National Accounting Statistics for the Less Developed Economies (with Special Reference to Transkei)¶
Iraj Abedian (1982). National Accounting Statistics for the Less Developed Economies (with Special Reference to Transkei). MA (Economics) thesis, University of Cape Town. University of Cape Town.
| Type | thesis |
| Scope | Economics and economic history of apartheid |
| Method | mixed |
| Period studied | 1954–1980 |
| Regions | Transkei, South Africa |
| Themes | bantustans removals influx, mining migrant labour, methods data sources, manufacturing industrial policy, macro fiscal finance debt, inequality poverty unemployment, agriculture land, comparative theory |
| Citation key | abedian1982 |
Argument¶
The thesis reviews the theory of national accounts and the UN SNA and argues that the SNA suits developed monetised economies and needs modification for less developed, dualistic economies. Part II applies a modified SNA to Transkei: it criticises the existing BEPA, Department of Statistics and BENSO estimates, builds primary-survey estimates of GDP, GNP and national income for 1980, and constructs time series for 1954-80. Transkei's economy is shown to be heavily dependent on migrant earnings and on South African transfers, with very low per capita output. Aggregate measures are said to conceal regional discrepancies and to risk legitimising discriminatory treatment of the homelands.
Evidence and method¶
Primary surveys carried out in 1981: a business survey (over 1,000 enterprises interviewed, 870 usable), full surveys of government, public corporations and big manufacturers, and district agricultural officers' questionnaires. Also Auditor-General revenue and expenditure records, Transkei government departments, labour recruiting agencies, and earlier BEPA, Department of Statistics and BENSO estimates.
Place in the debates¶
- Reliability and meaning of homeland national accounts (BEPA, Department of Statistics, BENSO) (see Data reliability and econometric method). Abedian rejects the 'Black' versus 'non-Black' split in the official estimates as having no theoretical or practical justification, calls BENSO's figures inconsistent, and finds the Transkei government's revisions overstated. He builds his own estimates, with a 4-5% stated margin of error.
- Whether UN SNA aggregates are appropriate for less developed or dualistic economies (see Data reliability and econometric method). He holds that the SNA is unsuited to LDCs but flexible enough to be adapted, and that a modified SNA remains the only workable framework.
- Homeland development and decentralisation policy (see Homelands, separate development, decentralisation and removals). He judges decentralised industry capital-intensive and subsidy-dependent, so it neither reduced migrant influx nor generated local income. Transkei is described as dependent on South Africa, not as an independent economy.
Findings¶
-
The GDP and GNP of Transkei diverge widely because of migrant remittances; the gap widened over 1970-77 in the BENSO series. (p. 79)
The GDP as a percentage of the GNP has declined from 42. 04~~ in 1970 to 34. 98~~ in 1977
-
Official Department of Statistics depreciation estimates are implausibly low, which inflates national income. (p. 82)
the depreciation - GNP(GDP) ratio is, usually .and for developed .~ economies, approximately 10 per cent.
-
Estimates of migrant numbers are mutually inconsistent across sources; the 1975 figure differs by 19%. (p. 80)
the number of migrants in 1975 was 326 800 whereas in the second survey for the same year the number was 265 000, i.e. 19% discrepancy!!
-
The 1980 GDP of Transkei is R574.5m (the text itself gives 547.5m in places), with about 69% from the market economy and 31% from subsistence. (p. 123)
The GDP of Transkei, in 1980-1981, thus consists of ~1% subsistence economy, and 69% market economy.
-
A large share of Transkei's 1980 GDP was externally originated through South African transfers. (p. 124)
241/547=44% of the GDP of Transkei in 1980 was externally originated.
-
Migrant earnings dominate GNP: the 1980 GNP at factor cost is about double the GDP. (p. 127)
547 503 +581 530 . -13 769 =1 115 264 49.1~~
-
Per capita GDP in Transkei is a fraction of South Africa's and a very low absolute level. (p. 182)
the GDP per capita in Transkei was 20%.of that of South Africa in 1980.
-
Industrial decentralisation produced capital-intensive branch plants that failed to reduce migrant influx or generate local income. (p. 177)
these industries, by very definition of decentralization, have only changed location and not technology.
-
Real GNP and real GNP per capita (BENSO series, adjusted) fell after the mid-1970s, in contrast to rising nominal figures. (p. 165)
The decline shown in the real GNP implies a deterioration of the overall economic performance.
-
Transkei's government revenue depends heavily on erratic South African transfers; budgetary assistance has no formula and is set by bargaining. (p. 159)
the size of the budgetary assistance is erratic and unforeseeable.
