## REPUBLIC OF UZBEKISTAN — Report on National Accounts Mission (July 24–August 4, 2023) ## Source details **Canonical URL:** [REPUBLIC OF UZBEKISTAN — Report on National Accounts Mission (July 24–August 4, 2023)](https://www.imf.org/-/media/files/publications/tar/2024/english/tarea2024070-print-pdf.pdf) ## Other formats - [Markdown version](/-/media/files/publications/tar/2024/english/tarea2024070-print-pdf.pdf.md) - [Structured JSON version](/-/media/files/publications/tar/2024/english/tarea2024070-print-pdf.pdf.json) --- ### Summary of Mission Outcomes and Priority Recommendations - Mission duration: July 24 to August 4, 2023. - Purpose: Assist the Statistics Agency under the President of the Republic of Uzbekistan (SA) in developing annual supply and use tables (SUT) and improving non-observed economy (NOE) estimates. - Key implementation: - Implemented the Python version of the balancing tool SUTB developed by IMF staff. - Compiled a symmetric 2021 SUT with 144 activity types and products. - Expanded the SUTB file to include 144 columns and rows of the output and intermediate consumption matrices and applied the SUTB balancing procedure after data preparation and analysis of statistical discrepancies by product. - NOE survey: - SA plans a sector-specific NOE survey covering transport, food and accommodation, education, and health care. - Stratified sample: 7,000 households. - Pilot: 16 households interviewed (mostly SA employees); results transferred from tablets to MS Access database; mission and SA created queries and compiled weighted household expenditures for questionnaire items. - Field work scheduled: start on August 7 and end on August 25. - Intended use: NOE survey results to be used in compilation of national accounts in 2025. - Recommendation on SUT timing: SUTs should be compiled before publication of annual GDP so SUT can be used to determine balanced GDP levels and allow changes to output and intermediate consumption during balancing. - Priority Recommendations (extract): - December 2023: Conduct logical control of the initial results of the NOE survey. — Responsible: SA - April 2024: Develop the NOE survey database and prepare the main results. — Responsible: SA - April 2024: Use the energy balance data for the SUT compilation. - April 2024: Improve accounting of goods produced for own consumption and adjust output of manufacturing (wood, meat, and milk processing, etc.). - August 2024: Improve data sources for the distribution of gross fixed capital formation by type of asset. - December 2024: Prepare and manually balance the SUT at current 2022 prices then remove final, minor imbalances by using the SUTB balancing tool. — Responsible: SA - April 2025: Use the results of the NOE survey in compiling GDP. - Priority action classification: H – High; M – Medium. ### Detailed Technical Assessment and Recommendations — Priority Action / Milestones - Target Completion Dates and High-priority actions: - December 2023: Conduct logical control of the initial results of the NOE survey. (H) - April 2024: Develop the NOE survey database and prepare the main results. (H) - April 2024: Use the energy balance data for the SUT compilation. (H) - April 2024: Improve accounting of goods produced for own consumption and adjust output of manufacturing (wood, meat, and milk processing, etc.). (H) - April 2024: Collect price indices for selected SUT indicators. (M) - April 2024: Use the tourism satellite accounts data in compiling SUTs. (H) - August 2024: Improve data sources for the distribution of gross fixed capital formation by type of asset. (H) - December 2024: Prepare and manually balance the SUT at current 2022 prices then remove final, minor imbalances by using the SUTB balancing tool. (H) - April 2025: Use the results of the NOE survey in compiling GDP. (H) ### A. Non-Observed Economy (NOE) — Findings and Recommended Actions Findings: - Planned NOE survey: sector-specific, covering transport, food and accommodation, education, and health care. - Sample design: - Total sample: 7,000 households. - Population base: approximately 7 million households. - Stratification: 27 strata (Tashkent city and 13 regions, each divided into urban and rural settlements). - Next geographical level: mohalla; total mohallas: 9,000; selected mohallas: 309. - From each mohalla: 22 or 23 households interviewed; reserve list includes another 23 households for refusals. - Data collection: - Electronic questionnaires on