A 1992 SOCIAL ACCOUNTING MATRIX (SAM) FOR TANZANIA Peter Wobst University of Hohenheim International Food Policy Research Institute TMD DISCUSSION PAPER NO. 30 Trade and Macroeconomics Division International Food Policy Research Institute 2033 K Street, N.W. Washington, D.C. 20006 U.S.A. August 1998 TMD Discussion Papers contain preliminary material and research results, and are circulated prior to a full peer review in order to stimulate discussion and critical comment. It is expected that most Discussion Papers will eventually be published in some other form, and that their content may also be revised. This paper was written under the IFPRI project “Macroeconomic Reforms and Regional Integration in Southern Africa” (MERRISA), which is funded by DANIDA (Denmark) and GTZ (Germany). TANZANIA MOZAMBIQUE MALAWI ZAMBIA ZIMBABWE SOUTH AFRICA MACRO ECONOMIC REFORMS AND REGIONAL INTEGRATION IN SOUTHERN AFRICA Trade and Macroeconomics Division International Food Policy Research Institute Washington, D.C. A 1992 Social Accounting Matrix (SAM) for Tanzania Peter Wobst August 1998 TMD Discussion Paper No. 30 Contents List of Abbreviations 1. Introduction and summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2. A macroeconomic social accounting matrix (macrosam) for 1992 . . . . . . . . . . . . 4 2.1. Data sources for the macrosam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.2. Documentation of macrosam cell entries . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 3. Microeconomic social accounting matrix (microsam) . . . . . . . . . . . . . . . . . . . . . 14 3.1. The disaggregation of the microsam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 3.2. Accounts of the microsam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 3.3. Documentation of data entries in the microsam . . . . . . . . . . . . . . . . . . . . . 17 3.4 Balancing the SAM using a cross-entropy approach . . . . . . . . . . . . . . . . . 28 3.5. The new macrosam after the balancing procedure . . . . . . . . . . . . . . . . . . . 31 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Annex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Abstract This paper documents the construction of a 1992 social accounting matrix (SAM) for Tanzania. On the basis of recently generated national accounts data, a 56-sector SAM is built focusing on the disaggregation of agriculture — which comprises 21 of the 56 sectors. First, a highly aggregated SAM (macrosam) is designed to set the macroeconomic framework that provides the control totals for the disaggregation procedure. Then, the sector disaggregation of the microeconomic SAM (microsam) is done. Data sources and the data adjustments made are presented. The microsam differentiates 4 household types and 5 labor categories. Special features of the microsam include non-monetary, own-household consumption and separate marketing margins on domestic products, exports, and imports — which play a crucial role in the low income economies of sub-Saharan Africa. Since the data base is to be used for economic policy modeling, consideration of these features will have a significant influence on the results of the analysis. Due to data insufficiencies the first microsam obtained from adjusted raw data (protosam) is highly unbalanced. A cross-entropy estimation method is applied to balance the protosam and generate the final estimated 1992 microsam for Tanzania that uses all available information in a consistent framework. List of abbreviations BOS Bureau of Statistics BOT Bank of Tanzania CGE Computable General Equilibrium C.I.F. Cost, Insurance, Freight GAMS General Algebraic Modeling System GDP (f.c.) Gross Domestic Product at Factor Costs GFCF Gross Fixed Capital Formation GNP (f.c.) Gross National Product at Factor Costs F.O.B. Free on Board HBS Household Budget Survey LFS Labour Force Survey macrosam Macroeconomic Social Accounting Matrix MERRISA Macroeconomic Reforms and Regional Integration in Southern Africa microsam Microeconomic Social Accounting Matrix NGO Non-Government Organization protosam Raw (unbalanced) Microeconomic Social Accounting Matrix ROW Rest of the World SAM Social Accounting Matrix TShs Tanzanian Shillings URT United Republic of Tanzania The MERRISA project consists of five country studieson Malawi, Mozambique, Tanzania,1 Zambia, and Zimbabwe and a trade focused regional component including South Africa. Social accounting matrices are developed for each of the six countries and a comprehensive data base of regional trade flows is developed. Applying this extensive new data, individual studies of the impact of domestic economic policies are carried out and opportunities and consequences for extended regional integration are analyzed. For a detailed analysis of the Economic Structure, Trade, and Regional Integration in Southern Africa, see Mukherjee and Robinson (1997). For a general description of the SAM concept see Pyatt and Round (1985).2 The difference of one account occurs because the activity account for Tourism does not require a3 respective commodity account since all its domestic production is exported and none of it enters the domestic commodity market. See Table 3 below for a complete list of accounts. 1 1. Introduction and summary As part of the Tanzania country studybeing done under the IFPRI research project on Macroeconomic Reform and Regional Integration in Southern Africa (MERRISA), a social accounting matrix (SAM)has been constructed for the base year 1992. A SAM is a square1 matrix consisting of row and column accounts that represent the different sectors, agents, and institutions of an economy at the desired level of disaggregation. The SAM, a useful framework for consistent multi-sectoral economic data preparation, represents the expenditure-receipt flows among all actors and sectors of the entire economy, capturing both input-output and national income and product data. 2 Because the analytical focus of the MERRISA project is agriculture and because agricultural production in Tanzania accounts for about half of GDP, the disaggregated Tanzania SAM contains 21 agricultural sectors out of 56 sectors in total. The Tanzania SAM consists of 56 activity accounts capturing the flows belonging to the domestic production process and 55 commodity accounts capturing the flows belonging to the marketing process of nationally and internationally produced goods. The factor disaggregation of productive3 factors consists of Capital, Land and 5 labor categories, while households are divided into Rural Farmers, Rural Non-Farmers, Urban Farmers, and Urban Non-Farmers. The construction of a SAM under poor data conditions is not only an exercise in putting together a complete data set, but also an estimation process on the basis of limited and often inconsistent existing data sources. In the case of Tanzania, data availability and data consistency are very limited and it is extremely difficult to obtain a complete and reliable national data base. To cope with limited data throughout the construction process of the disaggregated microeconomic SAM (microsam), a macroeconomic SAM (macrosam) has to The latest discussion on the cross-entropy approach is led by Golan and Judge (1996), but can be4 best related to the present context in Robinson, Cattaneo, and El-Said (Forthcoming). Both versions are financial SAM including domestic financial structure of the economy. Since5 the CGE approach that is applied within the MERRISA analysis focuses on real variables and aggregates of the economy, the nature of the SAM constructed here is non-financial and allows only limited comparison with the 1976 versions. 2 be constructed first. The macrosam provides the main macroeconomic characteristics and magnitudes of the economy and sets the basic data framework for the further development of the microsam. It is highly aggregated and consists of only one activity, commodity, factor and household account, capturing the basic macroeconomic features like total intermediate demand, value-added, factor payments, foreign trade, tax and savings characteristics, domestic supply and demand, and all domestic and international monetary transfers. The second step to achieve a consistent and balanced microsam is the construction of a preliminary SAM which typically is inconsistent and unbalanced and will be referred to as the protosam. The protosam incorporates all available raw data which are eventually adjusted to match various macroeconomic control totals. For those sub-matrices within the SAM for which an accurate disaggregation of the data is not available, rough estimates are made. The accounts of the microsam have to be balanced while also achieving the aggregate control totals from the macrosam. The resulting microsam will be used as the data base for computable general equilibrium (CGE) modeling within the MERRISA project. A cross- entropy approach to SAM estimation is used for the balancing process leading from the protosam to the balanced microsam. The protosam provides a “prior” for the parameter4 estimation using the cross-entropy method. Although inconsistent, it is a starting point and contains useful “information” which is used along with the various marcoeconomic constraints to estimate the new microsam. Section 2 describes the construction of the Tanzania macrosam for 1992, the data sources used, and the documentation of data entries. Section 3 documents the data sources and disaggregation criteria applied to the microsam, describes the construction of the protosam, presents the cross-entropy technique used for the balancing process, and gives a detailed presentation on figures and indicators of the final microsam. The construction of a 1992 SAM for Tanzania is quite a challenge considering the above-mentioned data inadequacies. Earlier work on SAMs for Tanzania by Rutayisire and Vos (1991) and Sarris (1994) are based on 1976 national accounts data and the 1976 input- output table for Tanzania — unfortunately the most recent input-output table available.5 3 However, CGE analysis of structural adjustment, trade liberalization, and elements of the macroeconomic transformation Tanzania experienced in the early and mid-1990s requires a more recent data base. Recently processed national accounts data for 1992 contribute to this requirement. Preparation of an input-output table for 1992 is still in progress, so the 1992 SAM for Tanzania uses the basic structure of the 1976 input-output table with substantial adjustments based on other sources of information. As soon as the 1992 input-output table for Tanzania is published by the Bureau of Statistics in Dar es Salaam — hopefully not later than by the end of 1999 — the current exercise can easily be repeated to obtain a more reliable data base for further analysis. Under the given circumstances, the current 1992 SAM for Tanzania incorporates all data available at this time and applies an extremely powerful method for the final balancing process, the cross-entropy approach. Furthermore, the 1992 Tanzania SAM has to be seen in the context of other SAMs built for African countries, in particular those being developed for comparative study under the MERRISA project, but also for other countries like Madagascar, Botswana, Uganda, and Ghana which are in process or planned to be constructed. Each of these country studies contributes to a broader and deeper knowledge of how to deal with insufficient raw data to construct a comprehensive and consistent data base incorporating African-specific features. Prepared by the Planning Commission at the President’s Office, in cooperation with the Bureau6 of Statistics (BOS) in 1997. Among which: The 1991/92 Household Budget Survey, the 1991 Informal Sector Survey, several7 Agriculture Surveys and the 1994 Survey of Construction, Trade and Transport. This data set was made available through staff members of the National Accounts Section of the8 Bureau of Statistics in Dar es Salaam. 