Afghanistan - Journey to economic development (Vol. 2 of 2) : Volume 2 : Source material and statistics on the economy of Afghanistan

Report No. 1777a-AF Afghanistan L OP The Journey to Economic Development (In Two Volumes) Volume II: Source Material and Statistics on the Economy of Afghanistan March 17, 1978 Country Programs Department I Europe, Middle East and North Africa Region FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Calculations in this report have been based on the average free rate for 1976/77 as follows: US$1.00 = Af 47.38 Af 1.00 = US$0.0211 By December 31, 1977 the rate had appreciated to Af 43 to the dollar WEIGHTS AND MEASURES 1 Jerib - 0.2 hectare 1 Seer = 7.14 kilogram ABBREVIATIONS AFC - Afghan Fertilizer Company ASC - Afghan Seed Company AgBank - Agricultural Development Bank DAB - Da Afghanistan Bank DAP - Development Assistance Fund FAO - Food and Agricultural Organization FPD - Food Procurement Department GM - Government Monopoly HAVA - Helmand-Arghandab Valley Authority IDA - International Development Association IDBA - Industrial Development Bank of Afghanistan MOPH - Ministry of Public Health NCD - Northern Coal Department NOC - National Oil Corporation Plan - First 7-Year Development Plan - 1976/77-1982/83 RDD - Rural Development Department UNDP - United Nations Development Program WAPM - Water and Power Ministry WFP - World Food Program FISCAL YEAR March 21 through March 20 POPULATION Throughout this Report an estimate of the mid-1977 population of 14 million has been used. This figure represents the best judgment of IBRD staff drawing on a number of different soures. The derivation of the figure is explained In detall In Volume 1. It should be noted that this figure differ* from wbat io given In the official publications of the Government of Afghaist a.  FOR OFFICIAL USE ONLY AFGHANISTAN THE JOURNEY TO ECONOMIC DEVELOPMENT VOLUME II SOURCE MATERIAL AND STATISTICS ON THE ECONOMY OF AFGHANISTAN Table of Contents Page No. PART I: SURVEY OF THE STATISTICS OF AFGHANISTAN i PART II: STATISTICAL APPENDIX 41 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.   PART I SURVEY OF THE STATISTICS OF AFGHANISTAN Table of Contents Page No. I. INTRODUCTION ....... ................................. 1 II. SUMMARY OF FINDINGS ...... ........................... 2 III. REVIEW OF AFGHAN STATISTICS . ........................ 5 A. Statistical Publications ...... ................. 5 B. Basic Statistics ........ ....................... 6 C. National Accounts ............... . .............. 19 IV. THE STATISTICAL SYSTEM .................... ........... 23 A. The Statistical Law .............. .............. 23 B. Organization and Co-ordination ................ . 24 C. Staffing of the CSO ............................ 28 D Training in Statistics ........... .. ............ 28 E. Data Processing ................................ 29 V. FUTURE PROGRAM OF WORK OF THE CSO ................... 30 Annex I: Statistical Law of Afghanistan ... 35 -1i-   SURVEY OF THE STATISTICS OF AFGHANISTAN I. INTRODUCTION 1. Afghanistan's statistics are notoriously inadequate for economic analysis and evaluation. The IBRD and other organizations have over the years commented that the absence of national income and population statistics and the poor data base in general made it difficult to quantify with any measure of reliability the overall trends in the Afghan economy. The Min- istry of Planning has also been pressing continuously for the collection and analysis of reliable statistical data needed to design and implement effective social and economic development plans. 2. It is not that the Government of Afghanistan is unaware of the problem. There were various reports and recommendations in the past for the improvement of Afghan statistics. 1/ The Government has also been making considerable use of experts from the UN Agencies and from various countries, and has sent abroad a number of Afghans for training in different fields of statistics since 1952. 3. Prior to 1974 data was compiled by the statistical offices of the various Ministries and Government Agencies (about 25 in all). In 1972, however, the Central Statistics Office (CSO) was established in the Prime Ministry to strengthen and centralize the statistical activities. All the statistical units of the Ministries were moved to the CSO in 1974 with the exception of those in the Ministries of Education and Health, and the Depart- ment of Statistics of the Ministry of Interior which has special responsi- bilities for population registration data. In 1975 the Statistical Law of Afghanistan (see Annex I) was promulgated for the purposes of creating a scientific system of statistics, centralization and unification of all statistical activities; collecting, comparing, analyzing and evaluation; and publication of reliable statistical information. The Law gives wide powers to the CSO for developing statistical programs and compiling statistics, and proclaims that "official statistics shall be the data and information which have been published, distributed or verified by CSO". At present the CSO is consolidating its authority, compiling and publishing available data, and planning for a population census of Afghanistan in the near future. 1/ a. Public Administration Service, Chicago - Survey of Statistical Reporting in the Royal Afghan Government with recommendations on improving present procedures (May 1958). b. Schiro, Bruno A. (US Bureau of Census) - The Statistical System of Afghanistan - a general overview with guidelines for improvement (prepared at the request of the Minister of Planning), USAID/Kabul (1970?). c. Tiwari, S. G. (Regional Adviser on National Accounts, ESCAP) - Report on Development of National Accounts and Long Range Program of Development of Statistics in the Republic of Afghanistan. (1974) - 1 -  4. The following report originated from the major findings of the Reconnaissance Mlission to Afghanistan in preparation fcr a Basic Economic Mlission in early 1977. The mission was in the field during August 12-26, 1976. The objectives of the statistical part of the mission uiere to study the statistical system of Afghanistan in order to assess the reliability of the existing data, identify the gaps in the system, and gather as much exist- ing data as possible. Following the Basic Economic Mlission in April-May, 1977, the earlier report has been considerably revised and updated benefit- ting from discussions with the statistical personnel and interested users in Afghanistan. 5. The missions did not attempt a full-fledged review of the statistical system. Rather an atterapt has been made to obtain as much factual information as possible on the present availability of data, par- ticularly in important sectors such as agriculture, industry, foreign trade, national accounts, etc. However, considerable difficulties were encountered as the publications were mostly in Dari, and contained minimal descriptions of concepts, definitions, and coverage of data. The missions, therefore, had to rely on the patience and cooperation of the officials to explain at length the various aspects of statistical programs. II. SUI-NARY OF FIIDINUCS 6. The statistical system of Afghanistan produces a great variety of statistical information as can be seen from the annual publication of the Central Statistics Office (CSO) - "The Statistical Information of Afghan- istan". The CSO compiles data on manpower, industry and foreign trade and conducts various surveys, Da Afghanistan Bank (DAB) prepares the balance of payments accounts and compile financial data and retail prices, the ISin- istry of Finance provides the central governnent accounts, the Ministry of Agriculture compiles information on agricultural output, and the Mlinistries of Education and Health collect elaborate statistics in their own spheres. The CSO, as the central co-ordinating authority for official statistics, is responsible for setting the standards and publishing the statistics. 7. In recent years important steps have been taken to improve Afghan statistics. The CSO is preparing for a population census in mid-1978. It has conducted surveys of wheat, sugarcane and apiculture, handicrafts, and banking, and instituted a montlily survey of modern industries including mines and electricity generation. DAB is preparing a comparable time series of external trade data, and export and import price indices. The Mlinistry of Planning has attempted the calculation of GNP despite lack of adequate data. 8. The system, however, still has major shortcomings due to the lack of: a) couLnrehensive economic statistics and national accounts; b) coor- dination among producers of statistics; c) user-orientation of data; d) an integrated pro2ran of statistical developr;ent; and e) qualified r.anploWer. Since the CSO is rather new and still undergoing the process of consolidation, it has not yet been able to ncet its obli-ation to provide the users of statistics witlh compreliensive data. A small example of this relates to  -3- publication of statistics. The statistical publications of the CSO are mainly annuals with a quarterly review of special interest topics, although monthly and quarterly internal data documents are prepared for circulation to interested Government agencies. Current information should be made available to all users, and the CSO should consider regularly publishing a compendium of current data in its quarterly review. 9. To summarize briefly the coverage and quality of statistics in the major categories, current population data are consolidated estimates derived from various population surveys and administrative records. While the vital statistics appear reasonable, a reliable estimate of total population may have to wait for the census. The preliminary results of the census due to be conducted in September, 1978 should be available by March, 1979 and the details in 1981. Manpower data for industry and agriculture are not completely reli- able since they are obtained from surveys of partial coverage. Other employ- ment data are mostly approximations. The sources of current agricultural statistics are the quarterly progress reports and annual reports on harvests by the agricultural extension agents. Although the data on major crops have a quantitative base, the reporting is on a sampling basis, and the agents are not trained enumerators. Minor crops, livestock and forestry data are rough estimates. In the industry sector, monthly surveys provide data on modern industries. However, due to non-response the data are incomplete and may not be comparable over time. Handicrafts data cover only 11 of the 26 provinces. Government Accounts are among the most reliable statistics in Afghanistan. But the Ministry of Finance only prepares the Central Government accounts. Attention needs to be paid to compilation of municipal and especially public enterprise acocunts. External trade and balance-of-payments data are subject to considerable error. Trade data, especially government imports, need better coverage and quality control. For BoP attempts should be made to improve non-trade data, especially for services and external aid. The only available price data are on retail prices of 24 food and non-food commodities, and major import and export unit values. Wholesale and farm-gate prices are needed. Consumer price indices for Kabul and the country as a whole are not of much use because of their unrealistic weighting pattern (a 90% weight for food) which was not based on any household expenditure survey. In other statistics, education, health, and communication data are detailed and have fairly complete coverage. However, only fragmentary data are available in important economic sectors such as construction, transport and commerce. Private construction, road transport and wholesale and retail trade are completely neglected. Banking data suffers from-the absence of any estimate of the monetary activities of the "bazaar". 10. National accounts estimates can only be as good as the basic statis- tics. Recent attempts at GNP estimates are based on calculations of GDP by industrial origin. These Ministry of Planning estimates have been used in the 7-year Development Plan. However, the absence of reliable estimates of population, employment, gross output of most sectors, cost-of-production surveys for estimating intermediate consumption, household surveys, and price data, all point to the difficulty of deriving any reasonable estimates of national accounts.  - 4 - II. Thus the task of the statistical development in Afghanistan is immense. Much detailed analysis remains to be done. In the meantime the CSO should actively promote the Professional Statistical Commission for guidance on user-orientation of statistics, formulation of statistical programs, and increased coordination with and participation from the Govern- ment agencies. This Commission, composed of- high ranking Government officials from various agencies, is required by the Statistical Law to advise the CSO on policieg, programs and professional matters of statistics. The Commission has met a few times since the inception of the Law in June 1975, but participation has not been adequate. The collaboration between the CSO and the Commission should improve the pro-ramming of statistics, particularly in the long-term development of Afghan Statistics. The CSO should also maintain a close rela- tionship with the lIinistry of Planning, the most important user of statistics. 12. The mission recommends immediate steps to strengthen the fol- lowing presently viable statistical collection processes for better quality control and coverage: (a) general government accounts including both central and the available municipal accounts, (b) analysis of the budgets of public enterprises and compila- tion of their accounts, (c) wholesale prices in addition to retail prices, (d) better estimates of the outputs of agricultural crops and industries, and (e) better external trade and balance of payments data. Marginal efforts in these areas should improve the reliability and usefulness of these statistics considerably. 13. The major problem of comprehensive data then should be tackled by a National Economic Survey. This Survey should be an integrated program of properly designed small sample surveys and collection of administrative data for the purpose of compiling comprehensive economic statistics for the global benchmark estimates of the Afglhan economy including especially the national accounts. The Survey should be a shiort-term project, carried out u;ithin a two-year period, by a small team of 3 experts, 8 professionals and 30 enumerators. The Survey will be highly effective if it is conducted during 1978-79. By the beginning of 1978 the population census frame withl sketchmiaps of villa-es and preliminary counts of population will be available. SamIplc surveys could be implenented on the basis of this frame. The results of the satiple surveys (with the help of census data in 1978), would then provide the global estimates. Niot only would this Economic Survey provide benchmark estiLmates, but in the process it wouild also identify the gaps in inforLiation, suitable procedures and areas of data collection, and the costs and benefits of compilation of selected data vis-a-vis the national economic system. Thus, in addition to the benctmiark estimates, the CSO would have a firmer basis  - 5 - for formulating an integrated program of statistical development for the future than envisaged in its 5-year development program of 1974. Since both the short-term requirements of the economic analysts and the need to prepare a viable program for statistical development in the longer term can be met adequately by undertaking the National Economic Survey, the mission recom- mends that it should be given the highest priority. 