tablets; results transferred to MS Access database maintained by Department of National Accounts. - Pilot test: - 16 households interviewed (mostly SA employees). - Mission and SA created MS Access queries and calculated weighted household expenditures by questionnaire item to assist full-database processing. - Pilot issues identified: - Questionnaire tables mix monthly and annual expense items; respondents may report annual instead of monthly values or vice versa. - Expenses must be entered in 1000 units of national currency; this was sometimes omitted by respondents. - Some expense items (e.g., restaurants) may require interviewing all adults; multiple household visits may be necessary. - Field work schedule: start August 7 and end August 25. Recommended Actions: - Conduct logical control of the initial results of the NOE survey. - Develop the NOE survey database and prepare the main results. - Use the results of the NOE survey in compiling GDP. ### B. Supply and Use Tables (SUT) Balancing Tool SUTB — Findings and Guidance: - Background: Prior SUTB implementation relied on a MATLAB-based mechanism requiring 64-bit Excel 2016 on one computer; risked obsolescence and hardware dependency. - Current mission: Implemented the SUTB balancing tool based on Python, deployable on any modern Excel-enabled computer. - SUTB function requirements (five arguments): - Unbalanced SUT, - Three sets of accounting constraints (grouped by type: two accounting-type constraints — supply-use and netting adjustments — and user-defined explicit constraints), - A set of reliability coefficients. - Reliability coefficients: - Scale from 0 to 100, where 100 represents fully reliable data not subject to adjustments. - Because SUTs were compiled after annual GDP publication, some indicators have reliability coefficients equal to 100 percent; mission recommends compiling SUTs before annual GDP publication to allow adjustments to output and intermediate consumption. - Application: - 2021 SUT was balanced using the Python SUTB. - SUT for Uzbekistan is symmetrical and includes 144 types of activities and products; SUTB file expanded to 144 rows and columns for output and intermediate consumption. The 2021 SUT — Product-level discrepancies, diagnostic actions, and adjustments: - Products with significant supply-use discrepancies (listed with NACE Rev.2 codes where provided): - Extraction of crude petroleum (NACE Rev.2 - 06.1) - Other food products except sugar (NACE Rev.2 - 10.8 except 10.81) - Sugar (NACE Rev.2 - 10.81) - Forestry and logging (NACE Rev.2 - 02) - Passenger rail transport, interurban (NACE Rev.2 - 49.1) - Dairy products (NACE Rev.2 - 10.5) - Plastics in primary forms (NACE Rev.2 - 20.16) - Crude petroleum: - Extraction and import data based on regular surveys; structure of intermediate consumption from 2019 detailed survey, making intermediate consumption relatively less reliable. - Main data source for crude petroleum is the energy balance; SA published an experimental energy balance for 2021 with volume indicators of output, changes in inventories, and consumption. - Mission used energy balance data to compile intermediate consumption of crude oil, which was about 40 percent higher than the preliminary estimate; after adjustment, the discrepancy between supply and use of crude petroleum was reduced. - Because GVA for crude petroleum had already been published and could not be changed, intermediate consumption was revised using a special technique: other products used in manufacturing of petroleum products were reduced by the same amount and redistributed proportionally; final result: intermediate consumption for manufacturing of oil products did not change in total but its structure shifted from crude oil to other products. - Sugar: - Two large sugar processing enterprises process raw sugar; raw sugar classified as two types: (i) sugar cane (agricultural product, NACE Rev.2 - 01.14) and (ii) processed sugar cane supplied to sugar refinery (10.81). - SA had classified processed sugar cane as NACE Rev.2 - 01.6 due to the 2019 intermediate consumption survey questionnaire containing only one question on sugar cane coded 01.6, producing inaccurate intermediate consumption structure. - Mission recommended reclassifying processed sugar cane to NACE Rev.2 - 10.81; SA implemented this change, significantly reducing the discrepancy between sugar supply and use. - Forestry and wood products: - Supply of forestry products far exceeded