4 2. A macroeconomic social accounting matrix (macrosam) for 1992 The Tanzania macroeconomic social accounting matrix (macrosam) for the year 1992 contains 31 non-zero entries. The initial macrosam balances the entire economy at a gross output level of TShs 2,759,506 million and a total domestic absorption of TShs 2,940,773 million. The difference results from Tanzania's high trade deficit in 1992 of TShs 387,681 million or 243 percent of total export earnings. As in many developing countries with insufficient data processing and publishing capacity, data from various sources are often inconsistent. It is necessary to adjust data from these different sources to gain a consistent economywide data base. This adjustment process needs to fulfill certain criteria. Most important is the decision about the core data source, which determines the macrosam's control totals to which all other data will be reconciled and balanced. In the case of the Tanzania macrosam, the Revised National Accounts of Tanzania 1987-1996 provides the control totals for the macrosam.6 This latest version of the National Accounts of Tanzania incorporates a variety of other surveys conducted in recent years and some new economic features. Besides various7 household expenditure categories, the most important new feature is the estimate for informal sector activities. Formal GDP (f.c.) is current TShs 935,247 million, informal — or non- monetary — GDP (f.c.) is estimated at current TShs 340,668 million or 36.4 percent of total GDP (f.c.). The informal GDP component considers agriculture, construction, and owner- occupied dwellings and therefore not only provides general information on the informal sector share in total GDP — whose magnitude estimated by different authors ranges from 30 to more than 60 percent — but also allows the explicit specification of own-household consumption. A second important data source is an unpublished supplementary data set which was used by the Bureau of Statistics for the preparation of the Revised National Accounts 1987- 1996. Listing 79 economic sectors, this data set provides sector-specific information on gross8 output, intermediate demand, imports, tariffs, sales taxes, exports, private consumption, These two tax accounts are pure auxiliary accounts for the collection of different taxes within the9 economy which then are transferred to government. In the commodity account column they enable the distinction between sales taxes and tariffs which otherwise would have to be combined into one payment from “Commodities” to “Government Recurrent.” 5 government consumption, investment, and changes in inventory, which are used as control totals for the Tanzania macrosam and as source of information for the disaggregated microsam. Table 1 is a schematic diagram which introduces the different features and functionality of the macrosam. The macrosam specifies two government accounts: Government Recurrent and Government Investment. The former deals with the recurrent budget activities of the government while the latter deals with the development (or investment) budget activities of the government. In the case of the Tanzania SAM, the development budget considers all government expenditures that are related to gross fixed capital formation. Keeping two separate accounts guarantees more flexibility for upcoming modeling purposes where government might boost the national level of infrastructure through increased investment spending. Two additional tax collection accounts are specified which capture the entire national tax scheme at macroeconomic level. Table 2 presents the Tanzania macrosam for9 1992 in million of current TShs, which is used as the macroeconomic point of reference for the construction of the 1992 Tanzania microsam. Table 1: Macroeconomic Social Accounting Matrix (MACROSAM) Activities Commodities Factors Households Enterprises Tariffs TotalDomestic Government Government Rest of the Capital Taxes Recurrent Investment World Account Activities Domestic Own HH Exports Total Supply Consump. f.o.b. Sales Commodities Intermediate Final HH Final Gov. Government Private Domestic Demand Consump. Consump. Investment Investment Demand Factors Value- Value-Added Added (f.c.) Households VA Operating Remittances Household Labor Surplus from Abroad Income Enterprises VA Government Enterprise Capital Transfers Income Domestic Indirect Sales Taxes Taxes Tax Dom. Taxes Subtotal Tariffs Import Import Tariffs Tariffs Government Income Corporate Indirect & Import Aid-related Recurrent Recurrent Taxes Taxes Sales Taxes Tariffs Grants Gov. Receipts Government Gov. Invest. Aid-related Financing Investment Deficit Loans Gov. Invest. ROW Imports Factor Paym. Interest Paym. Payments to c.i.f. abroad to ROW ROW Capital HH Enterprise Gov. Net Capital Total Account Savings Savings Savings Inflow Savings Total Gross Domestic Value-Added HH Enterprise Dom. Tax Import Gov. Recurr. Gov. Invest. ForEx Total Private Output Absorption (f.c.) Expenditure Expenditure Subtotal Tariffs Expenditure Expenditure available Investment 7 Table 2: Tanzania macrosam for 1992 in current million TShs Activities Factors Households Enterprises Tariffs TotalCommo- Domestic Government Government Rest of the Capital dities Taxes Recurrent Investment World Account Activities 2,320,484 273,340 165,682 2,759,506 Comm. 1,276,427 913,213 279,080 52,521 419,532 2,940,773 Factors 1,456,047 1,456,047 Households 841,879 368,663 96,755 1,307,296 Enterprises 550,669 27,620 578,289 Dom. Taxes 27,032 43,475 70,507 Tariffs 23,451 23,451 Gov. Rec. 16,656 65,054 70,507 23,451 172,671 348,339 Gov. Inv. 34,957 17,564 52,521 ROW 553,363 63,499 24,250 641,112 Capital Acc. 104,087 144,572 -17,568 188,440 419,532 Total 2,759,506 2,940,773 1,456,047 1,307,296 578,289 70,507 23,451 348,339 52,521 641,112 419,532 8 2.1. Data sources for the macrosam The following list of publications provides an overview of the major data sources used in the construction of the macrosam 1992 as presented in Table 2 and documented in the next section: Bank of Tanzania (1997): Economic Bulletin for the Quarter Ended 31 March, 1997, Vol.st XXV No.1, Dar es Salaam. Economic Research Bureau (1996): Tanzanian Economic Trends - A Bi-annual Review of the Economy, Vol.8 No.1 and 2, University of Dar es Salaam, Dar es Salaam. International Monetary Fund. 1996. Tanzania — Statistical Appendix. IMF Staff Country Report No, 96/2. Washington, D.C.: International Monetary Fund. The Economist Intelligence Unit ( 1996): County Profile: Tanzania / Comoros 1995-96, London. The United Republic of Tanzania (1997): Revised National Accounts of Tanzania 1987- 1996, Dar es Salaam. The United Republic of Tanzania (1996): Household Budget Survey 1991/92, Vol. IV, Dar es Salaam. The United Republic of Tanzania (1995): Revised National Accounts of Tanzania 1976- 1990, Dar es Salaam. The United Republic of Tanzania (1995): National Accounts of Tanzania 1976-94, Dar es Salaam. The United Republic of Tanzania (1993): Labour Force Survey 1990/91, Dar es Salaam. The World Bank (1996): Tanzania - The Challenge of Reforms: Growth, Incomes and Welfare, Vol. II. 9 2.2. Documentation of macrosam cell entries The notation for the cells of the macrosam cells is row, column. For example, “Commodities, Activities” represents an expenditure flow from the column “Activities” to the row “Commodities.” Note that Tanzania's financial year ends on June 30 and thus all data collected as financial year data have to be estimated for 1992 by computing the averages of 1991/92 and 1992/93 figures. 1. (Commodities, Activities): 1,276,427 — Import-ridden intermediate demand. Total intermediate demand including imports as from unpublished BOS data sources adjusted by other indirect production taxes (Domestic Taxes, Activities). 2. (Factors, Activities): 1,456,047 — Value-added at factor costs. Total value-added at factor costs as in Table 6(a) of URT (1997): Revised National Accounts of Tanzania 1987- 1996 adjusted accordingly to supplementary data to the Revised National Accounts of Tanzania 1987-96. 3. (Domestic Taxes, Activities): 27,032 — Other indirect taxes. Other indirect taxes are calculated as a residual of total net taxes on products as in Table 6(a) of URT (1997): Revised National Accounts of Tanzania 1987-1996 and tariffs and sales taxes as provided by the supplementary data to the Revised National Accounts of Tanzania 1987-96. 4. (Activities, Commodities): 2,320,484 — Domestic supply. Domestic supply is computed as the residual of the “Activities” column total (equal to gross output including informal sector activities) minus exports (Activities, ROW) and own-household consumption (Activities, Households). 5. (Domestic Taxes, Commodities): 43,475 — Sales taxes. Total sales tax as provided by the supplementary data to the Revised National Accounts of Tanzania 1987-96 and double-checked with data from Table 2.1 of Bank of Tanzania (1997): Economic Bulletin 1 /97, Dar es Salaam.st 6. (Tariffs, Commodities): 23,451 — Import duties. Total import duties as provided by the supplementary data to the Revised National Accounts of Tanzania 1987-96. 7. (ROW, Commodities): 553,363 — Imports of goods and services. Total imports of goods and services at f.o.b. prices as in Table 6(b) of URT (1997): Revised National Accounts of Tanzania 1987-1996, Dar es Salaam, adjusted by a 75 percent share of the A detailed description of the computations carried out for the generation of all value-added10 figures in the microsam is presented in the following section. See the discussion of the value-added split to labor, land, and capital under paragraph 211 (Factors, Activities) of the microsam documentation in section 3. 10 unrecorded trade and statistical discrepancy (TShs 19.023 million) — the remaining 25 percent are considered within the computation of total exports. 8. (Households, Factors): 841,879 — Labor value-added. Total labor value-added will be generated for the microeconomic SAM using the URT (1993): Labour Force Survey10 1990/91, Dar es Salaam (LFS 1990/91). For the macrosam it is set at 60 percent of total value-added adjusted for 50 percent of net factor payments abroad (ROW, Factors).11 9. (Enterprises, Factors): 550,699 — Value-added capital. Total value-added capital — including operating surplus and consumption of fixed capital — as a residual of total factor payment (Factors, Activities) minus labor value-added (Households, Factors) and adjusted for 50 percent of net factor payments abroad. 10. (ROW, Factors): 63,499 — Net factor income paid abroad. Total net factor income paid abroad as calculated from Table 1 of the URT (1995): National Accounts of Tanzania 1976-94, Dar es Salaam. The difference between GDP (f.c.) and GNP (f.c.) as a ratio to GDP (f.c.) (all 1992) is applied to the new GDP (f.c.) figure as in cell “Factors, Activities.” 11. (Activities, Households): 273,340 — Own-household consumption. Total own- household consumption of agricultural products as the equivalent of total non-monetary agriculture GDP as in Table 1 of URT (1997): Revised National Accounts of Tanzania 1987- 1996, Dar es Salaam. 12. (Commodities, Households): 913,213 — Final household consumption. Total final household consumption as in Table 6(a) of URT (1997): Revised National Accounts of Tanzania 1987-1996, Dar es Salaam, minus own-household consumption as in cell “Activities, Households” and adjusted accordingly to the supplementary data to the Revised National Accounts of Tanzania 1987-96. 13. (Government Recurrent, Households): 16,656 — Individual income taxes. Total individual income taxes paid by households as in Table 2.1 of Bank of Tanzania (1997): Economic Bulletin 1 /97, Dar es Salaam. Income taxes, other taxes, and non-tax revenue arest 11 combined into one figure and shared between households and enterprises according to their relative shares in 1990 as reported in Table 5.2 of World Bank (1996): The Challenge of Reforms, Vol. II, Washington, D.C. 14. (Capital Account, Households): 104,087 — Household savings. Total household savings as a residual for balancing purposes. 15. (Households, Enterprises): 368,663 — Operating surplus. Operating surplus distributed to households is calculated as the residual of total receipts of the enterprise account — value-added capital plus government transfers to enterprises — minus enterprise savings (Capital Account, Enterprises) and corporate taxes (Government Recurrent, Enterprises). 