14. The mission recommends that emphasis must be given to recruit qualified statisticians and to improve training facilities both in-house and through fellowships abroad to upgrade the capability of the CSO staff. III. REVIEW OF AFGHAN STATISTICS 15. Statistics - the compilation of data - are required to play a wide- ranging role. They are not end-products themselves, but used in decision- making and analysis by Government, economists, businesses and others. Not only, therefore, must statistics be user-oriented, but they also have to be consistent, comprehensive, and comparable over time. Most importantly the timeliness of statistics is essential for their use in decision-making. 16. In the process of formulation of the 1976/77-1982/83 (1355-1361) Seven Year Development Plan, the Ministry of Planning in Afghanistan has become acutely aware of the shortcomings of Afghanistan's statistical base. In recent years the quality and quantity of basic statistics has only im- proved marginally. Except for a few minor surveys the mainstay of Afghan statistics is still data generated by administrative processes. A. Statistical Publications 17. Since the CSO is the sole authority for publication of official statistics, no Government agency can release any data without the prior approval of the CSO. The only four regular statistical publications of the CSO are the Statistical Information of Afghanistan (latest 1975/76 (1354)), the Exports of Merchandise from Afghanistan (latest 1974/75 (1353)), the annually during the last 3 years, and the Quarterly Review of Afghan Statistics (latest, 4th quarter 1355 (1977)). The Quarterly Review is a collection of articles on general statistics and current developments. Also available at the CSO are Afghan Foreign Trade 1956/57-1963/64 (1335-1342) and A Provisional Gazetteer of Afghanistan, Vol. 1-3 (published in February 1975). The Gazetteer is a compilation of population data by province and village obtained from the Agricultural Census of 1967/68 (1346) and Ministry of Interior records. The CSO prepares internal documents of the weekly price index of four major food items, monthly precipitation and price data of 24 comodities, and quarterly statistical review of foreign trade and industry data, which are circulated to the interested Government agencies. 18. The latest Yearbook - Statistical Information of Afghanistan - available in July 1976, is a voluminous publication (about 500 pages) con- taining some descriptive material, tables and charts under chapter headings:  -6- Population, census and manpower, Agricultural statistics, National Accounts, Economic statistics, Social services statistics, International trade, and a short summary in English. The national accounts chapter mainly covers public finance, money supply, money and banking, and prices, with a single table on national income. Many data series are presented up to 1975/76 (1354). Thus the time lag of publication of only 4 months is highly commendable. However, the CSO concludes the English summary with a statement that "it should be reiterated that the contents of this vclume were collected from many sources. Because of a number of factors, including errors in reporting, compilation and processing, and incomplete coverage, most data are subject to error" (page 9). The publications on exports and imports provide computerized foreign trade data for 1974/75 (1353) by commodity by country according to SITC 7-digit codes with about 14 months lag in publication. Since a large amount of current information is available in internal documents for limited circulation, the CSO should consider publishing regularly in its Quarterly Review a compendium of current statistics for all users. 19. The General Directorate of Statistics, Ministry of Education pub- lishes the annual Educational Statistics (latest available issue 1973) with detailed information on educational structure and age composition, schools, students, teachers, vocational schools, teacher training institutions and Islamic schools, higher educational institutions, and budget with considerable detail of provincial data. Publication procedure is slow, the 1974 issue is being readied now, although preliminary data for 1976 are available from files. The publication is in both English and Dari. 20. The General Directorate of Health Information and Evaluation, Ministry of Public Health publishes a quarterly Report of Health Institute Activities (latest issue for 2nd quarter, 1355) and an annual Report on In- patients in Afghanistan Hospitals (latest issue for 1354). The quarterly publication provides data on polyclinic and Mother and Child Health centers (MCH) attendance by province and age, number of immunizations, lab. exams etc., and the annual one reports on hospitals, beds, discharges, average length of stay and bed occupancy rates by type and size of hospitals. The publications are in both English and Dari. 21. The Da Afghanistan Bank Magazine, issued every quarter (latest issue Jan-March 1977), contains a large section on statistics. The tables and charts include data on money, debt and investment, foreign exchange, foreign trade, and prices of exports in foreign markets in detail up to December 1977 (1355). Some internal documents on monetary data are avail- able for circulation to the interested ministries only. B. Basic Statistics 22. The following is a review of some selected sectors of the statis- tical system of Afghanistan. The review is necessarily incomplete for reasons discussed earlier.  -7- Demography and Manpower 23. There has never been a population census in Afghanistan. The various sources of data for population estimates are: i) Agricultural Survey 1967/68 (1346) - This survey attempted a complete enumeration of villages, and was oriented towards land utiliza- tion and agricultural production. As part of the enquiry, information was obtained from the village-headman (Malik) about the number of people living in the village. These were obviously approximate figures. ii) Civil registration data and distribution of citizenship cards - The Ministry of Interior is required to record birth, death, marriage, divorce, etc., and issue citizenship cards for school entrance, employment and military service purposes. The registration is mainly for the male population. A registration of all males in the population was undertaken in the early 1950's through visiting all villages. This was supplemented by issuance of registra- tions to men who were not registered during the initial phase. Since then the Ministry has attempted to update the records from the reports of male births which occurred since the initial complete registration. During 1973/74-1976/77 (1352-55) another complete registration of settled population was carried out, and registration cards (Tazkiras) were issued to all males and some females seeking school admission and employment. A demographic questionnaire was also canvassed at the same time recording age, sex, family composition, occupation, marital status and educational level (data not yet completely processed). However, even for males the registration data are incomplete since parts of the 4 border provinces are not required to register. The registration of nomads will be undertaken during 1977/78-1978/79 (1356-1357). iii) Number of tax payers - The Ministry of Finance provides such in- formation, which of course only covers a small part of the population. iv) Afghan Demographic Survey - This sample survey was conducted by the CSO and USAID (Contractors - State University of New York at Buffalo and Johns Hopkins University) during 1972-73 for sedentary population and in 1974 for nomads. The objectives of the survey were to provide improved demographic statistics, institutionalize within Afghan agencies the capa- bility of collecting and analyzing high quality demographic data, provide sociological and statistical data for the planning of the family guidance program, and institutionalize the capability of collecting data for the design, implementation and evaluation of family guidance programs. The sedentary part of the survey was conducted with household interviews and pregnancy history interview schedules, and data were collected from approx- imately 21,000 households and 9,000 women covering 120,000 individuals of "sedentary population". Information was collected on demographic parameters, literacy, socio-economic characteristics of the families, use of contracep- tion, radios, and travel. There were post-enumeration quality control checks with re-interviews. No details are available on the survey of nomads. The results of the survey, particularly the national population estimate, however, are in doubt since the sample may have been under-enumerated by about 5-10%,  - 8 - mainly due to the underreporting of females. On the other hand, it is con- sidered that vital statistics, and family characteristics obtained through the survey are fairly reliable. The organization of the survey, sampling method, interview procedure, and quality control of the data were of a very high professional standard. v) A population census of Greater Kabul was conducted in 1965 as a pilot project in preparation for conducting a national census of Afghanistan. No post-enumeration quality check was done. The results of the census were expected to involve a 5-10% under-enumeration. vi) Other estimates - The Ministry of Planning and the UN Specialized Agencies provide some estimates of their own derived from observations and fragmentary data. 24. The CSO, in conjunction with a Ministerial Committee, undertook a review of the above population estimates in 1975/76 (1354) and arrived at the following figures which were used in the 7-year Development Plan: Population in 1975/76 (1354) Number (000) % distribution Total 16,665 100.00 Nomad 2,405 14.43 Settled 14,260 85.57 Urban 2,390 14.34 Rural 11,870 71.23 Male 8,666 52.00 Female 7,999 48.00 These estimates are about 28% higher than those projected from the Afghan Demographic Survey results, although the total has been substantially lowered from the 1974/75 (1353) published figure of 18.70 mill. Some experts now believe that a reasonable total would be in the range of 12-15 million. 25. The vital statistics published by the CSO are also somewhat dif- ferent from those of the Afghan Demographic Survey (ADS):  -9- cso ADS 1975-76 1972-73 (Sedentary Population Only) Population Growth rate (p.c.) 2.3 2.2 Sex-ratio 108.3 115.6 Crude Birth rate (per thous.) 43.0 (under-estimate) Crude Death rate (per thous.) 20.8 ( " " ) Gross Reproduction rate 3.4 Infant mortality rate 184.9 (under-estimate) Age distribution: 0-14 years 43.3 45.6 15-64 " 54.1 51.2 65- " 2.6 3.2 Urban population (p.c. of total) 14.3* 15.1 Life expectancy (years at birth) 34.6 Literacy (p.c.) 11 (6 + years) * Urban as p.c. of settled population only, is 16.8%. The rates of the CSO are not unreasonable, especially since ADS rates are for sedentary population only. The difference in size of the population estimates, however, is quite substantial. 26. This problem can only be solved by a population census, and the CSO is now preparing to conduct a census in mid-1978. The census project, started in 1975 in cooperation with the UN Fund for Population Activities, consists of preparatory, enumeration and post-enumeration phases. In the preparatory phase the most important activity will be cartographic work, preparing "sketch-maps" for delineating boundaries of every existing village, and listing households and family sizes. This work has already started and will end by March, 1978, and will provide a preliminary estimate of sedentary population in villages. Other activities in this phase are designs of census questionnaire and tabulation programs, manuals, and conducting a pilot census. The general census questionnaire will probably have 18 basic questions for each individual in the household. In 5% of the households the details on social and economic characteristics, and information on fertility and mortality will be collected. The Kabul urban area will have a more detailed questionnaire than the other areas. The pilot census for testing the questionnaire and the enumeration procedure is planned for September, 1977. This will cover 2,500 urban households, 3,000 rural house- holds and 2,000 nomad tents (about 50,000 persons in all) in not more than 5 provinces such that urban (Kabul), mixed urban/rural, rural, nomad and mixed rural/ nomad areas are represented. In the enumeration phase 400 supervisors, 2,000 controllers and 10,000 enumerators (mostly teachers) will be recruited and trained. The actual enumeration will require about 3 weeks, and will be held between 20 Sept.- 10 Oct., 1978. The time of the year is important, since  - 10 - only at such times the nomads temporarily settle down. The post-enumeration phase will cover data processing and quality-control checks. It is expected that a preliminary population count will be available after two-months of the enumeration period (in Jan. 1979) and the final results will be published in early 1981. 27. The census project seems to be progressing well, although the sketch- mapping operation is behind schedule and therefore needed some reinforcement. About 200 staff of the CSO are detailed to work on the project now, and in 1978 the whole staff will be involved (unfortunately to the detriment of the regular statistical work of the CSO). The problems foreseen for the project are communication (publicity) for the census, accessibility to all parts of the country, and enumerator training and supervision. A training program for communication was held in September, 1976 with the help of the Asian Statis- tical Training Institute. The Census Project is being overseen by a National Census Committee with the Minister of Interior as Chairman and the President- General of the CSO as Secretary. 28. It may also be noted that the CSO's latest Yearbook published results of a survey of municipalities. This survey covered 69 municipalities in 25 provinces -(out of 87 municipalities in 26 provinces) and provided data for 1974/75 (1353) on population, housing, business establishments, schools, municipal employees and wage-earners, revenue, and ordinary and development budgets of the municipalities. The enumeration was only by mail questionnaire. 29. The vital statistics registration system of the Ministry of Interior is a new project undertaken in cooperation with the UN. The system is designed to record births, deaths and marriages throughout the country. A pilot project for 6 months covering 3 villages and 2 urban sub-areas has already been started. Under the legal provisions the village headman (malik) or the oldest male member of a family is required to report the events. The Ministry has the network of statistical clerks in all minor civil districts (MCDs) to record the statistics. Consideration is being given to appoint the maliks as registrars with periodical surveys by enumerators from the Ministry. It appears that with the tradition of civil registration in Afghanistan and the legal provisions in force, the vital statistics systems may prove quite successful, and thus become a valuable supplement to the population census. 30. In manpower, the only partially reliable data appear to be employ- ment in organized industry and estimates of the agricultural labor force. The employment data are obtained from the monthly survey of industries (incomplete coverage). The first survey of handicraft industries, which was conducted by the CSO in 1975/76 (1354) in eleven provinces, generated some employment data, and are available in the CSO's latest Yearbook. Data on the agricultural labor force are based on the 1967/68 (1346) Agricultural Survey and updated annually from the Ministry of Agriculture's Evaluation Department's reports on the state of agriculture. There is no Ministry of Labor, but a Department of Labor exists in the Ministry of Mines and Industry. For the 7-year Development Plan manpower requirement data have been obtained by the Ministry of Planning from project proposals and the various Ministries.  - 11 - 31. As will be seen later, the coverage of industrial surveys and the agricultural reports are not complete. Since the global estimates obtained from the surveys are not "true" totals due to non-probability sampling procedures, published data are either partial or unreliable. Employment data in the Yearbook presented with the population figures are, therefore, only rough estimates. 