their use; detailed analysis found output of roundwood (NACE Rev.2 – 02.2) was almost twice the output of manufacture of wood products and paper (NACE Rev.2 – 16 and 17). - More than half of forestry production is produced by households and mainly processed for own-account construction of residential buildings. - Own-account production chain (from production through own-account processing and use in own-account construction) needs full accounting: include own-account production of wood in industrial production index and national accounts. - Current status: own-account production of agriculture and food industry is included in industrial production index; own-account production of wood is not considered; output of food industry does not include own-account production of meat. - Recommendation: Value household processing of wood and meat at producer prices and add to output of these industries. - Passenger rail transport: - Output of passenger rail transport was much less than use. - Analysis found output, intermediate consumption, household consumption, and exports of services corresponded to 2021 situation, but imports of services were much lower than in previous years. - SA compiles exports and imports of services from two sources: (i) foreign economic activity survey (corporations) and (ii) tourism research (visitors and local households). - Mission judged tourism survey data more reliable than foreign trade surveys for rail passenger services; after changing the data source, imports of services increased and the difference between supply and use reduced significantly. ### Discrepancies in sectoral accounts: dairy and plastics - Dairy products: - Final consumption of dairy products was almost twice the output at basic prices. - Main reason: dairy products produced by households for own consumption, which are not fully accounted for when compiling output. - By considering household-produced dairy for own consumption, the statistical discrepancy will be significantly reduced. - Plastics in primary forms: - Resources far exceeded their use. - Main reason: underestimation of intermediate consumption. - Imports in 2021 significantly exceed intermediate consumption, with these products almost entirely used in the production process as intermediate consumption. - When imports are largely used as intermediate consumption, it is necessary to change the output or intermediate consumption estimates; if these estimates are fixed, balancing is particularly difficult. - This is cited as a specific example of why it is better if the SUT is compiled before publishing the annual GDP. ### SUT compilation and publication timetable - The 2021 SUT will be published in December 2023. - The preliminary 2022 SUT at current prices will be compiled at the end of October 2023. - The final 2022 SUT will be published in December 2024. - In 2024 the SA also plans to compile the SUT at the previous year’s prices. ### Recommended actions for SUT compilation and balancing - Prepare and manually balance the SUT at current 2022 prices then remove final, minor imbalances by using the SUTB balancing tool. - Use the energy balance data for the SUT compilation. - Improve the accounting of goods produced for own consumption and adjust the output of manufacturing (wood, meat, and milk processing, etc.). - Use the tourism satellite accounts data in compiling SUTs. - Improve data sources for the distribution of gross fixed capital formation by type of asset. - Collect price indices for selected SUT indicators. ### Officials met during the mission - Mr. Agzam Ikramov — Deputy Director of SA - Mr. Abrorali Mamadjanov — Head of the Macroeconomic Indicators and National Accounts Department (NAD), SA - Mr. Odiljon Mamadaliev — Head of the International Cooperation Department, SA - Mr. Sardor Juraev — Head of the Industrial Statistics Department, SA - Mr. Abdugafir Mirzakhidov — Head of the Service Statistics Department, SA - Mr. Akrom Sultanov — Head of the Coordination and Monitoring Department, SA - Ms. Zukhra Mamadjonova — Head of the Input-Output Statistics Division, NAD, SA - Mr. Umarjon Obidov — Deputy Head of the NAD, SA - Mr. Rakhmatjon Urusbayev — Head of the Satellite Accounts Division, NAD, SA - Mr. Farkhod Ganiev — Deputy Head of the Satellite Accounts Division, NAD, SA - Ms. Tatyana Kuksina — Chief Specialist, NAD, SA - Ms. Zioda Kamilova — Chief Specialist, NAD, SA - Mr. Jakhongir Mekhmonov — Chief Specialist, NAD, SA *Prepared by IMF CCAMTAC technical assistance mission on national accounts for the Republic of Uzbekistan (July 24–August 4, 2023).