16. (Government Recurrent, Enterprises): 65,054 — Corporate taxes. Total corporate taxes paid by enterprises as in Table 2.1 of Bank of Tanzania (1997): Economic Bulletin 1 /97, Dar es Salaam. As described under paragraph 13, income taxes, other taxes,st and non-tax revenue are combined into one figure and shared between households and enterprises according to their respective shares as from fiscal year 1990 information in Table 5.2 of World Bank (1996): The Challenge of Reforms, Vol. II, Washington, D.C. 17. (Capital Account, Enterprises): 144,572 — Enterprise Savings. Total enterprise savings as chosen for balancing purposes. Since the transfer of operating surplus from enterprises to households — paragraph 15 above — is calculated as a residual of total enterprise receipts minus corporate taxes and enterprise savings, enterprise savings can vary. The less enterprises save the more operating surplus is transferred to households. Since household expenditure on own-household consumption, final commodity consumption, and income tax are given, household savings remains the balancing cell of the household account. As a result, the less enterprises save the more households have to save and vice versa. This mechanism is used to adjust household and enterprise savings to obtain reasonable economywide rates. However, the sum of enterprise savings and household savings match the reported magnitude of domestic private saving. As a percent of adjusted GDP (m.p.) for the fiscal years 1992 and 1993, these are 10.9 and 14.0 percent, respectively (IMF 1996). 18. (Government Recurrent, Domestic Taxes): 70,507 — Domestic tax collections. Total domestic tax collections of net indirect taxes (Domestic Taxes, Activities) and sales tax (Domestic Taxes, Commodities) paid to the government account. 19. (Government Recurrent, Tariffs): 23,451 — Import duties collections. Total import duties collections (Tariffs, Commodities) paid to the government account. Since parastatals in this data base are not explicitly distinguished from private sector activities,12 the GFCF of parastatals as stated in the Tanzanian Economic Trends are combined with the private sector GFCF and not with the GFCF financed by the government. 12 20. (Commodities, Government Recurrent): 279,080 — Government consumption. Total final government consumption as in the supplementary data to the Revised National Accounts of Tanzania 1987-96. 21. (Enterprises, Government Recurrent): 27,620 — Government transfers to enterprises. Domestic government interest payments as reported for the financial years 1992 and 1993 in Table 11 of the Tanzania — Statistical Appendix, IMF Staff Country Report No. 96/2, adjusted in accordance with the applied final government consumption figure of the macrosam. 22. (Government Investment, Government Recurrent): 34,957 — Government investment account deficit. Non-foreign financed government spending within the development budget to balance the government investment account of the macrosam. 23. (ROW, Government Recurrent): 24,250 — Government transfers to ROW. Foreign interest payments by the government as reported for the financial years 1992 and 1993 in Table 11 of the Tanzania — Statistical Appendix of the IMF Staff Country Report No. 96/2, adjusted in accordance with the applied final government consumption figure of the macrosam. 24. (Capital Account, Government Recurrent): -17,568 — Government deficit. Total government deficit of recurrent government budget as a residual of total government revenue (Government Recurrent, Total) minus government consumption (Commodities, Government Recurrent), government transfers to enterprises, the balancing position of the government investment account, and government transfers to ROW. 25. (Commodities, Government Investment): 52,521 — Gross fixed capital formation (GFCF) by the government. Total GFCF by the government corresponding to government investment under the developing budget as in Table 5 of the statistical annex of Economic Research Bureau (1996): Tanzanian Economic Trends - A Bi-annual Review of the Economy, Dar es Salaam. The figure is adjusted to match the higher total GFCF applied12 in the macrosam. 26. (Commodities, ROW): 165,682 — Exports of goods and services. Total exports of goods and services at c.i.f. prices as in Table 6(b) of URT (1997): Revised National 13 Accounts of Tanzania 1987-1996, Dar es Salaam, adjusted by the remaining 25 percent share of the unrecorded trade and statistical discrepancy (TShs 19,023 million). 27. (Households, ROW): 96,755 — Remittances. Net direct transfers to households from abroad as computed from Table 4.4 of Bank of Tanzania (1997): Economic Bulletin 1 /97, Dar es Salaam, with respect to the relative household share of transfer inflows. Thisst figure probably contains a large share of grants to NGOs which are assumed to be part of households. 28. (Government Recurrent, ROW): 172,671 — Transfers to government from abroad. Net direct transfers to government from abroad as computed from Table 4.4 of Bank of Tanzania (1997): Economic Bulletin 1 /97, Dar es Salaam, with respect to the relativest government share of transfer inflows. This figure represents primarily foreign aid with a considerable grant component including recurrent government budget support. 29. (Government Investment, ROW): 17,564 — Other transfers to government. Net official sector capital flows in US$ as in Table 23 of the reference tables in The Economist Intelligence Unit ( 1996): County Profile: Tanzania / Comoros 1995-96, London, applying the exchange rate reported in this publication. 30. (Capital Account, ROW): 188,440 — Net capital inflow. Net capital inflow from Rest of the World — not elsewhere specified in the ROW column — as a residual to balance the macroeconomic reference data base and the implicitly generated current account. 31. (Commodities, Capital Account): 419,532 — Net private investment. Total private investment as in the supplementary data to the Revised National Accounts of Tanzania 1987-96 including changes in inventory as in the same data set, adjusted for government investment as described in paragraph 24 (Commodities, Government Investment). Tourism has been the largest foreign currency earning sector since the mid-1990s.13 For a detailed listing of the applied sector disaggregation refer to section 3.2. Accounts of the14 microsam. 14 3. Microeconomic social accounting matrix (microsam) This section describes the disaggregation of the microsam, the choice of data sources, the data manipulation carried out, and the estimation process for balancing the microsam. The documentation follows the section on the macrosam. The same numbering and notation is used for the description of the protosam and microsam entries, although in most cases it is not a single cell entry anymore, but a vector or matrix of entries depending on the disaggregation of the respective macrosam accounts. All crucial calculations, distributions, and data adjustments to estimate the protosam are documented. The cross-entropy approach for estimating the final microsam from the basis of the constructed protosam is discussed, including the implementation in the General Algebraic Modeling System (GAMS) program. In the following section the disaggregation of the microsam is described. 3.1. The disaggregation of the microsam For the construction of the microsam the activity and commodity accounts of the macrosam are disaggregated into 56 and 55 sectors respectively. Three additional commodity accounts are created to capture different marketing margins for exports, imports, and domestically produced and consumed goods. Since the analytical focus of the underlying study is on agriculture, 21 activities are agricultural. The disaggregation of activities and commodities is based on the 72-sector listing given by the Revised National Accounts of Tanzania 1976-90 which is the same disaggregation that is used for the 1976 I-O Table. In addition, the activity Tourism is created to capture this fast growing sector of the Tanzanian economy. Since the entire gross output of this sector is supposed to be exported, no13 commodity account for Tourism had to be created. The microsam distinguishes six typical export crop activities (Cotton, Sisal, Tea, Coffee, Tobacco, and Cashew Nuts), five cereals (Maize, Wheat, Paddy, Sorghum, and Other Cereals), Pulses, two drought-resistant staple crops (Cassava and Other Roots), Oil Seeds, Sugar, Other Horticulture, Other Crops, Livestock, Fishing, and finally Forestry & Hunting. The non-agricultural sectors are Mining, 22 manufacturing sectors, Electricity, Water, Construction, and nine service categories including Commerce, Transport & Communication, and Public Administration.14 15 The factor disaggregation of the microsam follows the breakdown of labor categories in the Labour Force Survey 1990/91. The following five labor categories are chosen — their respective sub-components presented in parentheses: Professionals (administrators/managers, professionals, associate professionals), White Collar (clerks/cashiers, services/shops), Blue Collar (craft workers, machine operators), Unskilled Labor and Agricultural Labor. Additional factors of production are Land and Capital. The disaggregation of households follows the Household Budget Survey 1991/92 (HBS 91/92) which provides detailed consumption data for Urban Farm, Urban Non-Farm, Rural Farm and Rural Non-Farm households. Since this categorization captures the desired distinction between rural and urban population and has been applied in many other data sources, it appears to be the most appropriate choice. All other institutional accounts of the microsam follow the macrosam. 3.2. Accounts of the microsam Table 3: Accounts of the microsam Acc. Code Description Acc. Code Description AG-Sectors 1 Cotton 2 Sisal (Activities) ACOTT ASISA 3 Tea 4 CoffeeATEA ACOFF 5 Tobacco 6 Cashew NutsATOBA ACASH 7 Maize 8 WheatAMAIZ AWHEA 9 Paddy 10 SorghumAPADD ASORG 11 Other Cereals 12 BeansAOCER ABEAN 13 Cassava 14 Other Roots & TubersACASS AROOT 15 Oil Seeds 16 SugarAOILS ASUGA 17 Other Horticulture 18 Other CropsAOHOR AOCRO 19 Livestock 20 FisheryALIVE AFISH 21 Forestry & HuntingAFOHU NON-AG Sectors 22 Mining 23 Meat & Dairy Prod. (Activities) AMINE AMEAT 24 Processed Food 25 Grain Mill ProductsAFOOD AGRAI Acc. Code Description Acc. Code Description 16 26 Beverages 27 Textiles nes.ABEVT ATEXT 28 Wearing Apparel 29 Leather ProductsAWEAR ALEAT 30 Wood & Wood Prod. 31 Paper & PrintingAWOOD APAPE 32 Other Chemicals nes 33 Fertilizer & PesticidesACHEM AFERT 34 Petroleum Refineries 35 Rubber ProductsAFUEL ARUBB 36 Plastic Products 37 Glass ProductsAPLAS AGLAS 38 Cement, Clay, etc. 39 Iron & SteelACEME AIRON 40 Manu. of Metal prod. 41 Machinery EquipmentAFMPR AMAEQ 42 Electrical Equipment 43 Transport EquipmentAELEQ ATREQ 44 Other Manufactures 45 ElectricityAOMAN AELEC NON-AG (cont.) 46 Water 47 ConstructionAWATE ACNST 48 Wholesale & Retail 49 Hotels & RestaurantsATRAD AHORE 50 Transport & Comm. 51 Financial InstitutionsATR_C AFI_I 52 Real Estate 53 Business ServicesAREAL ABUSI 54 Public Administration 55 Other ServicesAPUBA AOSER 56 TourismATOUR The same sector disaggregation as for activities applies to the respective commodity accounts — except for Tourism (ATOUR). In addition, three marketing margin accounts are specified under commodities. Marketing 112 Export MM 113 Domestic MM Margin (MM) Acc. (Commodities) CCOME CCOMD 114 Import MMCCOMI Factors of 115 Professional labor 116 White collar labor Production UPRO UWCO 117 Blue collar labor 118 Unskilled laborUBCO UNSK 119 Agricultural labor 120 LandRURA LAND 121 CapitalCAPITAL Households 122 Urban Farmers 123 Urban Non-FarmersHHUFA HHUNF Acc. Code Description Acc. Code Description In 1976 the share of subsidies in “Taxes on Production and Imports” was 5.6 percent and in15 1990 only 1.4 percent (URT 1990b). 