32. Until recently, the Manpower Division of the CSO was being used as a center for employment which introduced job applicants (high school and university graduates) to various Government agencies. All government employment, specially of university graduates, was channelled through the CSO office. This was a heavy burden on the CSO and diverted it from its primary responsibility. This function has now been taken over by the newly created civil service department. Agriculture 33. The basic source of detailed agricultural statistics is still the Agricultural Survey of 1967/68 (1346). The survey was conducted by the Ministry of Agriculture and covered 15,620 identified villages (current estimate of villages is 22,750). Enumerators obtained their information .mainly from the maliks and other local leaders. Data were collected by province on the number of villages, the agricultural population, the number of landowners and livestock owners, the area of arable land and orchards by irrigated and non-irrigated, and productions of crops, livestock and poultry. These data have been used for deriving annual estimates of crop and livestock production in conjunction with the reports from agricultural extension workers on the area sown and harvested and yield. 34. Another agricultural sample survey was conducted in 1969/70 (1348) by the Statistical Department of the Ministry of Planning covering 7 provinces (Baghlan, Kunduz, Parwan, Laghman, Nangarhar, Ghazni and Kandahar). Twelve predominantly agricultural villages were selected in these provinces for com- plete enumeration. Data were collected on population, household composition, system of land ownership, area under cultivation and yield of main crops, and per capita income and other economic characteristics of the households. A total of 737 households (6,926 persons) was included in the survey. The survey results have been published in the Survey of Progress of 1970/71 and 1971/72 (The Ministry of Planning's annual report on the Afghan economy). The survey presents some interesting results about the agricultural families, such as the numbers of sharecropper and landless laborer families, economi- cally active population varying from 48 to 56% in the 7 provinces, and the average annual per capita income ranging from 2,041 to 6,670 Afs. However, these results are for the surveyed population only, and no attempts have been made to ascertain how representative they are or produce any global estimate. 35. In 1974 the CSO, with the help of the Ministry of Agriculture, conducted 2 agricultural surveys on sugarcane and apiculture. The survey on area and production of sugarcane has covered all landowners in 160 villages  - 12 - of the Nangarhar, Laghman and Pakhtia provinces which are the major sugar- cane producing areas. Data were collected on area and production by type of sugarcane, yield, use of sugarcane, revenue from sales, and income of farmers. The survey on apiculture (number of honeybee families and honey production) was conducted in 12 provinces which supposedly covered the api- culture activities of the whole country. Information was collected for 1973 and 1974 on the number of apiculturists, number of hives, yield, sales of honey and beeswax, and income. For these two items, therefore, the surveys cover essentially the production of the whole country and provide up-to-date reliable information. 36. In addition, a forest survey was undertaken by the Ministry of Agriculture in 1969 throughout Afghanistan using aerial photography and maps. The survey identified forest and pasture areas. 37. The current agricultural statistics are obtained by the CSO from the Ministry of Agriculture, whose Evaluation Department compiles reports on agricultural activities from the extension agents. The agents send a quarterly progress report on areas under crop, current and prospective yield, number of farmers, use of fertilizer, seed and machinery, etc., and an annual report at the end of the harvesting season (December) on area and production of crops in irrigated and non-irrigated land. While the quarterly report could be used for forecasts of crops, the annual report has been the basis of updating agricultural crop production. Until 1974/75 (1353) the agents provided information for the total area. However, from 1975/76 (1354) a stratified random sampling procedure has been adopted to improve the quality of the data by reducing the workload of the agents. Under this procedure the stratification of land is done in the Kabul head office, and following given instructions the agents pick out 10-15 land holdings out of 100-400 holdings under each of their jurisdictions for detailed reporting of area and produc- tion. The reporting form enumerates the pieces of land held by a landowner, and for each piece the area, production and yield of various crops; foodgrains, legumes, fruits, alfalfa, melons, vegetables and industrial crops. In all 37 products are itemized. After receiving the reports in late December, the Evaluation Department checks the information, and in February sends the data to the CSO for processing. The Evaluation Department is fairly satisfied with the progress of the sampling procedure and the quality of information, except for some problems in checking of data. The Department occasionally sends its staff on field inspection of coverage and quality of data. These data may be processed by computer from next year. 38. As far as other items are concerned, the updating of livestock and poultry data are done by the CSO in consultation with the Livestock Department of the Ministry of Agriculture. The monthly precipitation data obtained from 13 meteorological stations around the country are also published by the CSO. These data are not very useful since rainfall and snow are not separated. No temperature data are published. However, the detailed meteorological data on snow, rain, temperature, humidity, sunshine, windspeed, etc. are collected and published monthly and annually by the Directorate General of Meteorology of the Afghan Air Authority. The data are collected from about 100 precipitation and climate registration stations which  - 13 - include 8 agricultural stations for special observations. The Government is instituting a land registration system in preparation for the land reform. The registers will provide valuable information on land-ownership, size of holdings, quality of land according to production of crops, and availability of water, etc. 39. In January 1977, FAO urged the Government of Afghanistan to develop an early warning system on basic food supplies with particular reference to crop forecasting. The report emphasized the need of pre-harvest forecasting of agricultural area and yield from weather data and inputs, food supply situation from prices, demand, imports and procurement, and asked the CSO to coordinate the information system. The report is under consideration of the concerned ministries. 40. From the above it appears that the collection of current crop data through the network of the Ministry of Agriculture's Evaluation Department is well-organized. However, the reliability of data and effectiveness of the sampling method in obtaining global estimates, especially the procedure of updating from the 1967/68 (1346) Agricultural Survey, are in doubt. The extension agents responsible for collecting the data lack statistical train- ing. In sampling, the minor crops usually need special treatment since they are not as widely or uniformly distributed over the agricultural area as the major ones. There is also no possibility of crosschecking production of commodities with export, import, and especially consumption data. Hence in agricultural statistics only the major crops such as wheat, cotton, beets, sugarcane, etc., seem to have good data. The minor crops, livestock, and forestry products are rather crude estimates. Emphasis should be put on proper probability sample surveys, crop-cutting experiments for yields, and training of enumerators for improved agricultural statistics. Industry 41. The Economic Statistics Department of the CSO conducts a monthly survey by mail enquiries of all mines, industry (excluding handicrafts) and electricity and power generating establishments. The list of establishments has been compiled with the help of the Ministries of Mines and Industry, Commerce, and Planning, and covers all public, private and mixed enterprises. The questionnaire is rather elaborate and includes (i) location and type of enterprise, (ii) employees by different grades and their wages (in cash and in kind), (iii) main products, production capacity, last year's production (check item), electricity generated, (iv) quantity and value of sales of products and services, average price per unit, stock, other income, (v) pro- duction costs - raw material, machinery and parts, fuel, wages and deprecia- tion, also average work hours per day, shifts, monthly total hours and days, and (vi) explanations for unused capacity. A quarterly report is prepared from these questionnaires with about 3 months lag. The initial response of the survey is about 70%. Non-response is followed up by telephone reminders and visits to the establishments with some success. Data published in the Statistical Information of Afghanistan show indices of industrial production (base 1971=100) by important products, groups and total; and revenue from sales. The aggregate indices are, however, not very useful since they do not  - 14 - seem to be based on any definite quantity or price weighting structure. No attempt has been made as yet to analyze the production cost structure of the industries, although data are available in the questionnaires. The list of registered enterprises is complete (due to the requirement of annual registra- tion) and the rates of final non-response is about 5%, but the mission could not establish whether they bias the data (especially for comparison over time), and how the quality of information is checked. 42. In handicrafts, the first survey was conducted by the CSO in 1975/76 (1354) in 11 provinces (Kabul, Kandahar, Herat, Balkh, Faryab, Kunduz, Ghazni, Farah, Helmand, Nangarhar, Parwan). The report of the survey is under prep- aration but some results are available in the statistical Yearbook. The survey indicated that 172,000 people are engaged in handicrafts in these provinces. Data have been collected on type of products such as carpet weaving, embroidery, etc., volume of production, sales, machinery used, investment, and raw material used. The survey was by complete enumeration, but it is not known how representative the results are for the whole country. No analysis of the structure of cost of production by industry has been attempted. 43. Thus in the industrial sector only partial data are available. For modern industries, emphasis must be put on obtaining a complete list of enter- prises and continuously updating it. The enterprises should be classified by industrial activity and by size, so that replies to questionnaires can be monitored to ensure that all industries and the large enterprises are ade- quately covered by the survey. For reliable global estimates it is essential that the large enterprises respond to the survey and follow-up for non- responses should be concentrated on them. For handicraft industries it is suggested that a quick survey of other provinces, for output and manpower only, be conducted to obtain global estimates. Production costs, etc., available from the 1975/76 (1354) Handicraft Survey could be used for the present to calculate value-added in national accounts. In future all surveys should be based on properly designed probability samples so that useful global estimates can be obtained at a considerable savings of resources. Government Accounts 44. The Government revenue and expenditure accounts are provided- to the CSO for publication by the Ministry of Finance. The detailed accounts of Government revenue by source, ordinary and development expenditure by goods and services and by ministries, development expenditure by economic sectors and projects, and foreign aid for development projects are published. However, these accounts are for the Central Government only. The municipal budgets are sent to the Ministry of Finance for information, but no consol- idated General Government Account is available. 45. The Accounting Department monitors the execution of budget and the disbursements of the central and provincial (branches of ministries) Govern- ment. The ministries submit to the Accounting Department monthly reports on revenue and expenditue with a time lag of about I month. Consolidated quarterly reports are available within about 2 months after the end of the  - 15 - quarter. The annual accounts of the Government are presented with the next year's budget. The Accounting Department is hampered by the shortage of qualified accountants in the country. While the accounting manual is now 15 years old, it could not be fully implemented due to lack of qualified personnel. Recently the Ministry of Education has started a special training course for accountants, and in 2 years' time about 200 should be available. The Accounting Department uses NCR machines for processing the accounts. 46. The Central Government accounts are the most reliable data in the statistics of Afghanistan. However, the municipality accounts need to be compiled to prepare the consolidated General Government accounts. The CSO in its latest Yearbook published 1974/75(1353) data on a survey of 69 muni- cipalities (out of 87) which included summary information on ordinary and development budgets and revenue. 47. The mission could not obtain aggregate information on the budgets of public enterprises although they are supposed to be sent to the Ministry of Finance. There are about 40 public enterprises engaged in agricultural, industrial, and commercial activities. While the public enterprises con- tribute to the Government account through payment of corporate income taxes and transfer of any surplus, for national accounting purposes the public enterprises are classified as part of the relevant economic sector. The detailed accounts of the public enterprises are, therefore, essential for estimating GDP by industry. External Trade and Balance-of-Payments 48. Since 1974/75 (1353) external trade data have been compiled by the CSO. Previously the Ministry of Commerce had this responsibility, but the staff was transferred to the CSO. The CSO has statistical agents stationed in customs houses around the country. From the customs invoices, the agents compile export and import data on statistical forms with description of the commodities, their quantity and value (in declared currency), origin and destination, and whether the transaction was made by the private, Government or mixed sector. These forms are sent to the CSO on a monthly basis. The Foreign Trade Section of the CSO checks the data for reasonableness (e.g., plausible unit price) and codes the data by SITC. The data are processed at the end of the month. All foreign currencies are also converted to Afs. by the relevant monthly average exchange rates provided by the Price Section of the CSO which obtains the figures from Da Afghanistan Bank. The data sheets are then sent to the Data Processing Section for computer processing and are published annually. The annual publication is delayed by about 14 months, and the data only show commodity by country compilation. The Data Processing Section is planning to produce a monthly report in the current year. Mean- while the Foreign Trade Section produces a partial summary report every month (with about 15 days lag) showing all exports and about 65 major imported commodities by commodity (not by SITC) and by country. These preliminary reports, which are also compiled quarterly, are sent to the interested Ministries and Da Afghanistan Bank. The annual figures and the totals from partial monthly summaries show some discrepancies. The CSO occasionally sends staff on field inspection for checking the quality of data.  - 16 - 49. These detailed customs data by commodity are mostly for commercial trade only, since the commodity description of imports financed by foreign aid for projects and other purposes are not always shown on customs invoices. For analytical purposes the aid-financed imports are sometimes included in commercial trade when the commodity description is available. Otherwise they are lumped together and shown in a separate category. Since the valuation of these imports in Afs. are done at a fixed rate, the commercial imports as shown in Afs. will have a mixed exchange rate. The aid-financed imports may also be under-recorded. The CSO has, therefore, initiated a system from March, 1976 whereby supplementary information on loan and grant-financed imports will be collected from the various Ministries and Government agencies, but with little success. It is important that every effort is made to obtain commodity detail for all imports which is essential for any meaningful economic analysis. 50. The Balance-of-Payments accounts is essentially compiled by Da Afghanistan Bank and reviewed and published by the CSO (although not in the standard IMF format). In the BoP the recorded merchandise exports are adjusted for the discrepancy between customs and exchange surrender data. This poses a problem since one of the major export commodities (Karakul) is held in inventory abroad. The value of commodity exports also need adjust- ments in many cases, since some commodities are exported on consignments and the sale prices are later obtained from the banks and government commercial attaches abroad. The services data are generally considered to be under- estimates. Data on loans and grants from external sources are also not very satisfactory since reports from the Ministry of Planning and international and foreign Government sources show considerable discrepancies. 51. Thus the BoP data are subject to considerable error. In trade, there are difficulties in classifying imports and lack of external trade data by country and commodity. Also, exports are not classified into domestic and re-exports. Transit trade, which has increased in recent years, is not recorded. It is also generally conceded that there is a large amount of unrecorded trade (specially exports), and there may be some under- and over- invoicing of imports and exports. The more serious problem is that although there is the mechanism to check whether all customs invoices have been statis- tically recorded or not by stamping the invoices as a requirement before the goods are released, the CSO has no source material to recheck the information for consistency and validity. Hence, the CSO should obtain copies of invoices, and use those copies for statistical coding and processing. Furthermore, in the published data there are noticeable errors in aggregation of data and in groupings by SITC. In BoP, non-trade data are generally unreliable. Monetary transactions, particularly workers' remittances from Iran in recent years are also not accounted for. Da Afghanistan Bank is trying to improve the situa- tion, and a comparable time series of external trade data for 1967/68-1974/75 (1346-1353) is being finalized. An attempt is also being made to prepare export and import price indices.  - 17 - Prices 52. The CSO collects the average monthly prices of 24 consumer goods from 11 provincial centers and Kabul area (4 markets) during the first 10 days of each month. Price indices based on them are then compiled for Kabul and the country as a whole. The commodities cover 17 food items: grains (3), meat (2), fruits (3), vegetables (5) and others (4); and-7 non-food items. The Yearbook published by the CSO in July 1976 contains data up to March 1976. The weekly price index for four major food items in 12 important centers with one week's lag and the monthly price indices for all 24 items with one month's lag are also prepared for internal circulation. 53. Detailed price data are also collected by the branch offices of Da Afghanistan Bank and reported to the head office on each Monday by telephone. All the branch offices (63) are required to send these information regularly for all commodities. It is not clear, however, whether the comparability of specification and quality of items are maintained over time. These weekly retail price data are sent to the CSO by Da Afghanistan Bank for processing. 54. The national index of retail prices was first prepared in 1964/65 (1343). For the construction of the index, the distribution of population is used as weight to derive the average prices from price quotations around the country. These average prices are then weighted by the relative importance of the consumer goods (based on some very rough guesses of national income) to calculate the index. The Kabul index was instituted later. However, both the national and Kabul price indices reflect a 58.1% weight for cereals, 10.7% for meat, 10.6% for fruits, 5.7% vegetables, 5.1% for other food items, and 9.8% for non-food items. The base periods of the indices are 1961/62(1340) and 1968/69 (1347) respectively. 55. It is obvious that both the prices and the price indices cannot be relied upon. The prices are collected by untrained people without sufficient checks on reliability of data. Price quotations are given in terms of units of 1 seer (16 lbs.) or say 100 eggs which are obviously not the usual units of retail sales to consumers. The weighting of price indices, with a 90.2% weight for food, is most unlikely for any consumption pattern. Two household consumer surveys have been undertaken in Afghanistan - the Kabul Household Expenditure Survey in 1968/69 (1347) (a 2.4% sample enquiry covering 1,685 households in Greater Kabul) and the Lashkargah Urban and Rural Household Expenditure Survey also in 1968/69 (1347) (829 urban households and 1,114 rural housholds in 7 villages from surrounding areas). The distribution of consumer expenditures, available only for the Lashkargah survey, is as follows:  - 18 - Lashkargah Consumer Expenditures (per capita) Urban Rural Total Food & drink 60.7 65.6 63.0 Rent, etc. 5.3 1.2 3.4 Fuel & light 9.6 7.3 8.6 Clothing 8.6 7.6 8.1 Miscellaneous 15.8 18.3 16.9 Total 100.0 100.0 100.0 (Source: Survey of Progress 1968/69, Ministry of Planning) The Kabul Survey indicated only 50% of expenditure on food items. In view of these, the present consumer price indices need urgent revision. 56. Regarding other price data, the import and export unit values of major commodities have been available for some years. However, there has been no attempt in the past to compile and analyze them systematically. There are no wholesale prices or farm-gate prices of agricultural commo- dities available. However, wholesale and retail prices of imported consumer goods can be obtained from the Internal Trade Division of the Ministry of Commerce which collects this information in order to control prices in the local market. Other Statistics 57. For construction, the total expenditure in Central Government construction is the only available information. This is obtained from the Public Works Department. 58. For the transport sector, reliable statistics are available only for air transport from Ariana and Bakhtar airlines. Only partial data are available for road transport, however, which is the major transport activity in Afghanistan. Statistics are available for registered motor vehicles in Kabul. The Government collects transit fees from passenger vehicles and trucks at various points of the road network. Detailed data on transit fee collections and passage of vehicles at these locations are available from the Ministry of Finance. Also, the Ministry of Commerce issues permits to truckowners for transportation of goods from specified origins to destinations when the distance is over 24 kms.. The consignors of goods then contact the Ministry to have trucks assigned to them. It appears that the Ministry has some information on volume of transportation by kinds of goods for shipments of Government departments but not of private sector. However, no attempt seems to have been made to utilize this for estimation of the volume of road transport activity. For communications, detailed data are available for posts, telegraph and telephones, and radio, from Government departments. 59. For banking and insurance, Da Afghanistan Bank compiles data on banking every month in considerable detail. However, no information is available on the monetary activities of the "bazaar", which is quite active  - 19 - in exchange of currencies and financing of trade. Data on insurance com- panies are obtained directly. The CSO completed a survey of banking sector in 1975/76 (1354) but the results are not yet available. 60. No information is available on domestic wholesale and retail trade (except for Government monopoly operations), or on professional and personal services. 61. Tourism data are published by the number of tourists by national origin arriving by air and surface transportation and also their expenditure in Afghanistan. The expenditure estimates are very approximate. Probably better estimates could have been obtained from the Tourist Currency Declara- tion Forms (Ministry of Finance, Customs Department) which record import and export of foreign currencies, but these have recently been eliminated. Data are also required on hotel occupancy. 62. For social services, data on education are published in great detail. The Ministry of Education, with the help of UNESCO, has an elaborate system for collecting information on schools, students and teachers by province, type of educational establishment, enrollment and graduation by sex, data on vocational schools and higher education institutions, and budget. Presently education data are being computerized which will facil- itate their timely publication. Statistics on Afghans sent aboard for training on scholarships are also available. Data on health are available in detail from the Ministry of Health. Information is obtained from all Government civilian hospitals and basic health centers, and also from some hospitals attached to the army, police, University and public enterprises, every month by mail questionnaire. Number of hospitals, physicians, beds and other medical facilities by provinces, records of contagious diseases, vaccinations, etc., are published. There are plans to computerize the data. In information and culture, statistics are published on number and location of libraries, movie houses and their attendance, museums, publication of newspapers and periodicals, etc. Annual reports on crimes by type and province are published. The activities of the Afghan Crescent Society and the Afghan Family Guidance Council are also reported in the Yearbook of the CSO. 63. Thus important economic sectors such as construction, transport and commerce have only fragmentary data. Emphasis must be put on road transport statistics, and especially commerce which is generally a major activity in developing economies. Data on professional and personal ser- vices are usually obtained through household expenditure surveys. Since such surveys are also necessary for consumer price indices, the CSO should give priority consideration to conducting them. C. National Accounts 64. The earliest attempt at estimation of the national accounts in Afghanistan was for 1958/59 (1337), made by the research and statistics department of the Ministry of Planning (Gross National Product for Afghanistan 1337 - Ministry of Planning, pp. 19). The publication contained a discussion  - 20 - on national income concepts and six tables - Gross National Product, Personal income and expenditure, Corporate income account, Capital account of Afghanistan, Current and Capital accounts of the Central Government, Foreign transactions - with some notes on data. Although a recommendation was made for compilation of annual GNP estimates, the effort apparently was not continued. The prepara- tion of national accounts on a regular basis is fairly recent. Again, the estimates for 1971/72 (1350) were compiled by the then statistics department of the Ministry of Planning, but on a very modest scale. The CSO has con- tinued to publish these estimates in the Yearbook in the form of Table 1 below. The CSO calls total production the Gross National Product, which is a misnomer according to the accepted concepts and definitions of the national accounts. GNP is normally defined as the sum of GDP (value-added, i.e., total output less intermediate consumption, of all production and services sectors) and net factor income from abroad. In Table 1, only the net produc- tion (national revenue) figures are relevant for inclusion in GNP. 65. In view of this, the Ministry of Planning prepared in 1975/76 (1354) a different estimate of GNP at market prices following the accepted definitions more closely, and used this for the 7-year Development Plan. These estimates are shown in Table 2 below. These start with net material production figures of all production sectors. Further items in Table 2 include contributions of trade and transport, services, housing, depreciation, factor income, indirect taxes, and subsidies. The sequence of items shows an attempt to build GNP from net material products. Of course the first two sub-aggregates, i.e., National Income and Gross Domestic Product at Factor Cost, should be termed Net Domestic Product at Factor Cost and Gross National Product at Factor Cost according to standard definitions. 66. While the basic accounting procedure of Table 2 corresponds to usual national accounts concepts, albeit with the wrong labels on some items, the problem of estimation must have been tremendous. National accounts can only be as good as the basic statistics. As we have seen earlier, total output data of agriculture and industry including handi- crafts are unreliable and likely to be underestimated. Only fragmentary data are available for construction, transport and commerce. Thus the gross production figures are based on partial data and liable to be underestimated. Furthermore data on material expenses, i.e., intermediate consumption, are mostly non-existent as there have been no cost-of-production surveys of any significance. In a number of cases the Ministry of Planning assumed the ratios of intermediate consumption to total product in Afghanistan to be those observed in neighboring countries with similar production character- istics. The estimation of the factor cost was also based on unreliable information since published data are for retail prices only. The Ministry calculated Net Domestic Product by sector which required deducting deprecia- tion in each sector. The total depreciation, on the average, amounted to only 2% of net product (see Table 2). The whole computational procedure seems to consist of rather rough estimates. Estimates of private services and housing are again based on the experience of similar countries. The factor income item, which represents factor income from abroad, raises serious questions. The BoP data are deficient in this respect and it is hard to conceive this item being negative in 1354 for the Afghan economy given the size of worker's remittances received by Afghanistan in the recent years.  :ab1e - TOTAL IIAT7NAL PROU 8TS, KATR IAL EXP1DITURE, AND NATIONAL INCOSE (Tn bill. Af4. at 1975/76 (1354) constent prices) 1968169 (1347) I972/73 (1151) 1975/76 (1354' Total Katerial wet production Total Matr.-Il Pot productlon Total Noteril Net production productiun expenditure (national revenue) production expenditure (nationl reven) production expenditure (national revenue) 7nduutries exel. hdicr.fts 4.61 3.06 1.55 7.46 5.09 2.37 11.06 7.66 3.40 Nadicrafts 10.77 8.19 2.5t 12.21 9.25 2.93 12.84 9.76 3.08 Agriculture and forestry 66.77 15.74 51.03 57.73 14.35 43.38 71.44 17.63 53.81 constructlon 3.36 1.93 1.43 3.37 1.98 1.39 5." 