico (US) 1.00 652 3,300 Table 8: Top 10 countries of 2010 population (thousands) by Agricultural GDP (millions of USD) and share of Ag GDP exposed to WCI areas Rank Country Pop (2010) Ag GDP Share of Ag GDP 1 India 990,000 233,000 0.87 2 China 950,000 400,000 0.70 3 Pakistan 170,000 44,400 1.00 4 Indonesia 140,000 42,000 0.46 5 Bangladesh 120,000 17,000 0.91 6 Egypt, Arab Rep. 76,000 25,800 0.94 7 Nigeria 71,000 32,300 0.38 8 Mexico 56,000 10,400 0.32 9 United Kingdom 51,000 11,300 0.81 9 Germany 51,000 10,700 0.45 20 Additional countries exposed to WCI area with the 1.00 share of Ag GDP include: West Bank and Gaza; Cyprus; Kuwait; Gambia, The; Qatar; Hong Kong (SAR, China) 37 AppendixA Tables 38 Table A1: Regional account descriptive statistics and source Country First Last Number Source year year of regions Albania 2012 2014 12 EUROSTAT Argentina 2004 2004 24 Instituto Nacional de Estad´ ıstica y Censos Australia 2009 2011 8 Australian Bureau of Statistics Austria 2012 2014 9 EUROSTAT Belarus 2011 2013 8 BELSTAT Belgium 2012 2014 3 EUROSTAT Bolivia 2009 2011 9 Instituto Nacional de Estad´ıstica Brazil 2010 2012 31 Instituto Brasileiro de Geografia e Estat´ıstica Bulgaria 2012 2014 2 EUROSTAT Canada 2009 2011 13 Statistics Canada Chile 2013 2015 13 Banco Central De Chile China 2009 2011 32 National Bureau of Statistics China Colombia 2009 2011 32 Departamento Administrativo Na- cional de Estad´ ıstica Croatia 2012 2014 3 EUROSTAT Czech Republic 2012 2014 7 EUROSTAT Denmark 2012 2014 5 EUROSTAT Ecuador 2006 2006 23 Banco Central De Ecuador Estonia 2012 2014 5 EUROSTAT Finland 2012 2014 2 EUROSTAT France 2012 2014 22 EUROSTAT Georgia 2009 2011 9 National Statistics Office of Georgia Germany 2012 2014 16 EUROSTAT Greece 2012 2014 13 EUROSTAT Hungary 2012 2014 3 EUROSTAT India 2011 2013 32 Central Statistics Office Indonesia 2009 2011 31 INDO-DAPOER Iran, Islamic Rep. 2014 2014 28 Iran Statistical Yearbook 1389 Ireland 2012 2014 2 EUROSTAT Continued on next page 39 Table A1: Regional account descriptive statistics and source Country First Last Number Source year year of regions Italy 2012 2014 20 EUROSTAT Japan 2009 2011 47 Cabinet Office Government of Japan Kazakhstan 2010 2012 15 Agency of Statistics of the Republic of Kazakhstan Kenya 2017 2017 48 Kenya National Bureau of Statistics and World Bank Korea, Rep. 2009 2011 15 Korean Statistical Information Ser- vices Latvia 2012 2014 6 EUROSTAT Lithuania 2012 2014 10 EUROSTAT Malaysia 2010 2012 16 Department of Statistics Malaysia Mali 2009 2009 9 Cellule d’Analyse et de Prospective Malta 2012 2014 2 EUROSTAT Mexico 2009 2011 32 Instituto Nacional de Estad´ ıstica y Geograf´ ıa Mongolia 2015 2017 23 Mongolian Statistical Information Service Morocco 2005 2007 7 Ministry of Finance Nepal 2019 2019 7 Central Bureau of Statistics Nepal Netherlands 2012 2014 12 EUROSTAT New Zealand 2009 2011 14 Statistics New Zealand North Macedonia 2012 2014 8 EUROSTAT Norway 2012 2014 19 EUROSTAT Panama 2009 2011 9 Instituto Nacional de Estad´ ıstica y Censo Peru 2009 2011 25 Instituto Nacional de Estadistica e informatica Philippines 2009 2011 17 Philippine Statistics Authority Poland 2012 2014 15 EUROSTAT Romania 2012 2014 4 EUROSTAT Russian Federa- 2009 2011 82 Mordoviastat: Federal Service of tion State Statistics Continued on next page 40 Table A1: Regional account descriptive statistics and source Country First Last Number Source year year of regions Slovak Republic 2012 2014 4 EUROSTAT Slovenia 2012 2014 2 EUROSTAT South Africa 2009 2011 9 Statistics South Africa Spain 2012 2014 19 EUROSTAT Sri Lanka 2009 2011 9 Economic and Social Statistics of Sri Lanka Sweden 2012 2014 3 EUROSTAT Switzerland 2009 2011 25 Federal Statistical Office of Switzer- land Thailand 2009 2011 76 Office of the National Economic and Social Development Board urkiye T¨ 2009 2011 81 Turkish Statistical Institute Ukraine 2010 2012 25 State Statistics Service of Ukraine United Kingdom 2012 2014 4 EUROSTAT United States 2009 2011 51 Bureau of Economic Analysis Uruguay 2008 2008 19 Instituto Nacional de Estadistica Vietnam 2009 2011 64 General Statistics Office of Viet Nam Zambia 2015 2015 9 Central Statistics Office of Zambia 41