17 124 Rural Farmers 125 Rural Non-FarmersHHRFA HHRNF Other 126 Enterprises 127 Domestic Indirect Taxes Institutional Accounts ENTR ITAX 128 Tariffs 129 Government RecurrentTTAX GOVR 130 Government Investment 131 Rest of the WorldGOVI WORLD 132 Capital Account 133 Change in InventoryKACCOU DST Note that: AG = Agriculture, NON-AG = Non-Agriculture, nes. = not elsewhere specified 3.3. Documentation of data entries in the microsam Following the documentation in the macrosam chapter, each corresponding cell entry, vector or sub-matrix of the microsam is discussed. Data sources are presented which provide clean data, raw data or structural information for data adjustments. Decisions for data manipulations are justified in this section as well. 1. (Commodities, Activities): To derive the intermediate demand matrix of the microsam, normalized input-output coefficients — coefficients of the intermediate demand sub-columns add up to one — are applied to sectoral intermediate demand figures, given by the supplementary data to the Revised National Accounts for Tanzania 1987-96. The applied coefficients are adjusted values of the 1976 input-output table of Tanzania, the most recent available source of information. Consequently, the general production technology and economic structure which is assumed to characterize the Tanzanian economy is that of 1976. It provides a starting point (“prior”) for the protosam and determines the estimation of the final technology of the microsam. Because the 1976 input-output table covers domestically produced intermediates only and contains tax and subsidy information that does not match with the economic characteristics of the early 1990s, the coefficients have to be adjusted. Since the structure of subsidies changed completely between the mid 1970s to the early 1990s the respective15 coefficients are netted out and thus subsidies are removed from the system. To adjust the (domestic) intermediate demand coefficients for imported intermediate demands, an import matrix is developed. Control totals for total imports per commodity are provided in the Revised National Accounts — see also paragraph 7 (ROW, Commodities). The Challenge of Reforms: Growth, Incomes and Welfare, Vol. II, p. 21.16 This is the intermediate use of a sectoral output by all sectors of the economy — e.g. total17 intermediate use of cotton by all productive activities including cotton itself — and it is not the total intermediate use of inputs by the cotton sector. 18 Shares of investment demand (capital goods), household and government demand (consumer goods), and intermediate demand for imports are derived from Table 3.5 of the most recent World Bank (1996) Country Report on Tanzania. On the basis of the domestically produced16 intermediate demand for each commodity with (strictly positive) imported intermediate demand, row coefficients are calculated and applied to imported intermediate demand totals. Imported intermediate demand is further adjusted using information from Balsvik and Brandemoen (1994), who compile intermediate demand use of imports for some crops on the basis of the annual crop reviews by the Marketing Development Bureau. The derived imported intermediate demand matrix is eventually combined with the matrix of domestically produced intermediate demands to obtain the “total” coefficients. Since the Revised National Accounts provides information on total intermediate demand of sector products row coefficients are calculated from the intermediate demands17 and applied to these control row totals. This technique ensures the consistency of sectoral production and absorption data, and also guarantees the correct magnitude of the total intermediate demand matrix. The derived structure of the intermediate input matrix is translated into final (sub-) column coefficients which can be applied to sectoral figures for total intermediate demand by sector. Since the supplemental information on the latest national accounts publication distinguishes total expenditure on inputs and value-added, the two sub-technologies — intermediate use and generation of value-added — can be applied separately instead of applying the entire sector technology to the respective gross output. For sectors with indirect taxes (Domestic Taxes, Activities), the intermediate demand entries of the respective column entries are reduced relatively, so that the sum of intermediate demand and indirect taxes meet the Revised National Accounts of Tanzania 1987-96 information. 2. (Factors, Activities): Information on GDP (f.c.) per sector is provided by supplementary data to the Revised National Accounts of Tanzania 1987-96. Data from the Labour Force Survey 1990/91 is used to calculate the value-added paid to the different labor factors. First, the survey provides detailed information on sectoral employment for 9 different labor categories — for urban and rural areas respectively — which have been aggregated to the 5 categories used in the microsam. Second, the survey differentiates primary and secondary occupations, including their respective average weekly working time in hours. Many economic activities in low income countries like Tanzania are carried out as secondary18 occupations by employees who either have a primary paid job or carry out an unpaid (mostly self- employed) main activity (e.g. farming). Information on the primary factor splits for agriculture derived by the Global Trade Analysis19 Project (GTAP) of the Center for Global Trade Analysis, Purdue University, indicates a 15:60:25 split of agricultural value-added between land, labor, and capital for Sub-Saharan African countries (Republic of South Africa and Republic of Zimbabwe). Consequently, a 40:60 split of non-labor value-added between land and capital appears reasonably close to the reported 15:25 — this is 37.5:62.5 — split of the GTAP data. As mentioned earlier, the applied coefficients are adjusted for their respective subsidy20 coefficients — these used to be substantial shares of gross output in 1976 but accounts for only one percent of GDP (f.c.) in 1990. The only exception is the activity account for Tourism which does not have a commodity21 account since all of its produce is exported and none enters the domestic market. 19 Third, average monthly wages for the different labor categories for rural and urban areas are derived from the survey. This employment and wage information is used to compute a18 sector-specific spread of total labor value-added among the different labor categories. Subsequently, the sum of value-added to land and capital for each sector is calculated as the residual of sectoral GDP (f.c.) and total labor value-added. Since information on the spread between land and capital value-added is not available, 40 percent is arbitrarily allocated to land and 60 percent to capital, following general practice in the literature.19 3. (Domestic Taxes, Activities): Indirect taxes related to the production process are calculated through their respective input-output coefficients. The coefficients are applied to20 the sectoral gross output figures given by supplementary data to the Revised National Accounts of Tanzania 1987-96 and subsequently adjusted to match total other indirect taxes as specified in the macrosam. 4. (Activities, Commodities): The entries of the main diagonal of this matrix represent the domestic supply of domestically produced goods. The off-diagonal entries are zero for all commodity accounts, because the present SAM does not include multi-commodity producing activities nor multiple activities producing the same commodity. In other words, each activity produces exactly one good, which is exclusively supplied to its commodity market. However, three marketing margin accounts are introduced to capture the21 transportation and marketing costs related to each commodity. Since the national accounts statistics of Tanzania categorize transportation and marketing costs for all commodities with the retail sector, the share of final demand for the retail sector is about 14 percent of total The marketing margin flow for exports is considered under paragraph 1 (Commodities,22 Activities) since activities buy the marketing margin flow associated with their export component out of the marketing margin account “CCOME.” This practice follows the assumption that exports are reported in F.O.B. prices including their respective marketing margin. Since exports (at F.O.B. prices, including marketing margins) are bought out of activities (Activities, ROW), but the activity columns usually report on gross output at farm/factory gate prices (excluding marketing margins), the marketing margins for exports have to be incorporated somewhere in the production process (the column technology). However, this practice violates the clean definition of the column sum being sectoral gross output since the net gross output figure of each column is adjusted by its respective export marketing margin flow. Consequently, gross output for Trade (ATRAD) is adjusted for the sum of all export marketing margin flows and the ATRAD column entries are recalculated. 20 final demand. Consequently, each commodity is associated with a certain amount of transportation and marketing costs of the retail sector for its “delivery” to the destination of its final demand. Therefore, each commodity account buys its relevant transport and marketing costs out of the three marketing margin accounts, depending on its shares of domestic, imported or exported production. Simultaneously, all final consumption and investment demands are adjusted for these additional costs and in turn, the final demand for retail (the commodity CTRAD) is eliminated. To balance the three marketing margin accounts their totals are bought out of the activity account Trade (ATRAD). The remaining flow from the commodity trade to the activity trade account (ATRAD, CTRAD) represents the part of gross output in trade that enters the domestic market for intermediate demand. The supply of domestic produce entries are obtained by subtracting own-household consumption and exports — as described in the paragraphs 10 and 26 below — from total sectoral gross output (obtained from previous calculations). 4.a. (Commodities, Commodities): All commodity accounts showing marketing margins for domestic produce and/or imports buy their marketing margin values out of the respective marketing margin commodity accounts (CCOMD and CCOMI). That way final22 household and investment demand for trade is eliminated and each commodity is associated with its own marketing margin. 5. (Domestic Taxes, Commodities): Sales taxes are directly adopted from supplementary data to the Revised National Accounts of Tanzania 1987-96. 6. (Tariffs, Commodities): Tariffs are directly adopted from supplementary data to the Revised National Accounts of Tanzania 1987-96. 7. (ROW, Commodities): Import values for all 55 commodities are obtained from sectoral information in the supplementary data to the Revised National Accounts of Tanzania The remaining 25 percent is deducted from total exports.23 The remaining 50 percent of Net Factor Income Paid Abroad is deducted from the operating24 surplus paid from enterprises to households. The non-farm households are assumed to have no Agricultural Labor endowment and should25 therefore not being considered in its value-added distribution. Nevertheless, most of the non-farm households — although not commercially involved in agriculture — carry out some non-commercial farming for own-household consumption which is reported by the “HBS 91/92.” This production volume is assumed to be also considered in the informal sector GDP reported by the Revised National Accounts of Tanzania 1987-96 and as such being part of value-added to Agricultural Labor. 21 1987-96 and adjusted for the import control total given by Table 6 (a) of the Revised National Accounts of Tanzania 1987-96. The import control total is increased by 75 percent of the Unrecorded Trade and Statistical Discrepancy figure reported in this table. Further23 adjustments are made for the sectors Machinery Equipment and Other Manufacturing. The two sectors show higher exports than gross output figures, which implies negative domestic supply of domestic produce. This phenomenon indicates re-exports of imported goods because the economy exports more than its produced gross output. The two export figures are netted out and their respective import figures are lowered by the equivalent values. 