3.27 2.17 Trasport nd comunications 2.41 0.89 1.52 3.19 1.16 2.03 3.66 1.45 2.21 Food supply *nd diftribution (7 cmerce) 9.73 2.36 7.37 10.95 2.69 8.26 14.48 3.62 10.86 Other sectors 1.46 0.48 0.98 1.66 0.61 1.05 1.76 0.69 1.09 rotal produwtion sectors 99.11 32.65 66.46 96.57 35.16 61.41 120.70 44.06 76.62 Imports - exports + 0.t4 - + 0.84 + 0.22 - + 0.22 + 1.15 - 4 1.15 Net foreign trade, aid and credits + 2.69 - + 2.69 + 2.12 _ + 2.12 + 3.18 + 3.16 Total 102.64 32.65 69.99 9.91 35.16 63.75 125.03 4.08 80.95 Source: CSO - Statisttcal Iniforation of Afghanistan (1975-76), p. 102, ed initry of Planning.  Table 2 GNP AT MARKET PRICES (In bill. Af8. at 1975/76 (1354) constant prices) 1968/69 (1347) 1972/73 (1351) 1975/76 (135l) Total Net Material Output (value added) 56.59 50.07 62.46 Trade and Transport 9.87 11.34 , 14.17 Services 5.88 6.14. 8.46 Private (1) 3.32 3.07 3.81 Public (2) 2.56 3.07 4.65 Housing (3) 4.33 4.15 5.00 National Income 76.67 71.70 90.09 Depreciation (4) 1.53 1.47 1.94 Factor Income (-) (5) 0.06 0.10 0.08 Gross Domestic Product at Factor Cost 78.14 73.07 91.95 Indirect Taxes 2.92 3.84 6.01 Subsidy (-) 0.37 0.70 1.90 Gross National Product at Market Price 80.69 76.21 96.06 (1) 5% of total material products. (2) 63% of govt. ordinary budget and 25% local financing of development budget expenditures. (3) 6% of total material products and services. (4) Average 2% of net national product at factor cost calculated on the basis of separate depreciation co-efficients in various sectors. (5) Interest on public debt is used as proxy factor income assuming that the other inflows and outflows are balanced. Source: Ministry of Planning.  - 23 - 67. The obvious discrepancies in the present GNP estimates are of course the possible large underestimations of handicrafts, private construc- tion, road transport, commerce (the income tax authorities assume 10% rate of profit on turnover in trade), services, depreciation and factor income. The view has been expressed by a number of those familiar with the data that the GDP might have been underestimated by about 15%. As discussed earlier, the absence of reliable estimates of population, employment, data on commerce and road transport, private construction and housing, household expenditure survey data, personal services, and most importantly price data and balance-of-pay- ments all point to the difficulty of arriving at a realistic estimate of national accounts. Nevertheless, given the state of statistics in Afghanistan, the difficulties in adhering to concepts and definitions, and the lack of experienced national accounts statisticians, the efforts of the responsible persons in the Ministry of Planning and the Russ.ian Advisers in preparing approximate GNP estimates should be highly appreciated. 68. The CSO has a National Income Unit with two persons trained abroad for short periods, but they are not presently working on national accounts. The CSO is studying the methodology and has expressed interest in having an expert for at least six months to initiate the work. 69. At the invitation of the Government of Afghanistan the Regional Adviser on National Accounts of ESCAP visited the CSO from March 24-May 15, 1974 to train statisticians in initiating work in the field of national accounts, review a long-range development program of statistics and make suitable recommendations. According to the Adviser's report the participants could only be taught the basics, since they had only limited practical experience in economic accounting. He recommended that for compilation of national accounts extensive economic data would have to be collected and an expert on national accounts would be required for 5 years to assist in actual estimation work and in-service training. IV. THE STATISTICAL SYSTEM A. The Statistical Law 70. Until recently there has been no legal basis for collection and coordination of statistics in Afghanistan. Before the Central Statistics Office was organized in 1972, there were some 25 statistical units in various ministries and agencies throughout the Government. There was no legal basis for coordinating the work of these statistical units, to estab- lish program priorities, to encourage the adoption of uniform standards of concepts and definitions, and to avoid duplication of effort. Further, the Government had no right to compel statistical reporting by establishments, households or persons, nor was there any guarantee of confidentiality to the respondents of statistical inquiries.  - 24 - 71. With the formal establishment of the CSO in April 1974 the need for the statistics law became more urgent, and in June 1975 the Statistical Law of Afghanistan was enacted (see Annex I). That the statistical system of Afghanistan is undergoing radical changes is apparent from the fact that the Law refers to Center for Statistics whereas it is now known as the Central Statistics Office. The Law has 24 articles divided into four sections - objectives, authority, organization, and miscellaneous regulations. Under the Law the CSO is to create a scientific system of statistics, centralize all statistical activities, and collect, evaluate and publish reliable data. The CSO has wide ranging authority which includes streamlining all statistical activities of the agencies, firms and municipalities and all public and private institutions and projects. The only exception is the Bureau of Population, Ministry of Interior, which is responsible for registration of population and issuance of identification cards. Provision has been made in the Law to protect the confidentiality of information. Hindrance of statistical activities or withholding of information is punishable by law. 72. The CSO is to be headed by a director under supervision of the Prime Minister's Office and will have a professional commission, cDmmittees on surveys, evaluation and consultation, and a technical committee. The pro- fessional statistical commission, composed of professional and high ranking Government officials from various agencies, is supposed to have an advisory role on policies, programs and professional matters of statistics. As far as the mission could determine the Professional Commission has met on about 3 occasions since the inception of the Law. However, the representation was inadequate for reaching any substantive decision. The Commission is a highly desirable institution, since the needs of the users of statistics could be formulated by it. The CSO should, therefore, actively promote the Commission and seek its guidance on suitable statistical programs. B. Organization and Co-ordination 73. Most of the statistical activities in Afghanistan are now central- ized in the CSO. However, apart from the research and study department of Da Afghanistan Bank, a few major units remain outside the CSO - the statistical departments of the Ministries of Interior, Education and Health. Although the statistics unit of the Ministry of Agriculture has been absorbed in the CSO, the Ministry has an Evaluation Department which compiles voluminous information on agricultural activities. The Central Statistics Office (CSO) 74. The following organization chart of the CSO shows that there are seven major departments - Administration; Planning; Printing and Publication; Data processing; Demography, census and cartography; Implementation; and Design and Analysis. The department chiefs report directly to the President General. The Departments have divisions and sections for various statistical activities. The Implementation Department controls the field organization with offices in Kabul (central section) and 15 provincial headquarters. Total staff in August 1976 numbered 373.  - 25 - IR ANIZATI -NT A STAIFl. nF TR. CCMRAI STATISTICS OFFICE, APCANISTAN. ALUUST A976 PriL iinietry President Ce-era I Central c. Lti-t'. Offirc | eIrEtsit Tech. iO,nnittee| hie hief Chief }Chief i|hihief Chief Chi tdnin incrar ion Planning PrintinD h Da onrecing Dt . D f - ep: P . -nftit DiT pnbiicOcnI2) C. -d 6 2 Dir.{ Dnr. Si-. -} Dr c G Dir Fe renr 1 lniiLha3 rgenc o.Dr . Di.rhd A A liraili ~~~~~~~~~~~~~~~~~~~~~~~~~~~~(8) | :=<Dir Acontei sug - in ord,narst Indct Codingxp Cinu 76ar nan) Ttlstt 7 FT, i.~~~~~~~~~~~~~~~~~~~~~~~~~~eita rio irources C Trainn D'oiiati Cptic ier _3 i Ce. Dir I ID. C--9 Di L!~P' n Dt moDroni pot i. Dir. ~ Ditsr. i Ep Dir. Conn1aSI-i. m Off 13' A Cartography ion. Dir. E3~~~~~~~~~~ hi ~ ~ ~ ~ ~ ~ c1Sta.-I  - 26 - 75. The CSO visualises its role as a central institution for standard- ization of concepts and definitions and compilation of data. There would be Liaison Officers in the various ministries, and they would ensure proper co- ordination with the CSO and implementation of revised procedures, especially for all administrative data. 76. The organization seems to be well thought out. The Co-ordination unit of the Planning Department is responsible for certifying data compiled outside the CSO. The President General could use one or two deputies to administer the departments and thus free himself to devote more time on policy and the programming of statistical activities. The Implementation department is rather small to be effective, particularly when there is no pool of enumerators who could be drawn upon for conducting surveys, etc. The CSO has the difficult task of winning confidence of the ministries, which had statistical offices in the past but now must rely on the CSO. Unless the CSO can demonstrate fairly quickly its usefulness and expanded capability in providing reliable statistics, the whole centralized structure may prove ineffective. 77. As mentioned earlier, the manpower unit of the CSO is operating as a job placement agency for the university and school graduates seeking employment in the Government and its agencies. Although this is a useful function, it is not a statistical activity. The placement operation, there- fore, should be transferred elsewhere, to an institution such as the Civil Service Department, so that the CSO can properly concentrate on manpower statistics. Da Afghanistan Bank 78. The Research and Study Department of the Bank is headed by a Director reporting to the Governor, and is organized into four sections - Foreign Trade, Central Banking, Foreign Markets, and Commercial Banking. It compiles all data on banking sectors, prepares the balance-of-payments accounts by compiling non-trade data and adjusting trade data for foreign exchange receipts, and collects data on retail prices of commodities through- out the country. The CSO clears all data compiled by the Bank, and publishes it in the statistical Yearbook. The department has a staff of about 40 peo- ple including about 10 graduates. Ministry of Interior 79. The Population Registration and Vital Statistics Department of the Ministry is headed by a President, and is organized into two major divisions - population registration, and analysis and evaluation. The Department com- piles data and records on civil registrations, issues the registration cards (Tazkiras), and has recently initiated a pilot project on vital statistics with the help of a UN expert. Besides a large staff at the headquarters, the Department has about 400 field offices with 700 statistical personnel - Directorates of Statistics in every province with 4/5 staff each, and one or two statistical clerks in each of the 325 Minor Civil Districts (MCDs) for registration of data.  - 27 - Ministry of Education 80. The General Directorate of Statistics in the Ministry of Education has about 20 people and a UN adviser on Education Statistics. The statistics unit compiles data on about 4,000 schools, training institutions and Univer- sity. Data are collected by mail questionnaire once a year in March but from 1977 will be collected semi-annually. Forms are returned within 4 months and the coverage is about 90% complete (other 10% cover only about 1% of students). The Unit has conducted seminars for proper reporting of data. Presently all tabulation is done manually which takes about 7/8 months. However, the system is being computerized which will reduce processing time to 5/6 months. Computerization will also help in timely publication of Education Statistics by eliminating typing error and reducing preparation time (now about 3 years) for publication. Ministry of Public Health 81. The General Directorate of Health Information, Ministry of Public Health is organized into 3 directorates - health resources and training, statistics on health services, and hospital statistics, and a data processing unit. There are about 20 people with 4/5 of them trained in medical record keeping of public health institutions. Data are collected every month by mail questionnaire and compiled and published every quarter. Concepts and definitions of data follow the WHO guidelines as much as possible. There is a Standing Committee on Health Information, with representatives from the various departments of the Ministry and the CSO, which provides technical guidance and advice to the Directorate. Evaluation Department, Ministry of Agriculture 82. The Ministry has a vast network of agricultural extension agents (approximately 2,600 now, and projected to increase to 5,000 at the end of the 7-year Development Plan), who provide valuable information on the effec- tiveness of programs and state of agriculture to the Evaluation Department. This information is the major source of the CSO for updating its agricultural statistics. The 2,600 extension agents are spread over 24 provinces and divided into 350 units. The agents report to their provincial directors, who in turn send all statistical information, collected on prescribed forms, to the headquarters. The extension agents live in the same villages for about 2 years and thus become familiar with local agricultural practices. However, most of them are school graduates, and have no statistical training. The Evaluation Department in Kabul is staffed with 25 graduates and headed by an agricultural statistician. 83. This is an excellent network for collecting agricultural statistics which would be both difficult and unnecessary to duplicate. The agents are familiar with data collection. Given proper training and incentives, at least at the field supervisory level, the CSO and the Ministry of Agriculture should be able to improve the quality and coverage of data at a minimal cost.  - 28 - C. Staffing of the CSO 84. The 373 staff positions of the CSO in August 1976 are almost all filled, There is no shortage of applicants, but it is rather difficult to get qualified people, or to send them to regional offices. There are about 125 graduates on the staff, with about 30 people who have had some training in statistics. The mission visited a few divisions of the .CSO and found the heads of the divisions generally well-qualified young men, quite often trained abroad, but short on experience. The CSO is aware of the training needs and during 1971-72 sent 22 persons abroad on grants and fellowships. Recently it has also actively participated in initiating teaching of Statistics in the Faculty of Economics at the University of Kabul, and hopes to attract all students with statistics majors graduating in the next 2/3 years. There is also a training unit in the CSO for organizing in-service training of the statistical personnel. An annual basic course in general and agricul- tural statistics is organized for new CSO staff. The course is given for 2 hours every day over a period of 3 months, and is presently taught by 2 foreign experts. The other training programs, however, are of an ad- hoc nature to meet the needs of surveys. 85. The promotion policy puts almost equal importance on qualification and experience. Initial placement in grades, however, is determined by qualification. The staff morale is fairly high considering the fact that most of the staff have been moved to the CS0 from their parent ministries only during the last two years. Salaries are comparable to other ministries, and working conditions and environment apparently the same. D. Training in Statistics 86. Formal training in Statistics has recently been made available in Afghanistan at Kabul University. The Faculties of Economics and Agriculture have statistics courses, and the Faculty of Engineering has an elective course in computer science taught by a visiting IBM expert. 