* ### Section 1 ### REPUBLIC OF UZBEKISTAN — Report on National Accounts Mission (July 24–August 4, 2023) ### Summary of Mission Outcomes and Priority Recommendations - Mission duration: July 24 to August 4, 2023. - Purpose: Assist the Statistics Agency under the President of the Republic of Uzbekistan (SA) in developing annual supply and use tables (SUT) and improving non-observed economy (NOE) estimates. - Key implementation: - Implemented the Python version of the balancing tool SUTB developed by IMF staff. - Compiled a symmetric 2021 SUT with 144 activity types and products. - Expanded the SUTB file to include 144 columns and rows of the output and intermediate consumption matrices and applied the SUTB balancing procedure after data preparation and analysis of statistical discrepancies by product. - NOE survey: - SA plans a sector-specific NOE survey covering transport, food and accommodation, education, and health care. - Stratified sample: 7,000 households. - Pilot: 16 households interviewed (mostly SA employees); results transferred from tablets to MS Access database; mission and SA created queries and compiled weighted household expenditures for questionnaire items. - Field work scheduled: start on August 7 and end on August 25. - Intended use: NOE survey results to be used in compilation of national accounts in 2025. - Recommendation on SUT timing: SUTs should be compiled before publication of annual GDP so SUT can be used to determine balanced GDP levels and allow changes to output and intermediate consumption during balancing. - Priority Recommendations (extract): - December 2023: Conduct logical control of the initial results of the NOE survey. — Responsible: SA - April 2024: Develop the NOE survey database and prepare the main results. — Responsible: SA - April 2024: Use the energy balance data for the SUT compilation. - April 2024: Improve accounting of goods produced for own consumption and adjust output of manufacturing (wood, meat, and milk processing, etc.). - August 2024: Improve data sources for the distribution of gross fixed capital formation by type of asset. - December 2024: Prepare and manually balance the SUT at current 2022 prices then remove final, minor imbalances by using the SUTB balancing tool. — Responsible: SA - April 2025: Use the results of the NOE survey in compiling GDP. - Priority action classification: H – High; M – Medium. ### Detailed Technical Assessment and Recommendations — Priority Action / Milestones - Target Completion Dates and High-priority actions: - December 2023: Conduct logical control of the initial results of the NOE survey. (H) - April 2024: Develop the NOE survey database and prepare the main results. (H) - April 2024: Use the energy balance data for the SUT compilation. (H) - April 2024: Improve accounting of goods produced for own consumption and adjust output of manufacturing (wood, meat, and milk processing, etc.). (H) - April 2024: Collect price indices for selected SUT indicators. (M) - April 2024: Use the tourism satellite accounts data in compiling SUTs. (H) - August 2024: Improve data sources for the distribution of gross fixed capital formation by type of asset. (H) - December 2024: Prepare and manually balance the SUT at current 2022 prices then remove final, minor imbalances by using the SUTB balancing tool. (H) - April 2025: Use the results of the NOE survey in compiling GDP. (H) ### A. Non-Observed Economy (NOE) — Findings and Recommended Actions Findings: - Planned NOE survey: sector-specific, covering transport, food and accommodation, education, and health care. - Sample design: - Total sample: 7,000 households. - Population base: approximately 7 million households. - Stratification: 27 strata (Tashkent city and 13 regions, each divided into urban and rural settlements). - Next geographical level: mohalla; total mohallas: 9,000; selected mohallas: 309. - From each mohalla: 22 or 23 households interviewed; reserve list includes another 23 households for refusals. - Data collection: - Electronic questionnaires on tablets; results transferred to MS Access database maintained by Department of National Accounts. - Pilot test: - 16 households interviewed (mostly SA employees). - Mission and SA created MS Access queries and calculated weighted household expenditures by questionnaire item to assist full-database processing. - Pilot