8. (Households, Factors): Labor Value-added is distributed to households according to the calculations described in paragraph 2 (Factors, Activities). The information on rural and urban employment allows the distribution of labor value-added to rural and urban households. Farm households (HHRFA and HHUFA) are assumed to be endowed with the factor Farming Labor (RURA) and non-farm households (HHRNF and HHUNF) are endowed with the remaining four labor factors. Consequently, household shares for each labor factor are calculated and applied to the respective total value-added by factor. The obtained distribution of labor value-added among household groups for the microsam — after applying the cross- entropy balancing approach — is reported in Table 3. Note that Professional Labor (UPRO) value-added is decreased by 50 percent of Net Factor Income Paid Abroad as specified in paragraph 10 (ROW, Factors). Furthermore, distribution of Agricultural Labor (RURA)24 value-added to non-farm households (HHUNF and HHRNF) results from Total Agricultural Own-Household Consumption of the respective household. Total land value-added is25 distributed to rural and urban farm households according to their relative labor income. 22 Table 4: Distribution of Labor Value-Added Values and Shares per HH Group in Million TShs & Percent UPRO UWCO UBCO UNSK RURA Total in % HHUFA 102,358 12.10 0 0 0 102,,358 HHUNF 291,689 34.662,914 58,158 111,930 52,966 5,721 HHRFA 402,218 47.70 0 0 0 402,218 HHRNF 47,432 5.620,809 4,595 10,255 4,254 7,519 9. (Enterprises, Factors): Total capital value-added — net of 50 percent of Net Factor Income Paid Abroad — is transferred to enterprises for further distribution. The enterprise account represents all productive enterprises in the economy and acts as a collecting point for total capital value-added. Besides value-added capital, the enterprise account receives transfers from the government. On the expenditure side, the enterprise account pays corporate taxes to the government, enterprise savings to the capital account, and distributes the remaining “operating surplus” to households (as described in paragraphs 14 to 16). 10. (ROW, Factors): Total net factor income is paid abroad as described in paragraph 10 of the macrosam documentation. Since there is no information on which part of this factor payment is related to labor or capital, half of it is netted out from factor payments Professional Labor — considering the relative shares of different households — and half of it is netted out from factor payments Capital. 11. (Activities, Households): The disaggregation of Own-Household Consumption is calculated according to the respective household shares in Own Produced Food as provided by the HBS 91/92. The survey not only reports on total household shares of non-monetary food expenditures, but also reports on the commodity distribution of these expenditures for each household. The sum of own-household consumption and final household consumption has to match the sector control total for total consumption for each household given by the supplementary National Accounts data. Consequently, sector-specific coefficients are applied to these control totals, which are calculated on the basis of first estimates for the distribution of own and final household consumption. Own-household consumption is a direct demand by private households for output of activities. The goods are produced by activities, but not fed through the commodity accounts to enter the domestic markets for intermediate or final consumption. The reason is that own consumption is assumed to be valued at producer prices, with no trade margins. In other 23 words, own-household consumption reduces sectoral gross output supplied to the domestic commodity market (as described under paragraph 4 (Activities, Commodities)). The consideration of own-household consumption within the SAM framework captures agricultural subsistence demand, an extremely important feature of the Tanzanian economy. The incorporation of own-household consumption into the SAM is important for policy analysis of its effects on household income, effective consumption, and welfare. Usually, this kind of transaction is netted out within the Production Consumption Process of a Household Production Unit and appears neither on the income nor on the expenditure side. Consequently, reported income of small-scale farmers often is underestimated. 12. (Commodities, Households): Total private household consumption reported in the Revised National Accounts of Tanzania 1987-1996 and adjusted for own-household consumption, is distributed among the four household groups according to (a) the sectoral distribution of total household consumption in 1992 given by the supplementary National Accounts data and (b) the relative commodity distribution among households given by the HBS 91/92. As described in the previous paragraph, the distribution of final household consumption is calculated in connection with household spending on own-household consumption. The coefficients which are derived from the first estimates of own and final household consumption are applied to the respective sector control totals. Since the relative sector spread of total household consumption gets distorted by this procedure, the relative shares of total consumption expenditure among households in the microsam differ from their shares in the HSB 91/92 as is shown in Table 4. Table 5: Final Household Consumption Values and Shares per HH Group in Million TShs & Percent HHUFA HHUNF HHRFA HHRNF Total HBS 91/92 710,116136,563 197,244 348,782 27,527 in % 19.23 27.78 49.12 3.88 100.00 microsam 913,213155,260 254,857 455,108 47,989 in % 17.0 27.9 49.8 5.3 100.00 13. (Government Recurrent, Households): Total individual income taxes, as computed in paragraph 13 of the macrosam documentation, are distributed among the four household groups according to their relative labor income. Each labor type is taxed with a specific rate. In 1992, the Tanzanian income tax scheme ranged from 0 to 30 percent of monthly taxable income, and permitted several deductions. The ratio of total individual income taxes over total labor value-added shows an average income tax rate of only 2.1 24 percent for the economy as a whole. The enormous difference between this de facto average income tax rate and the nominal income tax rates is partly due to the high share of small-scale agriculture and informal sector activities in the Tanzanian economy, which are not taxed at all. Furthermore, the efficiency in tax collection is affected by extremely low administrative and technical standards. As a result, it appears quite reasonable that de facto and nominal income tax rates show substantial differences. A tax rate scheme is developed on the basis of a Monthly Income Distribution of Government Employees According to the Individual Tax Rate Brackets provided by the Tanzanian Income Tax Department. Although the sample contains government employees only (and thus does not represent the entire formal labor force of the economy) and the data reports on 1996 incomes, it delivers persuasive evidence for an extremely narrow range of effective income tax rates over all income brackets. For an aggregation of all nine income brackets to three brackets which represent the SAM labor categories (a) Professionals, (b) White and Blue Collar and (c) Unskilled and Rural Labor, the effective average income tax rates are 3.7, 3.4 and 1.1 percent, respectively. To guarantee a gradual progression across the three income groups in the microsam, tax rates of 5.0, 3.0 and 1.0 percent are applied to the respective income flows in the microsam. Finally, these rates are adjusted to match total individual income tax revenue of TShs 16,656 million for the fiscal year 1992. After the adjustment of the applied tax rates and the balancing procedure, the final income tax rates for the four household groups are 0.6, 1.0, 1.5, and 2.5 percent, respectively. Individual household rates, values, and shares of total income taxes are shown in Table 5. Table 6: Individual Income Tax Rates, Values and Shares per HH Group in Million TShs & Percent HHUFA HHUNF HHRFA HHRNF Total / Average Rate in % 1.60.7 1.0 2.3 2.9 Value in TShs 16,6561,258 2,845 11,011 1,542 Share in % 100.07.6 17.1 66.1 9.3 14. (Capital Account, Households): Saving shares for different household groups are derived from non-consumption expenditure information of Appendix 6 of the HBS 91/92. The saving shares and respective values for the four household groups after the balancing procedure are presented in Table 6. 25 Table 7: Household Savings Rates, Values and Shares per HH Group in Million TShs & Percent HHUFA HHUNF HHRFA HHRNF Total / Average Rate in % 8.419.0 11.7 2.0 7.0 Value in TShs 83,44636,476 33,923 9,443 3,603 Share in % 100.043.7 40.7 11.3 4.3 15. (Households, Enterprises): The distribution of operating surplus to households is calculated according to group-specific income-expenditure deficits that result after all household expenditures and all household incomes except capital income and remittances are taken into account. The operating surplus distributed to households is calculated as a residual of total enterprise receipts net of enterprise savings and corporate taxes. 16. (Government Recurrent, Enterprises): Since the microsam contains only one representative enterprise — as in the macrosam — no changes occur in total corporate taxes being paid by enterprises to recurrent government in the order of TShs 65,054 million as described earlier. 17. (Capital Account, Enterprises): Again, since the microsam contains only one enterprise account, no distribution of the macrosam figure for enterprise savings is necessary. However, after the balancing procedure total enterprise savings equals TShs 195,335 million. 18. (Government Recurrent, Domestic Taxes): Since there is no difference in the function of the intermediate tax collection accounts in the macrosam and the microsam, the flow from the intermediate tax account Domestic Taxes to the government account (representing the sum of all domestically collected taxes as described under paragraphs 3 and 5) is the same as described in the macrosam section. 19. (Government Recurrent, Tariffs): The same argument holds for the Tariffs account. The flow of total collected tariffs — as described in paragraph 6 — from the intermediate tax account Tariffs to the government account works the same way as described in the macrosam section. 20. (Commodities, Government Recurrent): Final government consumption follows the supplementary National Accounts information and consists of government demand for The SAM for Tanzania incorporates education and health services in Public Administration.26 The related private demand equals 2.5 percent of total government demand for Public Administration. There are two general practices to deal with public administration and government services.27 Either the government demand a bundle of commodities and provides these items to the public or the public administration activity does contain this consumption bundle as part of its intermediate demand. In latter case, the government has no final demand for any other item than public services produced under the public administration activity. 26 Public Administration only. This specification assumes none or very little private demand for Public Administration and that government consumes the major share and provides it to the26 public.27 21. (Enterprises, Government Recurrent): Government transfers to enterprises as specified in the macrosam section remain a single cell entry and are fixed at their initial value of TShs 27,620 million after the balancing procedure. 22. (Government Investment, Government Recurrent): Government spending on the development budget remains one entry as specified in the macrosam counting for TShs 34,957 million after the cross-entropy. 23. (ROW, Government Recurrent): Government spending on foreign interest payments remains a single entry in the microsam as well and remains at TShs 24,250 million after the balancing procedure. 24. (Capital Account, Government Recurrent): Government saving in the microsam is calculated as the residual between total government revenue (that is the row total of the government account) and final government consumption, government transfers to enterprises, and the payments to the government investment account as described in paragraphs 20 to 22. Since the macro control totals of all government revenue categories and the macro control totals for all other government expenditure categories are constrained during the balancing procedure, the government deficit remains at its initial value of TShs - 17,586 million. However, if the government recurrent account payment to the government investment account were eliminated, the government recurrent account would show a surplus of TShs 17,389 million and the government investment account a deficit of TShs -34,957 million. 