87. The Faculty of Economics has a four-year course leading to the B.A. degree. During the first two years students take compulsory courses and in the last two years major (with course work and a small thesis) in one of 6 fields - general economic theory, statistics and econometrics, planning and development, public finance, trade, industry and business economics. All students in the Faculty take two compulsory courses in basic statistics which include descriptive statistics, probability, sampling, multiple regression and time-series analysis. The major in statistics and econometrics involves courses in demography, census and sampling, mathemati- cal economics, econometrics, linear programming and input-output analysis, national accounting, project and cost-benefit analysis, and seminar and short thesis on applied Afghan economics and statistics. It is indended that the statistics major students will work in the CSO for 2/3 months each year to gain some practical experience. The first batch of statistics major students (about 25) will be graduating in 1977. The Faculty has about 600 students, and 40 teaching staff with about 12 Ph.D.'s (from foreign Universities). The  - 29 - languages of instruction are Dari and Pushtu, and the staff provides printed lecture notes to the students. The Faculty has no calculators for statistical laboratory work, and is rather short on books, statistical mannuals, and other teaching materials. There is a Quarterly Journal of Economic Studies published in Dari and Pushtu languages. The mission felt that while the teaching of basic statistics to all economics students was commendable, the course work for the statistics majors was rather ambitious for under- graduates. Also the students would probably have better analytical training by joining the Faculty members in their research projects and participating in some field work for data collection instead of doing desk studies for their theses. 88. The Faculty of Agriculture has a compulsory course in Principles of Statistics for all 4th year students in the Faculty whether they are majoring in plant science, animal science, or agricultural economics, extension and education. The course was initiated 2 years ago. It covers descriptive statistics, probability, sampling, frequency distributions and various statis- tical tests. There are plans to introduce a further course in design of experiments and field-plot techniques in the next 2/3 years. The languages of instruction are Dari and Pushtu, but all teaching materials are in English. The Faculty publishes a quarterly journal, called Science and Skill, and also has a series of technical bulletins (about 35), research notes (about 20), and general publications (about 5). Some of the interesting bulletins are on the cost of production and returns to grape farming (1968), wheat (1970), cotton (1971) and sugarbeet (1971) which reported results on regional sample survey studies conducted by the Faculty members and students. The Faculty has about 700 students with 80-90 in the current graduating class. The economics staff of the Faculty regretted that they had no contact with the Ministries of Agriculture and Planning, and the CSO. E. Data Processing 89. The data processing facilities are rather poor in the CSO. The offices are mostly equipped with some manually operated and electric cal- culators, usually about I machine for 4 assistants. The data processing unit has about 20 punched card machines and verifiers, although only 5 are in working order due to lack of proper maintenance and spare parts. The unit has five computer programmers and analysts who utilize computer time available at Afghan Computer Center (ACC) for data processing such as the external trade statistics. However, although the CSO has plans for establishing a data bank, no major effort has yet been made. 90. Afghan Business Machines (ABM), a private organization and subsi- diary of IBM, has recently been reorganized as the Afghan Computer Center. The major shareholders now are Da Afghanistan Bank, Ariana, Bank Melli, Afghan Textile Co. and Hotel Holding Corp. The biggest customers of ACC presently are the Agricultural Bank (fertilizer, seed distribution records), Ariana (all operations), Bank Melli (current and loan accounts), Afghan Textile Co. (inventory, etc.), and the Ministry of Education. The president of ACC hopes that Da Afghanistan Bank statistics department will utilize the computer in near future. The CSO is also considering processing the 1978 population  - 30 - census data at ACC. There does not seem to be any immediate possibility of attracting work from other Government agencies, though the Ministry of Finance and large public enterprises should be in a position to take advan- tage of computer processing rather effectively. 91. The hardware consists of an IBM 360/30 computer with 32 K memory operating on a tape system. It is supported by a 2540 card reader and 1403 printer. The computer is operated for about 120 hours per month. It is serviced by a locally stationed IBM customer engineer. Although the repair parts have to be flown in from Paris, the servicing is deemed satisfactory. ACC programming staff consists of a senior systems analyst and 10 programmers including 3 trainees (4 persons are expatriates). If the population census is processed at ACC, the staff would increase to about 5 systems analysts and 20 programmers. Although Government agencies show little enthusiasm for computerization, there is no shortage of qualified trainees since Afghan programmers would have ample job opportunities in Iran. Most of the data processing is done in COBOL programming language. RPG, Assembly and Fortran languages are also available. ACC thinks that at the rate of 4,900 Afs/hr. it offers the cheapest computing facilities among the neighboring countries. If it can attract more business the rate will be lowered, and a bigger computer could easily be installed. It may be noted that ACC offered some free time to the Kabul University faculties and students but with very little response. 92. There is also a small IBM 1130 computer at the Institute of Meteoro- logy. This machine was a gift to the Afghan Met. Organization from the World Met. Organization (WMO). The computer was lying idle for quite some time at ACC and has only recently been installed at the Institute. It will be used only for meteorological work, and there are two WMO experts initiating the operations. Presently there is a small unit with one programmer and few keypunch staff. 93. Since ACC is under-utilized both in computing and programming operations, it is regrettable that the CSO and Da Afghanistan Bank have not computerized their routine operations and data files. The CSO seems to be particularly well-equipped to do so with its own programming staff to mobilize and oversee the statistical aspects of data processing. V. FUTURE PROGRAM OF WORK OF THE CSO 94. The Regional Adviser on National Accounts of ESCAP, after his visit to Afghanistan in March-May 1974, reported that the CSO was considering a 5-year development program to be undertaken with the advice of a US Bureau of Census Team. The advisory team consisted of 5 experts specializing in general, industrial, agricultural and foreign trade statistics and data processing, and was in Afghanistan from 1972 to mid-1975. The program envisaged a large number of surveys:  - 31 - (a) Regular surveys (i) Agricultural statistics -- land use and yield estimates (ii) Agricultural sample surveys -- producers and field agents (iii) Surveys of prices paid and received by farmers (iv) Remote sensing census project (v) Quarterly survey of large establishments (vi) Annual census of large establishments (vii) Sample survey of cottage and small industries (viii) Multi-purpose surveys of urban areas (ix) Index of industrial production (x) Survey of Government-owned construction (xi) Survey of privately owned construction (urban and rural) (xii) Improvement of consumer price index (xiii) Wholesale price series (xiv) National accounts (xv) Multi-purpose household sample survey (xvi) Foreign trade statistics program (xvii) Data processing (b) Ad hoc surveys and work (i) National survey of nomads (ii) Development of establishment register for large, small and cottage industries. The Regional Adviser then concluded that for compilation of national accounts (including regional estimates for certain fields) and data needs for policy, the above program should be expanded to include: (i) Survey of wholesale and retail trade (ii) Survey of land transport  -32- (iii) Population census (iv) Survey of hotels and restaurants (v) Survey of recreational and cultural services (vi) Survey of personal and household services (vii) Livestock production (viii) Disposal of crops (ix) Cost of cultivation. He also recommended that to improve the data collected through the adminis- trative agencies, the CS0 should train the agency staff and provide the necessary guidance. Furthermore, to undertake all these activities, it was suggested that an integrated project of statistical development for 5 years should be formulated, with a team of 6 advisers on sampling and training, national accounts and economic statistics, agriculture statistics, industry, construction and trade, demography and social statistics, and data process- ing. Thus, as can be seen from above, in the opinion of the experts a tremendous effort is required for the proper development of statistics in Afghanistan. 95. The mission, during its visits to Afghanistann, found that only limited progress has been made in upgrading the statistics. The efforts did not give the impression of following any integrated approach. The CSO has undertaken regular monthly surveys of modern industrial establishments, completed a banking survey and a partial survey of cottage industries, and initiated improvements in foreign trade statistics. Furthermore, a major project on the population census, to be held in mid-1978, is under way. Expert advice to the CS0 is available only for the census from a team of four UN Fund for Population Activities advisers. The CSO has also indicated that serious consideration is being given to a road transport survey and an agri- cultural census in 1980. 96. In January 1977 the CS0 conducted a Survey of the Statistical System in Government Agencies. Questionnaires were mailed to about 135 agencies in the Ministries, municipalities, banks, public enterprises, etc., requesting reply within 3 weeks. By the middle of May only 84 replies had been received. The questionnaire asked factual information about what statistics were collected, what other data were needed, who used them, where they were published, whether there were any departments of statistics or liaison units, any plans for statistical surveys and computer use, and also opinions about who should collect data (the agency or the CSO), realiability of data used in the preparation of the Development Plan and those published by the CSO, and the appropriateness of the Statistical Law governing the statistical system of data collection and analysis. Unfortunately the replies have not yet been processed, and therefore there is no firm basis to  - 33 - make any judgment. It appears, however, that the majority of the agencies would prefer to collect their own data and send the relevant information to the CSO, and whether they collect data or not would like to establish a statistical liaison department in their agencies. The mission feels that the Survey results will be helpful to the CSO in planning and coordination of its activities with the various agencies, and, that the CSO should undertake the consolidation of replies and extensive discussions with the participants, as soon as possible. 97. As for the CSO, since the census operations require vast organiza- tion and manpower, it is obvious that for the next 5 years it will be hard put to improve statistics in other fields. The regular work itself may suffer from inadequate staffing and direction. Recognizing the strain which the census and a program of improving current data series will impose, the mission suggests, therefore, that the future development of Afghan statistics should proceed with a) the formulation of a long-term program of statistics in collaboration with the Professional Commission on Statistics, b) improve- ment of presently viable statistical data collection procedures, c) imple- mentation of a National Economic Survey in 1978-79, and d) population census in 1978. These could be followed at a later date by a household expenditure survey, agricultural census and other statistical programs. 98. The long-term program of statistics, particularly its integrated approach with user-orientation of statistics, and emphasis on coordination with and participation from the various Government agencies are essential to the proper functioning of the CSO. As mentioned earlier, the CSO should promote the legally established Professional Statistical Commission and seek its guidance on a suitable statistical program. In addition, the CSO should maintain a very close relationship with the Ministry of Planning since the latter is the largest and most important user of statistics, and could con- tribute to the development of statistics by setting down their data needs for planning and evaluation. 99. The presently viable statistical programs must be strengthened despite the pressure of the impending population census. Specifically, better quality control and expanded coverage are needed for the statistics of general government accounts including both central and the available municipal accounts, the analysis of the budgets of public enterprises and compilation of their accounts, collection of wholesale prices in addition to retail prices, improved estimates of the outputs of agricultural crops and industries, and better external trade and balance of payments data. Considerable improvement could be achieved in these areas with a relatively small effort. 100. A National Economic Survey is proposed for 1978-79 to compile comprehensive economic statistics for the global benchmark estimates of the Afghan economy. This Survey should provide adequate data for the national accounts estimates, and also identify the gaps in information, suitable procedures and relevant areas of data collection, and the costs and benefits of compilation of various statistics vis-a-vis the structure of the Afghan  - 34 - economic system. The Survey would mainly rely on the available data collec- tion processes, but supplement the system with an adequate number of properly designed small sample surveys not only to fill in the gaps in existing data but also gather new information on other aspects of the economy. The only way such a series of small sample survey could be effective in obtaining global estimates for the country is to link them with a census. Hence the Survey should be conducted in 1978-79, as a short-term project with field work carried out over a year, and the samples based on the population census frame of the sketchmaps of villages and preliminary counts of population which are expected to be available at the beginning of 1978. A team of 3 experts - one in sampling and 2 in National Accounts - with about 6/8 professionals and 25/30 enumerators should be able to complete the National Economic Survey within 2 years. While the experts would probably be expatriates, the CSO should not have difficulty in providing the small requirements of other staff. CSO professional staff will gain valuable training in the process. Since the Survey seems to be the only method of obtaining the national accounts estimates and comprehensive statistics of Afghanistan within a short time at reasonable cost, the Mission urges giving high priority and immediate consideration to this project. 