issues identified: - Questionnaire tables mix monthly and annual expense items; respondents may report annual instead of monthly values or vice versa. - Expenses must be entered in 1000 units of national currency; this was sometimes omitted by respondents. - Some expense items (e.g., restaurants) may require interviewing all adults; multiple household visits may be necessary. - Field work schedule: start August 7 and end August 25. Recommended Actions: - Conduct logical control of the initial results of the NOE survey. - Develop the NOE survey database and prepare the main results. - Use the results of the NOE survey in compiling GDP. ### B. Supply and Use Tables (SUT) Balancing Tool SUTB — Findings and Guidance: - Background: Prior SUTB implementation relied on a MATLAB-based mechanism requiring 64-bit Excel 2016 on one computer; risked obsolescence and hardware dependency. - Current mission: Implemented the SUTB balancing tool based on Python, deployable on any modern Excel-enabled computer. - SUTB function requirements (five arguments): - Unbalanced SUT, - Three sets of accounting constraints (grouped by type: two accounting-type constraints — supply-use and netting adjustments — and user-defined explicit constraints), - A set of reliability coefficients. - Reliability coefficients: - Scale from 0 to 100, where 100 represents fully reliable data not subject to adjustments. - Because SUTs were compiled after annual GDP publication, some indicators have reliability coefficients equal to 100 percent; mission recommends compiling SUTs before annual GDP publication to allow adjustments to output and intermediate consumption. - Application: - 2021 SUT was balanced using the Python SUTB. - SUT for Uzbekistan is symmetrical and includes 144 types of activities and products; SUTB file expanded to 144 rows and columns for output and intermediate consumption. The 2021 SUT — Product-level discrepancies, diagnostic actions, and adjustments: - Products with significant supply-use discrepancies (listed with NACE Rev.2 codes where provided): - Extraction of crude petroleum (NACE Rev.2 - 06.1) - Other food products except sugar (NACE Rev.2 - 10.8 except 10.81) - Sugar (NACE Rev.2 - 10.81) - Forestry and logging (NACE Rev.2 - 02) - Passenger rail transport, interurban (NACE Rev.2 - 49.1) - Dairy products (NACE Rev.2 - 10.5) - Plastics in primary forms (NACE Rev.2 - 20.16) - Crude petroleum: - Extraction and import data based on regular surveys; structure of intermediate consumption from 2019 detailed survey, making intermediate consumption relatively less reliable. - Main data source for crude petroleum is the energy balance; SA published an experimental energy balance for 2021 with volume indicators of output, changes in inventories, and consumption. - Mission used energy balance data to compile intermediate consumption of crude oil, which was about 40 percent higher than the preliminary estimate; after adjustment, the discrepancy between supply and use of crude petroleum was reduced. - Because GVA for crude petroleum had already been published and could not be changed, intermediate consumption was revised using a special technique: other products used in manufacturing of petroleum products were reduced by the same amount and redistributed proportionally; final result: intermediate consumption for manufacturing of oil products did not change in total but its structure shifted from crude oil to other products. - Sugar: - Two large sugar processing enterprises process raw sugar; raw sugar classified as two types: (i) sugar cane (agricultural product, NACE Rev.2 - 01.14) and (ii) processed sugar cane supplied to sugar refinery (10.81). - SA had classified processed sugar cane as NACE Rev.2 - 01.6 due to the 2019 intermediate consumption survey questionnaire containing only one question on sugar cane coded 01.6, producing inaccurate intermediate consumption structure. - Mission recommended reclassifying processed sugar cane to NACE Rev.2 - 10.81; SA implemented this change, significantly reducing the discrepancy between sugar supply and use. - Forestry and wood products: - Supply of forestry products far exceeded their use; detailed analysis found output of roundwood (NACE Rev.2 – 02.2) was almost twice the output of manufacture of wood products and paper (NACE Rev.2 – 16 and 17). - More than half of forestry production is produced by households and mainly processed for own-account construction of residential