25. (Commodities, Government Investment): Gross fixed capital formation by government as specified in the macrosam, and remains a single entry in the microsam, representing demand of construction at the same level as initialized. 27 26. (Commodities, ROW): Export values for all 56 activities are obtained from sectoral information of the supplementary data to the Revised National Accounts of Tanzania 1987-96 and adjusted for the export control total given by Table 6 (a) of the Revised National Accounts of Tanzania 1987-96. The export control total is reduced by 25 percent of the Unrecorded Trade and Statistical Discrepancy figure reported in Table 6 (a) as mentioned in paragraph 7. Furthermore, the sectoral adjustments for Machinery Equipment and Other Manufacturing as described in paragraph 7 are taken into account. 27. (Households, ROW): Remittances from abroad to the four household groups are distributed according to group-specific income-expenditure deficits as described for the distribution of operating surplus in paragraph 15 and remain at their initial level. 28. (Government Recurrent, ROW): Net direct transfers to government from abroad as described in the macrosam section remain at their initial level. 29. (Government Investment, ROW): Capital inflows from abroad supporting the government development budget as described in the macrosam section remain at their initial level. 30. (Capital Account, ROW): Net capital inflows from Rest of the World as described in the macrosam section changes to TShs 158,320 million after the balancing procedure. 31. (Commodities, Capital Account): Final private investment demand per sector as reported in the supplementary data to the Revised National Accounts of Tanzania 1987-96 net of government investment in Construction (considered under the government investment account). As opposed to the procedure in the macrosam section, the values represent GFCF only and are net of changes in inventories, which are considered under a separate account as specified in paragraph 33 (Commodities, Change in Inventory). 32. (Change in Inventory, Capital Account): The flow represents the balancing of the Change in Inventory account through the Capital Account. The Change in Inventory account is not an explicit feature of the macrosam where GFCF and changes in inventory are combined. However, for later modeling purposes, these two flows are separated in the microsam. Ti, j For a more detailed discussion of the cross-entropy approach to SAM estimation see Robinson,28 Cattaneo, and El-Said (Forthcoming). 28 33. (Commodities, Change in Inventory): Change in inventory demand per sector is as reported in the supplementary data to the Revised National Accounts of Tanzania 1987- 96 adding up to TShs 4,146 million. Due to the above mentioned limitations and inconsistencies of the available data for the “SAM-making” process, the commodity columns and rows of the generated protosam are showing high deviations which make it difficult for the solver of the balancing program to find an optimal solution. To cope with this problem and to provide the solver with a better starting point, several adjustment are made. The input-output matrix is cleared of unreasonable entries, some negligible exports and imports are netted out from their respective counter flows, and activity columns or commodity rows are scaled up or down according to their sectoral excess demand or excess supply values. After this adjustment sub-routine the obtained protosam is balanced through the cross-entropy balancing procedure as described in section 3.4. For detailed information on the microsam entries refer to Table A1 of the Annex. The new macrosam, which is recalculated after the balancing of the microsam using the cross- entropy estimation method, is presented in section 3.5, after the description of the cross- entropy approach. 3.4 Balancing the SAM using a cross-entropy approach28 The microsam entries presented in the previous section are not only the result of sectoral data information and relative spreads within the various sub-groups of accounts, but also the result of the final balancing procedure of the SAM. A cross-entropy approach to SAM estimation is used for the balancing process leading from the unbalanced protosam to the balanced microsam. Since data availability and data consistency are limited, the cross- entropy approach is an appropriate tool for estimating a balanced and consistent data base starting from an unbalanced data base that contains all available information. The SAM used so far is defined as a matrix T of monetary flows (a payment from account j to account i), representing receipts and expenditures of all economic agents. Following the convention of double-entry bookkeeping, total receipts and total expenditures of a particular agent i have to be equal, i.e. respective row and column sums are balanced: & I( p:q ) ' & j n i&1 pi lnpi qi Ei A yi ' j j Ti, j ' j j Tj,i Ai, j ' Ti,j yj with j i Ai, j ' 1 œ i y ' A y min j i j j A ( i,j @ ln( A ( i, j Ai,j ) s.t.: j j A ( i, j y ( j ' y ( i and j j A ( i, j ' 1 œ i A Following information theory developed by Shannon (1948) and further developed by Theil29 (1967) the expectation of separate information values can be described as the expected information of data points: , where q and p are prior and posterior probabilities regarding a set of events and -I(p:q) is the Kullback-Leiber (1951) measure of the “cross-entropy” distance between the two probability distributions. The cross-entropy approach minimizes the cross-entropy distance between the probability distributions that are consistent with the information in the data and the prior. As formulated by Golan, Judge, and Robinson (1994) to update an input-output table by solving30 for a new coefficient matrix A which minimizes the entropy difference between the underlying prior and the new matrix A. 29 (1) Dividing every cell entry of the flow matrix T by its respective column total generates a matrix A of column coefficients: (2) In matrix notation it follows that: (3) Balancing a SAM is an underdetermined estimation problem using information from many sources and various years. The cross-entropy approach allows the incorporation of29 errors in variables, inequality constraints, and prior knowledge about any part of the SAM — not just row and column sums. These features of the cross-entropy estimation technique allow great flexibility in incorporating specific information and implementing certain limits to which the estimation results are restricted. The general cross-entropy approach is described by the30 following optimization problem (4) where is a coefficient matrix representing any (perhaps inconsistent and unbalanced) prior that was chosen as a starting point of the cross-entropy balancing process to achieve the A y ' A y j i Ai , j ' 1 œ j A ( min j i j j A ( i,j @ ( ln A ( i,j& ln Ai, j ) j i j j G (k) i, j @ Ti,j ' ((k) ((k) Tij y ' x % e x ei ' j w Wi,w @ vi,w j w Wi,w ' 1 with 0 # Wi,w # 1 This means that the prior does not need to satisfy the model , but the sum of its31 column coefficients adds up to one, i.e. . Note that if the error distribution is symmetrically centered around zero and all weights are32 equal — as their initial prior values — the respective error equals zero. 30 desired new coefficient matrix . The described problem is set up to minimize the entropy31 difference between the two coefficient matrices which becomes more obvious by rearranging it to (5) Additional equality and inequality constraints can be formulated as linear “adding-up” constraints on various elements of the SAM. For an aggregator matrix G, which has ones for those microsam entries that correspond to a certain macrosam aggregate and zeros otherwise, the formulation for k such aggregation constraints is given by (6) where is the value of the aggregate and the 's are the microsam flows. Measurement errors in variables can be incorporated into the system through (7) where y is a vector of row sums and the initially known vector of column sums measured with error. The error e is defined as a weighted average of known constants (8) where w is a set of weights W, v are constants, and weights are subject to (9) For the purposes of the Tanzania microsam, a symmetric distribution around zero given lower and upper bounds is chosen, using three weights. Consequently, the32 optimization problem of minimizing the entropy difference now contains a term for the weights W min j i j j A ( i,j @ ( ln A ( i,j& ln Ai, j ) % j i j w Wi,w @ ln Wi,w . 31 (10) The explicit application of the cross-entropy estimation procedure on the Tanzania microsam contains a set of additional constraints that constrain various sums over sub- matrices of the SAM to their respective macro control totals. First, within activities, the sum over all factor payments is fixed to their aggregate value as specified in the macrosam. As a result, total GDP (f.c.) is constrained to its original value. Sectoral production may change within specified lower and upper limits which are imposed through the error specification, allowing shifts in relative sector shares of production in the economy. Second, the foreign trade entries are constrained to their macro totals, although the relative commodity composition of imports and exports may change. Third, total final household, government, and investment demands are bound to their macro totals as reported in the Revised National Accounts of Tanzania 1987-96 as well as total own-household consumption. Finally, total income taxes, sales taxes, other indirect taxes, tariffs, and total remittances to households from abroad are fixed at their macro totals. Some single-cell entries are locked to their initial values if the data source applied is reliable, such as government investment demand and factor payments abroad. For a detailed description of the imposed constraints, refer to the GAMS program presented in the Annex. 3.5. The new macrosam after the balancing procedure The distribution of macroeconomic data according to the sector disaggregation of the microsam as described in section 3.3 and the cross-entropy estimation procedure to balance the protosam, as described in the previous section, leads to the final Tanzania microsam for 1992. The Tanzania microsam 1992 is presented in Table A1 of the annex. The sector-specific structure of the economy can be seen from Table A2 of the annex. After obtaining the final microsam the final macrosam is aggregated. This allows the comparison between initial and final macroeconomic structure and points out the changes that took place during the balancing procedure. The obtained values of the new macrosam are presented in Table 7 including their respective changes from the initial macrosam. Table 8: New macrosam for 1992 in current million TShs Activities Factors Households Enterprises Tariffs TotalCommo- Domestic Government Government Rest of the Capital dities Taxes Recurrent Investment World Account Activities 2,440,513 273,340 165,682 2,879,535 +120,029 Comm. 1,396,456 211,682 913,213 279,080 52,521 419,532 3,272,484 +120,029 new Factors 1,456,047 1,456,047 Households 884,150 275,629 126,875 1,286,655 +42,272 -93,033 +30,120 Enterprises 508,398 27,620 536,018 -42,272 Dom. Taxes 27,032 43,475 70,507 Tariffs 23,451 23,451 Gov. Rec. 16,656 65,054 70,507 23,451 172,671 348,339 Gov. Inv. 34,957 17,564 52,521 ROW 553,363 63,499 24,250 641,112 Capital Acc. 83,446 195,335 (17,568) 158,320 419,532 -20,642 +50762 -30,120 Total 2,879,535 3,272,484 1,456,047 1,286,655 536,018 70,507 23,451 348,339 52,521 641,112 419,532 33 For the cross-entropy balancing procedure certain macroeconomic control totals are imposed as constraints. In other words, some cell entries and some sub-matrix totals of the proposed microsam are fixed at their initial levels. In particular, this is true for: ! total value-added ! the three tax aggregates (other indirect taxes, sales taxes, and tariffs) ! the trade flows (total exports and imports) ! all aggregate final demand categories (household, government, government investment, and private