101. The population census of 1978 has been discussed earlier, and the preparation for census is already in progress. Since the statistics on population are the basis of various types of classifications and estimation procedures in other statistical fields, and since there has never been a census in Afghanistan, the census is a crucial element in any program of statistical development. 102. After the census of 1978, the long-term statistical program should be programmed in collaboration with the Professional Commission on Statistics on the basis of the information obtained from the National Economic Survey. The need for a Household Expenditure Survey in Afghanistan deserves emphasis however. It should be conducted as early as possible after the Census to utilize the population census frame. In considering the long-term program, the CSO might also wish to examine the advisability of instituting a system of annual household surveys, such as the Indian National Sample Survey and the UN African Household Surveys Program, which would cover various subjects (e.g., demographic characteristics, health and nutrition, housing, education, economic level of household etc.) in a four/five year repetitive cycle for up-to-date socio-economic statistics.  - 35 - ANNEX I Page 1 STATISTICAL LAW OF AFGHANISTAN Order'No. 421 (12 June 1975) from the Office of the Prime Minister: To: Mr. Abdol-Karim, Director of the Center for Statistics, Office of the Prime Minister The Council of Ministers, through Its resolution No. 922 of the session on 10 June 1975, approved the following, after its consideration of the request No. 282, June 7, 1975 from the Center for Statistics regarding the Law on Afghanistan Statistics: "Law on Afghanistan Statistics containing twenty-four articles and bearing the seal of approval of the Secretariat is herewith submitted" Provisions of the enactment by the High Council of Ministers pre- sented and approved by the Honorable President of the Republic of Afghanistan, and registered under No. 593 dated 11 June 1975 is submitted for your proper action. Dr. Mohammed Hassan Sharq Archives entry No. 271 Secretariat entry No. 317 12 June 1975 14 June 1975  - 36 - ANNEX I Page 2 Statistical Law of' Aflhatistan. Sectioi-I- data and statistics based vities and for establishing on scientific and interna- a continuous and constant tional standards. system of data collection Objectives and distribution. C. To obtain, analyse, te- ARTICLE ONE: Section l 1- Cen- st, compile and unify stati- stical data and figures and Considering Act. No. 436 + I publication of compiled of the Cabinet 20-1-1353, tral ULtlistis data. which has been signed and D. To design programS approved by the President Office (CSO) and carry out the data pro- of the Republic of Afgh- " cessing programs of the anistan and Eas been en- data and information co-I acted by Executive Order Authority lected as a result of surv- No. 102 dated 22-1-1353 eys and other statistical of the Prime Ministry, the programs. Central Statistics Office wi- th the following objectives ARTICLE TWO: E. To print and publisli. has been established dire- The Central Statistics annual statistical informat- ctly under Prime Minist- Office has been empower- ion about Afghanistan er's anthority: ered ,with the following functioids and responsibilit- F. To arrange and con- A. To establislh a scienti- ies: duct professional training fic system of statistics, programs for the purpose for collection of statistical A. To design programs of furnishing and provid- data and information to pr for data collection, carTy- ing more information for ovide for an effectJve and ing out statistical surveys, the CSO personnel in var- better execution of activit- conducting population cen- ious branchs of statist ies, programs and plans of sues and compilation of re- and raising their level of guidance and direction of iated data, figures and sta- understanding and qualifi- economic and social aff a- tistical infor mation within cation. irs. the framework of approv- ed statistical objectives and G. To establish ways and B. To centralise, unify policy. means to utilize all pers- and coordinate all statisti- onnel, equipment, comm- cal activities and avoid du- B. Design and preparati- odities and funds assigned plication of functions. on of fornis, procedures, and provided for the statis- questionaires and tables for tical activities in the Gov- C. To collect, measure, the implementation of ap- ernment agencies, enterp- analyse, evelaute and pu_ proved statistical programs rises, municipalities, pub- blish continuous and cons- and activities and for st. lic and private institutions tant reliable a'nd required atitical programs and acti- to *a maximum extent.  -37 - ANNEX I Page 3 11. To establish and inu- comlpilation and analysis c.rnccinillg publications of intain coordination and of the statistical informa- the collected data and in- harmonization among all tion and data concerning formation of the various statistical activities carried following objectives: activities and sectors of otit by Government agen- the country. cies, enterprises, nimunicip- 1. Agricultural statistics alitres, projects and public 2. Energy and industrial AI'FICLE SIX and private institutions by statistics. introducing the following: Exchange of statistical 3. Mining and construc- information with the Uni- 1. Unified ternminology, tion statistics. ted Nations. national and nanuals, definitions, pro- 4. Communication anid international organizations (edures, forms and tables. and scientific statistical transportation statistics. centres. 5. Domestic and foreign 2. Centralization of sta. trade statistics (balance of ARTICLE SEVEN: tistical surveys and resear- payment). ches. 6. Social services statis- When considered neces- 3' Assigning professional tics. sary, participation in var- statistical personnel. 7. Price and banking std- ious international statisti- 4. Implementation of the tistics. cal seminars, conferences required statistical servi- 8. Manpower and labour and workshops. ces for the Government ag- statistics (labour force). cncies, enterprises, muri- 9. National accounts (f1- ARTICLE EIGHT: cipalities and all other pu- nance and national inco- blic and private institut- me). All personnel and offic- ion. 10. Household income and ial staff of the Central Sta- expenditure statistics. tistics Office ewho are eff- 1. Determination and 11. Demography and icially assigned to prepare, approval of the qualificati- censuses. analyse and evaluate rep- on ~~~~~~~~~~~~~~~~~~orts and data zrz- leftaliS on and definitir.n of furlc- 12. All other statistics bound to take an oath that tions of the professsional related to various aspects they will not disclose clas- statistical experts and ad- of the economy. wil (confiential data visors who are being em- :,iried (confidential) data ployed by Government ag- Note: All the above men- and statistical information. encies, enterprises, proj- tioned activities and statis Note: The above menti- ects and public and priva- tical terminology vill hle oned decree also applies to te instituitions for the pur- defined, interpreted and all persons in the various pose of improving statisti- impleniented on the basis statistical activities in the cal activities. of international standai ds CSO and statistical definitions. J. To establish and de- selop a national statisti- ARTICLE FOIJR: ARTICLE NINE: cal information centre and a data bank for the Topeaeastii- OrTicials and persons as- purposp of storage and dis- Texconand pea esar s i se to C^ ar f tributiori of statistical in- leduicon ( fand celisary preo- dlen to divulge classified fonnation about Afg-h3 n- oeduinstion ofacollaetheg an- statistical data and ipiforin- istan. odnto fcletn n ation. If classified (confid- utilisation of data and sta- ential) statistical data and ARTICrLE THREE: tistical information, infornmation are revealed for 'he personal benefits ARiTICLE FIVE: and other purposes or if To design ar.d conduct the confidential or restrict- statistical program oscca- To study and approve all ed data and information sionally and continuously statistical programs, whi- -re intenationally altered in the various fields arid ch may be undertaken by or torged, such acts are activities of natioral eco- individuals and leal in- lezally considered a cr-ime, nomy which is required stititions, before such pro- and thP persons rommitting for socio-econonmic plan.. graliis are implfemented sucrI acts shall he punished ning inclu(ding collection. Alto issuance of prrmitrn *3U cording to law.  - 38 - ANNEX I Page 4 S ection I~I 0, CSO and professional equipment and vehicles, su- Sectioll IIIaznd high ranking repres- pplies and services wbich entatives of Government arc needed during the cen- Organization agencies. Members of the suses and complete and ge- commission shall be app- neral statistical surveys ointed on the basis of the in accordance with the Go- According to the Budg- CSO President General's vernment decisions. et and Accounting and the proposal and approval of Organization's law of the thie Prime Minister. The state, CSO shall have its professional statistical co- ARTICLE SEVENTEEN: own budget and shall be mmission shall have advi- directed by a Presideut Ge- sory authority and shall During census numera- neral. The President Gene- exper ssits views on statis- tion and complete agricul- ral of CSO shall have full tical policies and programs tural, economic and social rauthofrCSO sand havesp ul- and other related specia- surveys all individuals and authority and responsffli- 117ed and professional ac- citizens of the country, ex- cial fornctilel and affas tivities and subject matt- cept those serving under cial functions arnd affars ers. Decisions of the com- national defense and se- of CSO. mission shall be erforced curity as wvell as enumera- ARTICLE ELEVEN: after the approval of the tors, are bound to stay at ARTICL ELEVENCSO President General. their regular places and The organization of CSO localities. is composed of a President ARTICLE EIGHTEEN General, professional com- Section IV mission and committees During cenisus enumera- concerned with inspection, and statistical surveys, evaluation, consultation, Miscellaneous CSOs assigned officials and and functional and admini- enumerators are bound to strative departnments. ARTICLE FIFTEEN: follow all CSO's instructi- ons and directions without ARTICLL TWEL" E: All Government agenc- any bias. ies, public enterprises and The President General institutions and all actual ARTICLE NINETEEN: and heads of the -departmn- and legal persons are leg- ents of CSO will be appo- ally bound to submit to All statistical data inted from persons who ha- CSO all information need- and information collected ine fm ped for arranging and org- and compiled by CSO as ve academic education in ve aaec o Inanising related statistical a result of census and sta- statistics or economics or information; systems for tistical surveys are preserv- from persons v:th suffici- the purpose of planning ed and their general res- ent experience in the men- and direction of economic tilts will be published. tioned fields. and social affairs of the ARTICLE THIRTEEN: courntry, vith the exception ARTICLE TWENTY: of State classified data and documents. Such reli- Persons whose statem- The organisat on and bu- able statistics should bc ents, speeches or actions dget of csn. its departm- based on facts and shall he disturb public order dur- ents and related offices sh- provided to authorized per- ing census enumeration or all be included In the Go sonnel of CSO when it is complete surveys shall be 'e:rnmcntos annual budget required. Information sub- ipdirted for the act of dis- in accordance weth the mitted by actual ar,d le- turbing public order. budgetary procedure. gal persons are only used for statistical purposes. ARTICLE TWENTY ONE: ARlTICLE FiOURTEEN: ARTICLE SIXTEEN: The competent authority CSO shall have a profes- for civil registration and siornal statistical comniis- Government agencies and issuance of citizenship car- sion. The comm'ission shall public enterprise s and ins- ds based on international include professional vnd titutions are hbound to pro- standards and in accordan- qunlified staff members vide tn CSO all required ce with legal procedures  - 39- ANNEX I Page 5 shall be the Dcpartment of which havc been publish- contradiction to the text Civil Registratioti of the ed, distributed or verified of this law shall be void Ministry of Interior. by CSO. from the date this law co- nes into force. ARTICLE TVENTY TWO: ARTICLE TWENTY THREE ARTICLE TWENT'Y FOUR: This law shall coine into' Official statistics shall be All legal procedures and force X hen published in the data and information regulations which are in the Official Gazette.   PART II STATISTICAL APPENDIX Table of Contents Page No. 1. POPULATION AND EMPLOYMENT 47 Notes on Population of Afghanistan 49 11 " Population Projections 51 Tables 1.1 Population and rate of growth, 1972/73-1976/77 61 1.2 Demographic characteristics of settled popula- tion 1972/73 62 1.3 Structural characteristics of the population of working age, 1972/73-1976/77 63 1.4 Labor force composition, 1975/76 64 2. NATIONAL ACCOUNTS 65 Notes on National Accounts 67 it " Absolute Poverty Levels 75 Tables 2.1 GNP estimates (1975/76 constant prices), 1968/69-1976/77 79 2.2 Expenditures on GDP (1975/76 constant prices), 1968/69-1976/77 80 3. FOREIGN TRADE AND PAYMENTS 81 Tables 3.1 Value of principal exports, 1966/67-1975/76 83 3.2 Value of principal imports, 1966/67-1975/76 84 3.3 Value of exports by regions, 1966/67-1974/75 85 3.4 Value of imports by regions, 1966/67-1974/75 86 3.5 Export price indices of major commodities (1968/69=100), 1968/69-1976/77 87 3.6 Major exports quantity indices (1968/69=100), 1968/69-1976/77 88 3.7 Tourist arrivals by Air and by Land, 1967/68- 1976/77 89 3.8 Balance of Payments, 1969/70-1976/77 90 3.9 International Reserves and related items, 1971/72-1976/77 91 3.10 Merchandise exports, 1976/77 and 1982/83 Seven Year Plan 92 3.11 Merchandise imports, 1976/77 and 1982/83 Seven Year Plan 94 3.12 Balance of Payments, Seven Year Plan 96 3.13 Composition of Foreign Economic Assistance, Seven Year Plan, 1976/77-1982/83 97 3.14 Projected Exports by Principal Commodoties, 1977/78-1982/83 98 3.15 Projected Imports by Principal Commodities, 1977/78-1982/83 99 3.16 Afghanistan: Free Market and related Exchange Rates 100 - 41 -  - 42 - Page No. 4. EXTERNAL DEBT 101 Tables 4.1 External Public Debt by Principal Lenders, March 21, 1977 103 4.2 Projected External Public Debt by Principal Lenders, March 31, 1977 104 5. FISCAL STATISTICS 107 Notes on Tax System 109 Presentation of Government Accounts 117 Exchange Rate Subsidy 121 Tables 5.1 Summary statement of Central Government Finances, 1972/73-1977/78 123 5.2 Central Government Revenue, 1972/73-1977/78 124 5.3 Central Government Ordinary Expenditure, 1972/73- 1977/78 125 5.4 Distribution of Central Government Expenditures, 1972/73-1977/78 126 5.5 Summary statement of Central Government Finances (1975/76 constant prices), 1972/73-1975/76 127 5.6 Central Government Revenue (1975/76 constant prices), 1972/73-1975/76 128 5.7 Central Government Ordinary Expenditure (1975/76 constant prices), 1972/73-1975/76 129 5.8 Data Concerning Tax Incidence, 1976/77 130 5.9 Consolidated Budgets of Municipalities and of Kabul, 1975/76-1977/78 132 6. MONETARY STATISTICS 133 Tables 6.1 Monetary Survey, 1971/72-1977 135 6.2 Changes in Money and Quasi-Money and Factors Affecting Changes, 1972/73-1976/77 136 6.3 Consolidated Statement of Specialized Banks, 1971/72-1975/76 137 7. AGRICULTURAL STATISTICS 139 Tables 7.1 Land Resources and Present Utilization 141 7.2 Climatic data: A. Total Recorded Precipitation 1968/69-1975/76 142 B. Climatic Data for Selected Towns 143 C. Monthly Mean Temperatures at Selected Stations 144 7.3 Condition of the Irrigation System 145 7.4 Area of Agricultural and Horticultural Crop Production 146 7.5 F.A.0. Agricultural Production Indices (1961-65=100), 1965-1976 147 7.6 Trends in Production