buildings. - Own-account production chain (from production through own-account processing and use in own-account construction) needs full accounting: include own-account production of wood in industrial production index and national accounts. - Current status: own-account production of agriculture and food industry is included in industrial production index; own-account production of wood is not considered; output of food industry does not include own-account production of meat. - Recommendation: Value household processing of wood and meat at producer prices and add to output of these industries. - Passenger rail transport: - Output of passenger rail transport was much less than use. - Analysis found output, intermediate consumption, household consumption, and exports of services corresponded to 2021 situation, but imports of services were much lower than in previous years. - SA compiles exports and imports of services from two sources: (i) foreign economic activity survey (corporations) and (ii) tourism research (visitors and local households). - Mission judged tourism survey data more reliable than foreign trade surveys for rail passenger services; after changing the data source, imports of services increased and the difference between supply and use reduced significantly. *Prepared by IMF CCAMTAC technical assistance mission on national accounts for the Republic of Uzbekistan (July 24–August 4, 2023).* ### Section 2 ### Section 2 ### Discrepancies in sectoral accounts: dairy and plastics - Final consumption of dairy products was almost twice the output at basic prices. - Main reason: dairy products produced by households for own consumption, which are not fully accounted for when compiling output. - By considering household-produced dairy for own consumption, the statistical discrepancy will be significantly reduced. - Plastics in primary forms: resources far exceeded their use. - Main reason: underestimation of intermediate consumption. - Imports in 2021 significantly exceed intermediate consumption, with these products almost entirely used in the production process as intermediate consumption. - When imports are largely used as intermediate consumption, it is necessary to change the output or intermediate consumption estimates; if these estimates are fixed, balancing is particularly difficult. - This is cited as a specific example of why it is better if the SUT is compiled before publishing the annual GDP. ### SUT compilation and publication timetable - The 2021 SUT will be published in December 2023. - The preliminary 2022 SUT at current prices will be compiled at the end of October 2023. - The final 2022 SUT will be published in December 2024. - In 2024 the SA also plans to compile the SUT at the previous year’s prices. ### Recommended actions for SUT compilation and balancing - Prepare and manually balance the SUT at current 2022 prices then remove final, minor imbalances by using the SUTB balancing tool. - Use the energy balance data for the SUT compilation. - Improve the accounting of goods produced for own consumption and adjust the output of manufacturing (wood, meat, and milk processing, etc.). - Use the tourism satellite accounts data in compiling SUTs. - Improve data sources for the distribution of gross fixed capital formation by type of asset. - Collect price indices for selected SUT indicators. ### Officials met during the mission - Mr. Agzam Ikramov — Deputy Director of SA - Mr. Abrorali Mamadjanov — Head of the Macroeconomic Indicators and National Accounts Department (NAD), SA - Mr. Odiljon Mamadaliev — Head of the International Cooperation Department, SA - Mr. Sardor Juraev — Head of the Industrial Statistics Department, SA - Mr. Abdugafir Mirzakhidov — Head of the Service Statistics Department, SA - Mr. Akrom Sultanov — Head of the Coordination and Monitoring Department, SA - Ms. Zukhra Mamadjonova — Head of the Input-Output Statistics Division, NAD, SA - Mr. Umarjon Obidov — Deputy Head of the NAD, SA - Mr. Rakhmatjon Urusbayev — Head of the Satellite Accounts Division, NAD, SA - Mr. Farkhod Ganiev — Deputy Head of the Satellite Accounts Division, NAD, SA - Ms. Tatyana Kuksina — Chief Specialist, NAD, SA - Ms. Zioda Kamilova — Chief Specialist, NAD, SA - Mr. Jakhongir Mekhmonov — Chief Specialist, NAD, SA *IMF | CCAMTAC Report – Republic of Uzbekistan National Accounts Statistics Mission* --- _Source: https://www.imf.org/-/media/files/publications/tar/2024/english/tarea2024070-print-pdf.pdf_