investment demands as well as own-household consumption) ! total factor payments abroad ! corporate and total individual income taxes ! government transfers to enterprises and abroad ! transfers from abroad to the government (recurrent and investment accounts). Consequently, the respective cells of the new macrosam do not show any changes compared to the initial macrosam. Other cells are not explicitly, but implicitly fixed. Since the government investment demand and the related inflow from abroad are fixed the balancing flow from government recurrent to government investment cannot change either. Consequently the government deficit is the last unconstrained cell of the government expenditure column, but since the totals of all government revenue sources are fixed, the government deficit is implicitly constrained. Total intermediate demand (Commodities, Activities) remains unrestricted for the balancing process since it has to adjust to total value-added, which is increased to match total final consumption of the national accounts data. The macro total for intermediate demand increases by TShs 120,029 million including TShs 18,213 million for total marketing margins for exports. The net change accounts for 8.0 percent of the original value and the resulting change in total gross output accounts for 4.3 percent. According to the mechanism of the macrosam where domestic supply equals gross output minus exports and own-household consumption (which are both fixed), the change in the total domestic supply value is the same as the reported increase of total intermediate demand. The new entry in the new macrosam (Commodities, Commodities) represents total import and domestic marketing margin values. In the protosam TShs 226,911 million are distributed from final demand for retail and wholesale trade (CTRAD) to the two marketing margin accounts for domestic products and imports. The final figure of TShs 211,682 million corresponds to a decrease of 6.7 percent. The initially chosen distribution of total value-added into capital value-added paid to enterprises and non-capital value added paid directly to households changed by TShs 42,272 34 million, a 5.0 percent change from the initial factor payments to households. Total household savings are adjusted from TShs 104,087 million to TShs 83,446 million whereas enterprise savings increase from TShs 144,572 million to TShs 195,335 million. Households receive TShs 93,033 million less value-added capital distributed through enterprises and TShs 30,120 million more remittances from abroad. Consequently, the payments to the capital account from abroad decrease by latter amount. 35 References Bacharach, M. 1970. Biproportional matrices and input-output change. University of Cambridge, Department of Applied Economics. Cambridge: Cambridge University Press. Balsvik, R. and A. Brendemoen. 1994. A computable general equilibrium model for Tanzania: Documentation of the model, the 1990 — Social Accounting Matrix and Calibration. Oslo-Kongsvinger: Statistics Norway. Bank of Tanzania. 1997. Economic bulletin for the quarter ended 31 March, 1997. Vol.st XXV, No. 1. Dar es Salaam: BOT. Economic Research Bureau. 1996. Tanzanian economic trends - A bi-annual review of the economy, Vol.8, No.1 and 2. Dar es Salaam: University of Dar es Salaam. Golan, A. 1998. Entropy, Likelihood and Uncertainty: A Comparison. In Erickson, G. Maximum entropy and Bayesian methods, ed. G. Erikson. Kluwer Academic Publishers, forthcoming. Golan, A. and G. G. Judge. 1996. A maximum entropy approach to empirical likelihood estimation and inference. Working Paper Berkeley: University of California. Golan, A., G. G. Judge, and D. Miller. 1996. Maximum entropy econometrics: Robust estimation with limited data. New York: John Wiley & Sons. Golan, A., G. G. Judge, and S. Robinson. 1994. Recovering information from incomplete or partial multisectoral economic data. Review of Economics and Statistics 76:541-549. International Monetary Fund. 1996. Tanzania — statistical appendix. IMF Staff Country Report No. 96/2. Washington, D.C.: International Monetary Fund. Mukherjee, N. and S. Robinson. 1997. Economic structure, trade, and regional integration in Southern Africa. In Achieving food security in Southern Africa, ed. L. Haddad. Washington, D.C.: International Food Policy Research Institute. Pyatt, G. and J. I. Round. 1985. Social accounting matrices: A basis for planning. Washington, D.C.: The World Bank. 36 Robinson, S., A. Cattaneo, and M. El-Said. Estimating a Social Accounting Matrix Using Entropy Difference Methods. TMD Discussion Paper Series. Washington, D.C.: International Food Policy Research Institute, forthcoming. Rutayisire, L and R. Vos. 1991. A SAM for Tanzania. Working Paper Sub-Series on Money, Finance and Development No. 39. The Hague: Institute of Social Studies. Sarris, A. 1994. A social accounting matrix for Tanzania. Washington, D.C.: Cornell University Food and Nutrition Policy Program. Schneider, M. H. and S. A. Zenios. 1990. A comparative study of algorithms for matrix balancing. Operations Research, Vol. 38, pp. 439-55. The Economist Intelligence Unit. 1996. County profile: Tanzania / Comoros 1995-96. London: The Economist Intelligence Unit. The United Republic of Tanzania. 1997. Revised national accounts of Tanzania 1987-1996. Dar es Salaam: Bureau of Statistics. ______. 1996. Household budget survey 1991/92, Vol. IV. Dar es Salaam: Bureau of Statistics. ______. 1995a. Revised national accounts of Tanzania 1976-1990. Dar es Salaam: Bureau of Statistics. ______. 1995b. National accounts of Tanzania 1976-94. Dar es Salaam: Bureau of Statistics. ______. 1993. Labour force survey 1990/91. Dar es Salaam: Bureau of Statistics and Labour Department. The World Bank. 1996. Tanzania - The challenge of reforms: Growth, incomes and Welfare, Vol. II. Washington, D.C.: The World Bank. 37 Annex TABLE A1: TANZANIA MICROSAM 1992 IN CURRENT TSHS MILLION ACOTT ASISA ATEA ACOFF ATOBA ACASH AMAIZ AWHEA APADD ASORG AOCER ABEAN ACASS AROOT AOILS ASUGA AOHOR AOCRO ALIVE AFISH ACOTT ASISA ATEA ACOFF ATOBA ACASH AMAIZ AWHEA APADD ASORG AOCER ABEAN ACASS AROOT AOILS ASUGA AOHOR AOCRO ALIVE AFISH AFOHU AMINE AMEAT AFOOD AGRAI ABEVT ATEXT AWEAR ALEAT AWOOD APAPE ACHEM AFERT AFUEL ARUBB APLAS AGLAS ACEME AIRON AFMPR AMAEQ AELEQ ATREQ AOMAN AELEC AWATE ACNST ATRAD ATOUR AHORE ATR_C AFI_I AREAL ABUSI APUBA AOSER CCOTT CSISA 0.080 CTEA CCOFF 0.711 CTOBA 0.112 CCASH CMAIZ 17.832 CWHEA 2.016 CPADD 10.726 CSORG 3.120 COCER 1.030 CBEAN 6.975 CCASS 0.801 CROOT 0.961 COILS 2.923 CSUGA 9.449 COHOR 7.413 COCRO 2.130 CLIVE 0.731 1.040 6.715 CFISH 0.813 1.186 5.448 CFOHU 0.590 0.902 CMINE 0.042 0.034 CMEAT 0.813 0.636 CFOOD 0.440 0.351 CGRAI 1.892 1.476 CBEVT 0.880 0.682 CTEXT 1.287 0.058 0.394 0.672 6.440 0.299 0.136 0.042 0.014 0.091 0.011 0.013 0.035 0.094 0.027 0.369 0.286 CWEAR CLEAT CWOOD CPAPE 0.218 CCHEM 6.315 1.945 1.090 0.696 0.603 6.341 0.906 0.050 0.017 0.110 0.013 0.015 0.042 24.516 0.113 0.033 CFERT 1.360 0.418 0.235 0.151 0.131 1.383 0.198 0.010 0.003 0.023 0.003 0.003 0.009 1.764 0.023 0.007 CFUEL 0.114 0.008 0.811 0.105 1.253 0.006 0.005 CRUBB CPLAS CGLAS CCEME CIRON CFMPR 0.307 0.065 2.345 0.102 0.081 CMAEQ 0.046 0.011 0.312 0.015 0.012 CELEQ CTREQ 0.228 0.022 0.002 0.037 0.052 0.042 COMAN CELEC 1.635 CWATE 0.498 CCNST CCOME 1.703 1.262 0.342 1.129 0.031 0.187 0.665 0.527 0.371 0.066 CCOMD CCOMI CTRAD 0.162 0.303 0.338 0.047 0.069 0.004 0.568 3.380 0.396 0.004 0.001 0.010 0.001 0.001 0.004 3.790 0.010 0.003 0.505 0.404 CHORE CTR_C 0.644 0.074 0.108 0.217 0.047 3.879 3.782 0.031 0.010 0.068 0.008 0.010 0.026 14.982 0.070 0.020 0.150 0.120 CFI_I 0.863 1.232 1.289 0.207 0.120 0.001 0.083 0.085 0.002 0.001 0.005 0.001 0.001 0.002 2.432 0.005 0.002 0.466 0.372 CREAL CBUSI CPUBA COSER UPRO UWCO UBCO UNSK RURA 8.406 0.656 2.270 5.159 5.582 6.071 101.377 2.581 22.863 8.160 1.895 34.243 18.296 21.793 19.178 16.977 68.079 17.319 44.725 33.176 LAND 1.107 0.086 0.279 0.631 0.731 0.794 5.092 0.277 1.278 0.661 0.152 2.836 1.491 1.780 1.567 1.295 5.790 1.411 4.463 3.276 CAPITAL 0.752 0.058 0.186 0.424 0.493 0.534 3.875 0.187 0.885 0.446 0.102 1.973 1.014 1.215 1.070 0.815 4.171 0.962 3.154 2.283 ENTR HHUFA HHUNF HHRFA HHRNF GOVR GOVI ITAX 1.653 0.191 TTAX KACCOUN DST WORLD TOTAL 20.897 5.243 8.533 9.935 12.913 9.986 146.900 9.550 41.436 12.526 3.226 46.522 21.638 25.793 25.521 83.095 86.296 22.283 64.854 48.875 TABLE A1: TANZANIA MICROSAM 1992 IN CURRENT TSHS MILLION (CONT.) AFOHU AMINE AMEAT AFOOD AGRAI ABEVT ATEXT AWEAR ALEAT AWOOD APAPE ACHEM AFERT AFUEL ARUBB APLAS AGLAS ACEME AIRON AFMPR ACOTT ASISA ATEA ACOFF ATOBA ACASH AMAIZ AWHEA APADD ASORG AOCER ABEAN ACASS AROOT AOILS ASUGA AOHOR AOCRO ALIVE AFISH AFOHU AMINE AMEAT AFOOD AGRAI ABEVT ATEXT AWEAR ALEAT AWOOD APAPE ACHEM AFERT AFUEL ARUBB APLAS AGLAS ACEME AIRON AFMPR AMAEQ AELEQ ATREQ AOMAN AELEC AWATE ACNST ATRAD ATOUR AHORE ATR_C AFI_I AREAL ABUSI APUBA AOSER CCOTT 3.538 CSISA 2.344 CTEA 0.743 CCOFF 1.516 CTOBA 11.243 CCASH 1.238 CMAIZ 34.508 9.184 CWHEA 0.935 7.109 CPADD 23.036 CSORG COCER 1.260 CBEAN CCASS CROOT COILS 1.140 0.133 CSUGA 24.315 11.008 COHOR 5.768 11.400 COCRO 0.811 6.366 CLIVE 1.943 1.358 1.917 CFISH 2.193 CFOHU 2.289 0.098 1.169 2.408 4.723 0.788 0.118 CMINE 0.014 0.326 0.011 1.165 8.057 0.015 CMEAT 0.261 0.429 CFOOD 0.148 CGRAI 0.605 CBEVT 0.273 0.003 0.018 0.004 0.001 CTEXT 0.120 0.414 1.147 0.736 0.661 CWEAR 3.285 0.252 CLEAT 0.129 CWOOD 0.994 2.003 1.256 0.781 0.028 0.121 0.798 0.583 CPAPE 0.230 0.891 0.633 4.955 1.281 1.266 0.490 0.017 0.004 0.588 0.123 0.075 0.505 0.500 0.366 CCHEM 0.188 0.048 0.095 4.353 0.184 1.752 1.049 3.197 0.006 0.175 CFERT CFUEL 0.002 0.291 0.196 0.150 1.306 0.285 0.030 0.119 0.005 0.050 0.030 0.047 0.002 0.108 0.235 0.049 0.030 0.201 0.035 0.025 CRUBB 0.061 3.178 CPLAS 0.709 1.454 0.061 0.253 0.011 0.098 0.062 0.856 0.030 2.170 0.054 0.036 CGLAS 0.248 0.516 0.032 0.288 0.010 0.412 CCEME 2.836 CIRON 20.631 10.638 CFMPR 0.034 1.595 1.136 2.322 0.521 0.101 1.005 0.745 0.026 0.126 0.223 CMAEQ 0.005 0.211 0.159 1.323 0.306 0.089 0.348 0.015 0.143 0.086 0.110 0.004 0.005 0.144 0.031 0.019 0.128 0.758 CELEQ 0.582 0.391 3.171 0.828 0.192 0.921 0.037 0.345 0.216 0.274 0.010 0.012 0.362 0.076 0.047 0.308 CTREQ 0.018 0.150 0.101 0.076 0.148 0.046 0.180 0.008 0.073 0.045 0.060 0.043 0.211 0.027 0.182 0.107 COMAN 0.041 0.346 0.079 0.023 0.090 0.004 0.037 0.021 0.028 0.001 0.001 0.036 0.008 0.005 0.033 0.759 CELEC 0.975 1.556 1.068 7.550 2.059 0.479 2.206 0.095 0.895 0.541 0.659 0.024 0.024 0.666 0.142 0.087 0.563 0.536 0.396 CWATE 0.281 0.450 0.303 0.561 0.172 0.636 0.030 0.263 0.164 0.203 0.007 0.007 0.201 0.027 0.180 0.184 0.132 CCNST 0.099 0.009 0.052 0.387 0.014 0.011 0.073 0.008 CCOME 0.181 0.090 0.303 0.091 1.542 6.790 0.457 1.181 0.250 0.039 0.020 0.054 0.411 0.521 CCOMD CCOMI CTRAD 0.170 0.605 2.968 1.987 13.005 3.693 0.838 3.719 0.170 1.531 0.953 1.740 0.065 0.017 1.313 0.282 0.174 1.101 1.167 0.868 CHORE 1.344 0.957 1.951 0.391 0.726 0.069 0.502 0.365 CTR_C 0.050 0.970 2.336 1.572 10.444 2.944 0.240 1.071 0.049 0.442 0.272 0.804 0.030 0.015 1.042 0.224 0.138 0.877 0.402 0.298 CFI_I 0.156 5.367 3.399 18.409 5.768 2.163 8.799 0.456 3.754 2.495 2.305 0.091 1.304 2.619 0.577 0.357 2.150 1.266 0.960 CREAL 1.227 2.750 12.795 0.545 5.157 3.106 2.930 0.106 1.551 3.260 0.690 1.678 CBUSI 0.071 0.410 0.289 0.572 0.144 0.681 0.028 0.272 0.162 0.153 0.005 0.080 0.169 0.035 0.022 0.145 0.097 0.069 CPUBA 0.987 COSER 1.436 0.105 0.502 0.184 1.252 0.022 0.180 0.532 0.024 0.099 0.596 0.368 UPRO 0.556 0.627 0.377 0.318 0.739 0.167 0.260 0.026 0.616 0.115 0.056 0.002 0.015 0.023 0.014 0.020 0.103 0.054 0.106 UWCO 0.202 0.469 0.294 0.248 0.576 0.130 0.201 0.021 0.484 0.090 0.046 0.002 0.012 0.019 0.011 0.016 0.082 0.043 0.084 UBCO 29.387 5.105 3.271 2.944 6.608 1.367 2.148 0.215 5.010 0.940 0.469 0.017 0.124 0.190 0.114 0.165 0.851 0.449 0.873 UNSK 1.318 0.665 0.398 0.335 0.864 0.181 0.279 0.026 0.667 0.123 0.061 0.002 0.016 0.023 0.014 0.021 0.110 0.058 0.114 RURA 54.402 7.603 4.660 3.597 8.750 LAND 5.454 CAPITAL 3.880 11.087 12.885 8.163 7.068 16.206 2.869 4.426 0.448 10.412 1.978 2.363 0.088 0.635 0.951 0.574 0.840 4.288 2.241 4.399 ENTR HHUFA HHUNF HHRFA HHRNF GOVR GOVI ITAX 3.976 10.398 0.341 0.451 0.535 0.431 0.176 0.414 0.044 0.350 0.124 0.187 0.005 0.037 0.137 0.032 