of Principal Crops, 1970/71- 1976/77 148 7.7 Agricultural Production Targets in Seven Year Plan, 1975/76 and 1982/83 149  - 43 - Page No. Tables 7.8 Production of Important Agricultural Products according to Seven Year Plan, 1968/69, 1975/76-1982/83 150 7.9 Projection of Main Agricultural Products, Seven Year Plan, 1975/76-1982/83 154 7.10 Actual and Projected Cotton Production by Provinces in Seven Year Plan, 1975/76 and 1982/83 155 7.11 Actual and Projected Fruit and Vegetable Production in Seven Year Plan, 1958/59, 1975/76 and 1982/83 156 7.12 Wheat Purchases and Imports of Food Procurement Department, 1964/65-1975/76 157 7.13 Fertilizer Use in Afghanistan, 1966/67-1976/77 158 7.14 Annual weighted average National Prices of Selected Agricultural Items, 1970/71-1976/77 159 7.15 Agricultural Exports Quantity Indices, 1968/69-1976/77 160 7.16 Volume and Value of Specific Fruit and Nut Exports, 1972/73-1975/76 161 7.17 Analysis of Lending Operations and Loan Recoveries, Agricultural Development Bank, 1973/74-1975/76 162 7.18 Capacity of Projects and Development of Irrigation during Seven Year Plan, 1976/77-1982/83 163 7.19 Agriculture and Irrigation Development Expenditures, 1969/70-1982/83 164 7.20 Agricultural Co-operatives 165 8. INDUSTRY AND MINING 167 Notes on Investment Law, 1974 169 Tables 8.1 Major Industrial and Mining Production by Principal Commodities, 1967/68-1976/77 173 8.2 Handicrafts and Handmade Products, 1974/75 174 8.3 Project Composition of Major Sub-Sectors of Industrial and Mining Activities, Seven Year Plan 175 8.4 Major Industrial Enterprises by Type and Ownership 179 8.5 Potential Large and Medium Scale Private Industries 181 8.6 Small-Scale Industrial Activities to be Financed Under IDBA's Development Assistance Fund 182 9. ENERGY AND POWER 185 Tables 9.1 Coal Characteristics 187 9.2 Characteristics of Locally Produced Oil and Gas 188 9.3 Primary Energy Reserves 189 9.4 Energy Balances, 1970/71-1976/77 and 1982/83 190 9.5 Natural Gas Production, Consumption and Exports, 1967-1976 191 9.6 Crude Oil Production, 1967/68-1976/77 192 9.7 Imports of Petroleum Products, 1970-1976 193 9.8 Main Coal Consumers 194 9.9 Fuel prices 195 9.10 Projected Coal Production and Consumption, Seven Year Plan 196  - 44 - Page No. Tables 9.11 Projected Natural Gas Production, Consumption and Exports, Seven Year Plan 197 9.12 Projected Production, Consumption and Imports of Petroleum Products, Seven Year Plan 198 9.13 Projected Energy Sector Investment (excluding Power), Seven Year Plan, 1976/77-1982/83 199 9.14 Electricity Generating Capacity, 1977 200 10. TRANSPORT AND TELECOMMUNICATIONS 201 Tables 10.1 Motor Vehicles Registration by Province, 1975 203 10.2 Summary of Total International Passengers Movement, Kabul International Airport, 1969-1976 204 10.3 Anticipated Development Expenditure of the Ministry of Public Works, Seven Year Plan, 1976/77-1982/83 205 10.4 Customs Duties on Motor Vehicles 208 10.5 Telecommunications facilities in Afghanistan, 1966/67- 1975/76 210 10.6 Telephone System in Afghanistan, 1966/67-1975/76 211 10.7 Expenditure flow of the Telecommunications System, 1966/67-1976/77 212 10.8 Public Sector Investment Programs in Telecommunications. Seven Year Plan, 1976/77-1982/83 213 11. PRICES 215 Notes on consumer price index 217 Tables 11.1 Major Consumer Goods Price Indices (1968/69=100) 1968/69-1975/76 219 11.2 Annual weighted average prices of consumer goods in the City of Kabul, 1972/73-1975/76 220 12. EDUCATION STATISTICS 221 Tables 12.1 Enrollment Growth in Education 1964/65-1974/75 223 12.2 Enrollment by Province and sex, 1974/75 224 12.3 Teacher Training Institutions, 1975 225 12.4 University of Kabul, Intake and Output, 1976-1978 226 12.5 Dropout and Repetition Rates, 1974 227 12.6 Primary Education, Seven Year Plan Targets, 1976/77- 1982/83 228 12.7 General Secondary Education, Seven Year Plan Targets, 1976/77-1982/83 229 12.8 Vocational Education, Seven Year Plan Targets, 1976/77-1982/83 230 12.9 Teacher Education, Seven Year Plan Targets, 1976/77- 1982/83 231 12.10 New Entrants into Kabul University, Seven Year Plan Targets, 1976/77-1982/83 232 12.11 Enrollments in Kabul University, Seven Year Plan Targets, 1976/77-1982/83 233  Page No. 12.12 Graduates from Kabul University, Seven Year Plan Targets, 1976/77-1982/83 234 12.13 Estimated Targets & Phasing of National Plan of Nonformal Education, Seven Year Plan, 1976/77- 1982/83 235 12.14 Qualifications of Primary Level Teachers, 1974/75 236 12.15 Qualifications of Secondary Level Teachers, 1974/75 237 12.16 Enrollment in Vocational Secondary Schools, 1974/75 238 12.17 Distribution of Vocational School Teachers by Type of School and Qualification, 1974/75 239 Charts 12.1 Organizational Structure of the Ministry of Education 240 12.2 Education Structure, 1975 241 12.3 Revised Educational Structure, 1975 242 12.4 Educational Pyramid, 1974/75 243 13. HEALTH STATISTICS 245 Tables 13.1 Health Facilities by Province, 1976/77 and 1982/83 247 13.2 Trained Medical Personnel, 1976/77 and 1982/83 (projected) 249 13.3 Family Planning Services, 1972 and 1976 250 14. INCOME DISTRIBUTION 251 Tables 14.1 Distribution of Income (I), 12 villages in 7 Provinces, 1970/71 253 14.2 Distribution of Income (II), Kabul City, 1968/69 254 14.3 Distribution of Consumer Expenditure, Lashkar Gah, 1968/69 255   - 47 - 1. Population and Employment Page No. A. Note on Population of Afghanistan 49 B. " " Population Projections 51 Tables 1.1 Population and rate of growth, 1972/73-1976/77 61 1.2 Demographic characteristics of settled population 1972/73 62 1.3 Structural characteristics of the population of working age, 1972/73-1976/77 63 1.4 Labor force composition, 1975/76. 64   - 49 - A. Population of Afghanistan There has never been a population census in Afghanistan. In the past, from various surveys and partial censuses, the population of Afghanistan was estimated as high as 19.28 mill. in mid-1975. However, for the 7-year Plan of Afghanistan, a ministerial committee reviewed all available sources and decided on the estimate of 16.67 mill. in 1975/76 (1354). Preparation for a population census in September 1978 is in progress in Afghanistan. In the view of the World Bank, available evidence from partial sketch-mapping of villages and preliminary population counts suggest an aggregate population in the 12-15 mill. range in 1976/77 (1355). The mission derives an estimate of 14 mill. in 1976/77 from the following: 1) Up to March 1977 mapping and prelisting of population had been completed in 68 minor civil divisions (MCDs) in the rural areas of 7 provinces out of a total of 325 MCDs. Assuming these rural areas are representative of the whole country, the total rural population is estimated at about 10 mill. Adding provisional estimates of 2.3 mill. for urban areas and 1.2 mill. for nomadic population, the total population of the country amounts to 13.5 mill. 2) During 1974-76 the Ministry of Interior completed the civil registration of all settled population in Afghanistan. Over 6 million registration cards (tazkiras) were issued which include all males and some females (e.g., those seeking school enrollment and employment). A slight underestimation is expected since parts of 4 border provinces do not fall under compulsory registration. From these recent data, the settled population for 1976/77 has been estimated at 12.4 mill. Adding 1.4 mill. for nomads (estimate of the Ministry), the total population of Afghanistan comes to 13.8 mill.  - 50 - 3) The Afghan Demographic Surrey (ADS), an extensive demographic and family guidance survey conducted in 1972/73# derived the estimate of 10.44 mill. in 1972/73 for the sedentary population. A major criticism of the survey was that the population might have been underestimated by about 5-10% due to underreporting of females (e.g., observed sex ratio of 116 males/100 females considered too high). Adjusting the figure of 10.441 mill. upwards generously by 10% , and applying the annual rate of natural increase of 2.2% observed by the ADS, the 1975/76 estimate of the sedentary population amounts to 12.26 mill. Assuming the nomadic population to be 1.2 mill, the total population is again 13.46 mill. While waiting for the results of the population census - preliminary estimates expected in early 1979 - the mission feels that the following workcing estimates in round figures are reasonable approximations of the population for 1976/77 (1355): Mission Pre- Registration Adjusted Plan Estimate Census Data ADS (75/76) Fairal 10.5 mill. 10.0 .. .. 11.87 Urban 2.0 " 2.3 .. .. 2.39 Total settled 12.5 n 12.3 12.4 12.26 14.26 Nomads 1.5 " 1.2 1.4z 1.20 2.41 Total 1J4.0 n 13.5 13.8 13.146 16.67 In comparison with other recent estimates, the mission's figure for urban population is slightly lower but for nomadic population slightly higher. This is partly due to rounding, and partly because the mission thinks that urbaniza- tion, except for Kabul, has not progressed rapidly in recent years and the nomadic population figure might be closer to the estimate of Ministry of Interior who ought to have a better feel with its network of offices around the country.  - 51 - B. Population Projections As broadly indicative of the population prospects of Afghanistan if fertility is not reduced, several simulations have been run under different assumptions of mortality improvement. Projection I - Mortality (life-expectancy) remains constant until the year 2025. This model is very unlikely to be realized, but does demonstrate the significant impact of reducing mortality levels. (Corresponds to Scenario 1 in Vol. 1, Ch. 3). Projection II - Mortality improves more slowly than in III to account for the extremely difficult environment in which health services must be delivered in Afghanistan and the necessarily slow rate at which some basic problems (e.g. drinking water) can be solved (see below). Projection III - Mortality (life expectancy) improves according to a decision rule developed by the MI as representative of past and current experience of most developing countries (also see below). This path would possibly be achieved if there were a sudden major improvement in health effort and program orientation (corresponds to Scenario II in Vol. 1, Ch. III). Table I: Life Expectancy at Birth Assumed in Alternative Mortality Assump- tions: 1975-2025;E a/ Projection I Projection II Projection III 1975 35.0 35.0 36.0 1980 35.0 36.6 38.5 1985 35.0 38.1 41.1 1ggo 35.0 39.8 43.5 1995 35.0 41.8 46.c 2000 35.0 43.8 48.5 2005 35.0 46.3 51.0 2010 35.0 48.8 53.5 2015 35.0 51.8 56.0 2020 33.0 54.7 58.2 2025 35.0 57.6 60.5 Note: a/ 1) Decision rules contained in Appendix 1. 2) Life expectancy at birth for 1975 are not all equal because they were initially set equal for 1973, and increased from there by different decision rules.  - 52 - Exact estimation of the nomadic population is impossible. Only 1/ most tentative estimates place their number in the 1.2-1.5 million range. But not on'y are their numbers and vital statistics unknown but also the rate at which they will be absorbed into the settled population and assume their 2/ demograph characteristics. In order to allow some idea as to the population of current nomads by the year 2000, two limiting estimates were made: High Estimate - Assumes all current nomads would be absorbed into the settled population through migration and immediately 3/ assume all its demographic characteristics. "Nomads" as such would cease to exist. Low Estimate - Assumes the current nomadic population remains nomadic and continues to display unchanged vital characteristics through the year 2000. It is assumed their natural rate of increase is 1.8% (the low end of the estimate range for the total population (see Vol. 1, para 3.4). 1/ A 1975 estimate of 1.4 million is used, based on 1972/73 ADS nomadic population estimate of approximately 1.3 million. 2/ It is believed, however, that the nomadic population has higher fertility, and higher overall and infant mortality rates (thus lower life expectancy). They are also less literate, but probably more economically active. 3/ In fact if nomads have higher fertility than the settled population, which they may well have, even this estimate could be low.  - 53 - Table II: Population Estimates 1975-2000 (millions) Nomadic Population Projection Projection Projection High a/ Low I II III Estimate Estimate 1975 12.0 12.0 12.0 1.l 1.4 1980 13.4 13.4 13.5 1.5 1.5 1985 15.0 15.2 15.5 1.7 1.6 1990 16.0 17.5 18.0 2.0 1.8 1995 19.2 20.3 21.1 2.4 1.9 2000 21.7 23.5 24.9 2.8 2.1 2005 24.5 27.5 29.5 3.3 2.3 2010 27.7 32.4 35.2 3.9 2.5 2015 31.3 38.6 42.3 4.7 2.7 2020 35.4 46.2 51.3 5.7 3.0 2025 39.9 55.6 62.4 6.9 3.3 TNote: a/ If nomads in fact have higher fertility than the settled population, this estimate could still be low. Source: IBRD estimates based on ISPC and ADS data. 0Oi the basis of the preceding analysis, by the year 2025, the population of Afghanistan could range between 42.9 and 69.3 million depending upon assumptions concerning the rate of expansion of health services, and the rate of nomad-settled population migration.L/ If one were to choose as the most likely outcome by the year 2025 Projection II and the average of the two nonadic pcpulations the resulting total population would be about 60.7 million. In the less-distant future, by 1/ Again, no improvement in the use of family guidance services is assumed. To that extent, estimates are high.  - 54 - the year 2000, a population of 26 million, or almost double that of 1975, will have resulted. While the population estimates for the year 2000 are in themselves not particularly alarming, the underlying rate of population growth is. Under Projection II, by the year 2000 the population has reached a rate of population growth of 3.0%, which implies a doubling of the population every 23 years (under Projection III the rate of growth has reached 3.3%, implying a doubling every 21 years). As a consequence, over the next 25 years, the population stands to almost triple under either Projection II or III. Two primary causes can be identified: the changing age structure of the population (which brings ever-increasing numbers of women into the fertile age group), and declining mortality (by far the stronger force). However, population growth will not continue unchecked - the use of family guidance services for the purposes of spacing births and limiting family size has already started and will undoubtedly accelerate in the future as the availability of health services increases and the knowledge of such family guidance services spreads. In further investigating this point, it is important to realize that achieving a stationary population under conditions of declining mortality is contingent upon declining fertility. But how rapidly will fertility decrease? This is of course extremely difficult to predict and depends upon many interacting forces (cf. Vol. I, Ch. III). Under the circumstances, it may be easier to turn the question around and ask how rapidly fertility must decline for the preconditions of a stationary population to be met, and what the long-run population size implied by this fertility decline taken in conjunction with mortality improvement are. From such an exercise, fertility goals could be identified, and policy measures taken to reach and maintain them if general social and economic forces proved inadequate.  - 55 - For illustration, three rates of fertility decline have been identified which correspond to reaching the necessary pre-conditions for a 1/ stationary population by the years 2030, 2050, and 2070. These are henceforth referred to as fast, moderate, and slow fertility declines, respectively. Each of these three assumptions is applied to life expectancy 2/ projections II and III above. Table III: Long-Term Population Estimates under Different Rates of Fertility Decline (1980-2070, in millions) M4ortality Assumption II Mortality Assumption III Rate of Fertility|(l) Fast (2) Medium (3) Slow| j(4) !ast (5' TMedium (6) Slow Decline I _ 1980 13.18 13.20 13.20 13.27 13.28 13.29 1