0.035 0.209 0.211 0.188 TTAX KACCOUN DST WORLD TOTAL 72.038 59.981 50.922 72.322 146.573 108.136 20.981 55.226 5.475 40.660 15.466 21.073 0.599 4.017 15.408 5.209 4.063 23.491 31.608 24.243 TABLE A1: TANZANIA MICROSAM 1992 IN CURRENT TSHS MILLION (CONT.) AMAEQ AELEQ ATREQ AOMAN AELEC AWATE ACNST ATRAD ATOUR AHORE ATR_C AFI_I AREAL ABUSI APUBA AOSER CCOTT CSISA CTEA CCOFF ACOTT 3.538 ASISA 2.428 ATEA 4.702 ACOFF 4.047 ATOBA ACASH AMAIZ AWHEA APADD ASORG AOCER ABEAN ACASS AROOT AOILS ASUGA AOHOR AOCRO ALIVE AFISH AFOHU AMINE AMEAT AFOOD AGRAI ABEVT ATEXT AWEAR ALEAT AWOOD APAPE ACHEM AFERT AFUEL ARUBB APLAS AGLAS ACEME AIRON AFMPR AMAEQ AELEQ ATREQ AOMAN AELEC AWATE ACNST ATRAD ATOUR AHORE ATR_C AFI_I AREAL ABUSI APUBA AOSER CCOTT CSISA 0.004 CTEA CCOFF CTOBA CCASH CMAIZ 2.557 CWHEA CPADD CSORG COCER 0.184 CBEAN 10.672 CCASS 6.464 CROOT 6.040 COILS 4.594 CSUGA COHOR 1.742 18.215 COCRO 0.004 0.247 3.100 CLIVE 2.373 17.938 CFISH 2.690 22.360 CFOHU 5.634 19.421 CMINE 0.001 61.934 CMEAT 0.285 0.451 CFOOD 0.161 0.255 CGRAI 0.787 CBEVT 0.015 1.880 0.297 0.034 0.010 0.002 0.036 CTEXT 1.444 0.416 0.388 0.833 CWEAR 0.547 0.134 0.134 0.369 CLEAT CWOOD 0.133 0.264 1.275 0.032 0.444 3.453 0.148 0.315 0.337 0.071 7.483 0.727 CPAPE 0.083 0.165 0.798 0.020 0.109 0.119 0.296 0.194 2.663 0.841 0.196 0.044 5.026 0.459 CCHEM 0.255 0.138 0.148 0.691 2.292 0.123 0.262 0.286 0.019 11.871 2.793 CFERT 1.552 CFUEL 0.005 0.010 0.005 0.001 0.559 0.446 0.682 0.725 0.025 0.055 8.618 0.098 0.029 0.006 1.591 0.153 CRUBB 0.006 23.842 CPLAS 0.300 0.023 2.524 CGLAS 0.003 0.005 1.299 0.018 0.965 CCEME 22.141 CIRON 0.113 0.228 1.304 0.242 13.860 CFMPR 0.096 0.191 0.724 4.169 0.133 1.373 CMAEQ 0.692 0.108 0.150 0.603 0.123 0.010 0.020 0.025 0.039 0.012 0.002 0.157 0.033 CELEQ 1.738 0.268 0.095 1.424 0.299 0.023 0.048 0.079 0.096 0.028 0.006 0.398 0.085 CTREQ 0.069 0.138 0.004 0.057 0.957 0.544 0.042 0.088 2.941 0.156 0.046 0.010 1.743 0.200 COMAN 0.692 1.372 0.027 0.075 0.166 0.031 0.002 0.005 0.012 0.001 0.039 0.008 CELEC 0.093 0.183 0.302 0.022 1.268 0.445 4.857 0.298 0.677 5.738 1.538 0.403 0.090 7.476 2.889 CWATE 0.028 0.056 0.093 7.799 0.117 1.060 0.092 0.197 1.381 0.436 0.123 0.028 1.611 0.834 CCNST 0.027 1.571 1.297 1.440 1.221 0.095 0.202 15.727 6.981 1.755 0.389 5.221 1.973 CCOME CCOMD 0.373 2.616 CCOMI CTRAD 0.208 0.408 1.633 0.050 1.261 1.106 2.334 4.940 0.407 0.768 0.217 0.049 7.054 0.654 CHORE 0.083 0.165 0.458 6.742 1.392 12.084 0.290 0.085 0.019 6.201 0.348 CTR_C 0.071 0.140 0.275 0.017 1.562 1.362 1.820 59.530 4.634 10.179 20.132 3.929 0.985 0.244 10.627 1.023 CFI_I 0.242 0.471 1.906 0.059 0.366 0.718 4.874 0.677 1.350 2.907 12.391 0.449 2.283 CREAL 0.288 0.569 2.377 0.070 0.430 0.997 15.425 0.880 2.013 6.635 2.393 3.268 CBUSI 0.015 0.029 0.124 0.004 0.022 0.082 0.972 0.045 0.096 0.400 3.852 0.959 0.162 0.175 CPUBA 10.879 COSER 0.053 0.115 0.308 0.012 0.745 0.023 0.048 0.748 0.706 0.083 3.974 8.966 UPRO 0.018 0.034 0.036 0.004 1.303 0.222 1.943 27.646 2.048 6.448 10.673 6.039 8.899 1.166 34.250 12.710 UWCO 0.015 0.027 0.030 0.003 1.711 0.291 0.783 17.891 1.240 3.936 10.994 3.715 5.474 0.706 8.758 4.150 UBCO 0.153 0.280 0.300 0.036 5.181 0.880 17.814 3.331 0.230 0.742 20.403 0.886 1.303 0.168 7.176 3.055 UNSK 0.020 0.037 0.039 0.004 0.476 0.083 3.165 25.099 1.778 5.628 7.822 0.669 0.983 0.128 4.360 1.655 RURA LAND CAPITAL 0.778 1.420 1.516 0.188 23.152 3.989 54.661 121.052 4.835 14.279 50.501 33.858 49.876 6.662 49.678 1.140 ENTR HHUFA HHUNF HHRFA HHRNF GOVR GOVI ITAX 0.094 0.054 0.093 0.007 0.109 0.340 0.943 0.146 3.822 0.786 0.159 0.015 0.023 0.315 0.179 0.588 TTAX KACCOUN DST WORLD TOTAL 4.047 8.100 15.821 1.106 50.110 18.257 200.959 297.350 26.448 61.463 205.397 73.627 75.686 10.765 295.753 51.133 3.538 2.428 5.254 7.251 TABLE A1: TANZANIA MICROSAM 1992 IN CURRENT TSHS MILLION (CONT.) CTOBA CCASH CMAIZ CWHEA CPADD CSORG COCER CBEAN CCASS CROOT COILS CSUGA COHOR COCRO CLIVE CFISH CFOHU CMINE CMEAT CFOOD ACOTT ASISA ATEA ACOFF ATOBA 11.715 ACASH 6.075 AMAIZ 78.280 AWHEA 9.471 APADD 37.287 ASORG 6.011 AOCER 2.078 ABEAN 25.173 ACASS 6.575 AROOT 9.634 AOILS 24.025 ASUGA 58.754 AOHOR 69.271 AOCRO 16.864 ALIVE 36.809 AFISH 43.745 AFOHU 46.886 AMINE 58.316 AMEAT 50.091 AFOOD 29.499 AGRAI ABEVT ATEXT AWEAR ALEAT AWOOD APAPE ACHEM AFERT AFUEL ARUBB APLAS AGLAS ACEME AIRON AFMPR AMAEQ AELEQ ATREQ AOMAN AELEC AWATE ACNST ATRAD ATOUR AHORE ATR_C AFI_I AREAL ABUSI APUBA AOSER CCOTT CSISA CTEA CCOFF CTOBA CCASH CMAIZ CWHEA CPADD CSORG COCER CBEAN CCASS CROOT COILS CSUGA COHOR COCRO CLIVE CFISH CFOHU CMINE CMEAT CFOOD CGRAI CBEVT CTEXT CWEAR CLEAT CWOOD CPAPE CCHEM CFERT CFUEL CRUBB CPLAS CGLAS CCEME CIRON CFMPR CMAEQ CELEQ CTREQ COMAN CELEC CWATE CCNST CCOME CCOMD 3.351 1.858 17.269 4.809 1.011 3.388 2.732 1.616 3.693 17.102 14.610 1.466 9.539 15.478 6.946 9.829 2.132 CCOMI 0.058 0.055 0.272 0.117 0.219 5.526 CTRAD CHORE CTR_C CFI_I CREAL CBUSI CPUBA COSER UPRO UWCO UBCO UNSK RURA LAND CAPITAL ENTR HHUFA HHUNF HHRFA HHRNF GOVR GOVI ITAX 0.107 0.574 0.005 0.004 0.073 0.214 3.531 4.233 TTAX 0.387 0.019 0.006 0.021 0.080 0.044 0.528 0.132 2.714 KACCOUN DST WORLD 0.625 3.938 1.376 0.216 0.192 0.953 0.410 12.657 1.546 40.453 TOTAL 15.066 7.934 95.549 10.203 37.287 10.819 7.987 28.561 9.307 12.645 28.002 75.856 84.154 18.330 47.725 59.223 54.402 71.715 65.348 84.557 TABLE A1: TANZANIA MICROSAM 1992 IN CURRENT TSHS MILLION (CONT.) CGRAI CBEVT CTEXT CWEAR CLEAT CWOOD CPAPE CCHEM CFERT CFUEL CRUBB CPLAS CGLAS CCEME CIRON CFMPR CMAEQ CELEQ CTREQ COMAN ACOTT ASISA ATEA ACOFF ATOBA ACASH AMAIZ AWHEA APADD ASORG AOCER ABEAN ACASS AROOT AOILS ASUGA AOHOR AOCRO ALIVE AFISH AFOHU AMINE AMEAT AFOOD AGRAI 146.501 ABEVT 107.589 ATEXT 7.765 AWEAR 39.948 ALEAT 4.283 AWOOD 33.574 APAPE 13.966 ACHEM 21.073 AFERT 0.599 AFUEL 4.017 ARUBB 15.298 APLAS 5.088 AGLAS 3.738 ACEME 21.023 AIRON 30.376 AFMPR 21.116 AMAEQ 4.047 AELEQ 8.100 ATREQ 15.821 AOMAN 1.106 AELEC AWATE ACNST ATRAD ATOUR AHORE ATR_C AFI_I AREAL ABUSI APUBA AOSER CCOTT CSISA CTEA CCOFF CTOBA CCASH CMAIZ CWHEA CPADD CSORG COCER CBEAN CCASS CROOT COILS CSUGA COHOR COCRO CLIVE CFISH CFOHU CMINE CMEAT CFOOD CGRAI CBEVT CTEXT CWEAR CLEAT CWOOD CPAPE CCHEM CFERT CFUEL CRUBB CPLAS CGLAS CCEME CIRON CFMPR CMAEQ CELEQ CTREQ COMAN CELEC CWATE CCNST CCOME CCOMD 11.253 1.994 0.720 0.157 4.302 1.723 2.576 0.918 0.458 0.473 2.691 2.213 0.509 0.971 1.916 0.128 CCOMI 0.451 2.124 0.438 0.601 0.115 1.233 5.175 0.409 1.321 3.990 0.237 0.453 2.618 7.375 5.727 18.085 2.267 CTRAD CHORE CTR_C CFI_I CREAL CBUSI CPUBA COSER UPRO UWCO UBCO UNSK RURA LAND CAPITAL ENTR HHUFA HHUNF HHRFA HHRNF GOVR GOVI ITAX 0.053 10.963 2.466 1.479 0.533 0.011 0.386 0.865 0.001 0.430 1.013 0.139 0.122 4.609 0.096 1.184 2.376 1.406 4.282 0.808 TTAX 0.163 0.239 2.145 0.633 0.448 0.019 0.292 0.650 0.001 0.343 0.460 0.203 0.070 0.207 0.871 0.589 1.425 1.121 9.158 0.483 KACCOUN DST WORLD 13.990 3.158 14.267 2.815 4.183 0.867 9.222 39.196 6.408 25.743 9.714 30.564 1.762 3.280 29.348 19.591 55.394 42.997 78.081 16.995 TOTAL 160.707 133.654 30.760 46.033 10.205 38.887 26.821 69.535 7.009 30.942 28.724 40.442 6.402 32.263 60.691 47.311 71.126 60.322 127.343 21.786 TABLE A1: TANZANIA MICROSAM 1992 IN CURRENT TSHS MILLION (CONT.) CELEC CWATE CCNST CCOME CCOMD CCOMI CTRAD CHORE CTR_C CFI_I CREAL CBUSI CPUBA COSER UPRO UWCO UBCO UNSK RURA LAND ACOTT ASISA ATEA ACOFF ATOBA ACASH AMAIZ AWHEA APADD ASORG AOCER ABEAN ACASS AROOT AOILS ASUGA AOHOR AOCRO ALIVE AFISH AFOHU AMINE AMEAT AFOOD AGRAI ABEVT ATEXT AWEAR ALEAT AWOOD APAPE ACHEM AFERT AFUEL ARUBB APLAS AGLAS ACEME AIRON AFMPR AMAEQ AELEQ ATREQ AOMAN AELEC 50.110 AWATE 18.257 ACNST 200.959 ATRAD 18.213 152.819 58.863 67.455 ATOUR AHORE 50.451 ATR_C 196.261 AFI_I 73.556 AREAL 75.686 ABUSI 10.765 APUBA 295.594 AOSER 25.224 CCOTT CSISA CTEA CCOFF CTOBA CCASH CMAIZ CWHEA CPADD CSORG COCER CBEAN CCASS CROOT COILS CSUGA COHOR COCRO CLIVE CFISH CFOHU CMINE CMEAT CFOOD CGRAI CBEVT CTEXT CWEAR CLEAT CWOOD CPAPE CCHEM CFERT CFUEL CRUBB CPLAS CGLAS CCEME CIRON CFMPR CMAEQ CELEQ CTREQ COMAN CELEC CWATE CCNST CCOME CCOMD CCOMI CTRAD CHORE CTR_C CFI_I CREAL CBUSI CPUBA COSER UPRO UWCO UBCO UNSK RURA LAND CAPITAL ENTR HHUFA 102.358 8.606 HHUNF 62.914 58.158 111.930 52.966 5.721 HHRFA 402.218 31.846 HHRNF 20.809 4.595 10.255 4.254 7.519 GOVR GOVI ITAX 0.184 0.556 0.004 TTAX KACCOUN DST WORLD 7.376 5.054 25.263 45.729 33.912 TOTAL 50.295 18.813 200.959 18.213 152.819 58.863 67.455 57.827 201.315 98.819 121.415 10.765 295.594 25.229 117.635 62.753 122.186 57.220 517.817 40.452 TABLE A1: TANZANIA MICROSAM 1992 IN CURRENT TSHS MILLION (CONT.) CAPITAL ENTR HHUFA HHUNF HHRFA HHRNF GOVR GOVI ITAX TTAX KACCOUN DST WORLD TOTAL ACOTT 17.359 20.897 ASISA 2.815 5.243 ATEA 3.831 8.533 ACOFF 5.888 9.935 ATOBA 1.199 12.913 ACASH 3.911 9.986 AMAIZ 3.272 1.711 62.058 1.502 0.076 146.900 AWHEA 0.079 9.550 APADD 0.145 0.076 3.852 0.075 41.436 ASORG 0.413 0.222 5.665 0.198 0.017 12.526 AOCER 0.202 0.117 0.717 0.094 0.018 3.226 ABEAN 0.543 0.356 19.011 0.953 0.486 46.522 ACASS 0.333 0.197 13.431 0.568 0.534 21.638 AROOT 0.385 0.233 14.819 0.668 0.053 25.793 AOILS 1.496 25.521 ASUGA 23.849 0.492 83.095 AOHOR 1.207 0.716 12.569 0.930 1.603 86.296 AOCRO 0.181 0.174 4.126 0.104 0.834 22.283 ALIVE 1.229 0.747 25.345 0.494 0.230 64.854 AFISH 0.239 0.098 3.233 0.084 1.477 48.875 AFOHU 1.571 0.843 21.639 0.645 0.453 72.038 AMINE 1.665 59.981 AMEAT 0.831 50.922 AFOOD 0.347 0.232 39.715 0.712 1.818 72.322 AGRAI 0.072 146.573 ABEVT 0.547 108.136 ATEXT 13.216 20.981 AWEAR 15.278 55.226 ALEAT 1.191 5.475 AWOOD 7.086 40.660 APAPE 1.501 15.466 ACHEM 21.073 AFERT 0.599 AFUEL 4.017 ARUBB 0.110 15.408 APLAS 0.122 5.209 AGLAS 0.325 4.063 ACEME 2.468 23.491 AIRON 1.232 31.608 AFMPR 3.126 24.243 AMAEQ 4.047 AELEQ 8.100 ATREQ 15.821 AOMAN 1.106 AELEC 50.110 AWATE 18.257 ACNST 200.959 ATRAD 297.350 ATOUR 26.448 26.448 AHORE 11.012 61.463 ATR_C 9.136 205.397 AFI_I 0.071 73.627 AREAL 75.686 ABUSI 10.765 APUBA 0.159 295.753 AOSER 25.909 51.133 CCOTT 3.538 CSISA 2.428 CTEA 1.016 1.674 1.686 0.134 5.254 CCOFF 1.037 1.760 2.096 0.131 7.251 CTOBA 0.912 1.454 1.223 0.121 15.066 CCASH 1.090 1.943 3.533 0.130 7.934 CMAIZ 12.930 6.350 10.883 1.300 0.005 95.549 CWHEA 0.014 0.017 0.076 0.003 0.033 10.203 CPADD 0.941 1.620 0.807 0.157 37.287 CSORG 1.202 3.225 3.103 0.169 10.819 COCER 0.604 1.562 3.257 0.090 7.987 CBEAN 2.394 3.981 4.273 0.266 28.561 CCASS 0.538 0.764 0.640 0.100 9.307 CROOT 1.470 2.104 1.806 0.263 12.645 COILS 3.901 6.998 7.746 0.569 -0.001 28.002 CSUGA 6.789 10.490 11.329 1.862 0.615 75.856 COHOR 10.630 15.831 11.312 1.843 84.154 COCRO 1.381 2.146 1.793 0.203 0.149 18.330 CLIVE 2.242 3.375 4.023 0.612 3.459 47.725 CFISH 5.393 7.035 10.796 1.309 59.223 CFOHU 2.961 4.566 8.055 0.680 54.402 CMINE 0.115 71.715 CMEAT 10.523 22.686 26.640 2.842 -0.216 65.348 CFOOD 9.125 17.928 49.366 2.920 3.863 84.557 CGRAI 25.592 55.897 70.422 6.779 -2.743 160.707 CBEVT 15.025 12.843 91.213 8.672 1.768 133.654 CTEXT 1.569 2.323 5.813 0.562 2.814 1.253 30.760 CWEAR 5.367 8.258 25.573 1.818 0.296 46.033 CLEAT 1.312 2.029 6.369 0.444 -0.078 10.205 CWOOD 0.424 0.782 1.333 0.127 14.548 0.428 38.887 CPAPE 0.675 1.057 2.132 0.230 -0.429 26.821 CCHEM 3.013 5.968 7.672 0.978 -20.827 69.535 CFERT -0.263 7.009 CFUEL 1.971 4.364 5.563 0.539 30.942 CRUBB 0.053 0.102 0.307 0.015 2.232 -1.071 28.724 CPLAS 2.170 4.250 9.734 0.616 15.180 -0.149 40.442 CGLAS 0.205 0.409 1.152 0.047 0.889 -0.096 6.402 CCEME 0.630 1.572 4.231 0.124 0.730 32.263 CIRON 13.674 60.691 CFMPR 0.836 1.844 2.368 0.222 26.869 -2.248 47.311 CMAEQ 0.006 0.014 0.025 0.002 62.220 2.602 71.126 CELEQ 0.421 0.838 0.868 0.112 46.448 -0.723 60.322 CTREQ 1.060 2.027 2.083 0.283 112.890 0.145 127.343 COMAN 0.868 1.861 2.017 0.249 12.924 -0.075 21.786 CELEC 0.075 0.096 0.250 0.015 1.425 50.295 CWATE 0.198 0.127 0.330 0.040 -0.033 18.813 CCNST 0.011 0.013 0.035 0.002 52.521 109.703 0.123 200.959 CCOME 18.213 CCOMD 152.819 CCOMI 58.863 CTRAD 67.455 CHORE 1.156 10.507 10.006 1.826 0.161 57.827 CTR_C 3.054 4.915 11.759 5.577 8.669 2.340 201.315 CFI_I 0.075 0.166 0.296 0.020 0.006 98.819 CREAL 10.853 12.330 24.370 2.718 0.005 121.415 CBUSI 0.101 0.194 0.199 0.027 -0.098 10.765 CPUBA 0.875 1.299 2.292 0.183 279.080 295.594 COSER 0.572 1.263 2.252 0.061 25.229 UPRO 117.635 UWCO 62.753 UBCO 122.186 UNSK 57.220 RURA 517.817 LAND 40.452 CAPITAL 537.985 ENTR 508.398 27.620 536.018 HHU