-
Migrant labour weakens agricultural incentives, especially for women, because remittances follow needs. (p. 170)
the incentive for women to succeed is dramatically reduced by the poor rwards offered them
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| BEPA 1968 Interim Report on the Use of the National Accounts as a Basis of a Development Programme for the Economy of the Transkei | official statistical series | 1954/55 and 1959/60-1966/67 | Supplies GDP by industry for 1954-1966/67, used to build the 1954-80 GDP time series and the public sector share. |
| Department of Statistics, National Accounts of the Bantu Homelands 1969-70 to 1973-74 (Report No. 09-17-01) and National Accounts of the Republic of Transkei 1971-1975 | official statistical series | 1969/70-1975/76 | Gives GDP by sector, public sector value added, public fixed investment, expenditure by purpose, and depreciation for 1969-75; used in the time series and as an object of critique. |
| Department of Statistics, Draft Report: Gross Domestic Product of Transkei 1971-1974 | official statistical series | 1971-1974 (published 1976) | Cited for the treatment of sorghum beer in GDP and for the non-market services remark. |
| BENSO Statistical Survey of Black Development 1979 and 1980 (and The National Economy of Transkei 1976) | official statistical series | 1970-1977 (with data to 1980) | Source of GDP/GNP 1970-77, migrant numbers and earnings, commuters and population series; the author critiques it as inconsistent and uses it with revision. |
| Transkei Government Estimates ("This is Transkei", 1978) | official statistical series | 1959-1973 | Alternative GDP estimates compared with BEPA/Dept of Statistics/BENSO and rejected as about 15% overestimated. |
| Tomlinson Commission estimates of Bantu regions' domestic production | commission or official report | 1950-51 | Mentioned as the pioneering estimates but judged unusable because they excluded white-populated areas. |
| Transkei Government Auditor-General's Office expenditure and revenue records 1980/81 | company records or reports | 1979/80-1980/81 | Source of government salaries, expenditure categories, taxes and transfers used to build public administration value added and government accounts. |
| 1981 Business (and Employment/Expenditure) Survey of Transkei | author survey or fieldwork | 1981 (for 1980/81) | Provides sampled value added, profit, foreign payments, depreciation and investment for manufacturing, trade, transport, finance and services sectors. |
| Department of Commerce, Industry and Tourism 1979 business census / business licence records | other | 1979-1980 | Used as bench-mark for the total number of establishments to gross up sample value added (with 25-30% adjustment). |
| District Agriculture Officers' crop estimates and questionnaire on communal and homestead production | author survey or fieldwork | 1980/81 | Basis for estimating field, communal garden and homestead crop output in the subsistence sector, and the household count of 485 997. |
| 1980 Annual Report of the Veterinary Section, Dept. of Agriculture and Forestry, Transkei | commission or official report | 1980/81 | Livestock numbers, births and exports used to calculate offtake and meat value added. |
| Agriculture Products Brokers (APB) sales data for hides, skins, wool and mohair | company records or reports | 1979/80-1980/81 | Gives quantities and farmer income for hides, skins, wool, with a 30% adjustment for unrecorded sales. |
| Transkei Meat Industries livestock prices and TRACOR commercial milk data | company records or reports | 1980/81 | Prices for valuing livestock offtake and commercial milk output. |
| Development Strategy for the Republic of Transkei (1980-2000) | commission or official report | 1979/1980-2000 | Source of value-added-to-gross-output ratios for crops, irrigation, tea and sugar. |
| Transkei Development Review 1981, Vol. 1, No. 2 (Department of Interior labour survey) | official statistical series | 1969-1980 | Number of migrants and their earnings used for 1980 foreign factor receipts and the migrant time series. |
| Labour recruiting organisations in Umtata (VEZAMFA, ACRO-HANISE) wage and remittance information | chamber of mines or recruiting records | 1979-1980 | Supplies average migrant wage, length of employment and remittance ratios (55-60%) for contract workers. |
| Pilot survey "Household Expenditure Income of Transkei" | household or labour survey microdata | December 1981 - January 1982 | Supports the wage rates suggested by recruiting agencies but its remittance ratios are judged inapplicable to all migrants. |
| Swidinsky, Labour Resources in the Economic Development of Transkei (1981) / BMR 1979 survey | secondary literature compilation | 1979 | Cited remittance ratios (25.9% of cash earnings) contrasted with recruiting agency figures. |
| Department of Statistics population estimates and 1970 census; population projections | population census | 1970-1980 | Used to construct high, low and adjusted population series for per capita GDP and GNP. |
| South African Consumer Price Index | official statistical series | 1965-1980 | Used as the deflator (1970=100) to compute real GDP and GNP for Transkei. |
| Reports and estimates of the Transkei Department of Local Government, municipalities, TRESCOR and TDC 1980 Annual Report | company records or reports | 1977-1980 | Wages, interest and factor payments for electricity, water, local authorities and TDC value added. |
| Rosmarin ELS & Taylor town planning data on erfs and rents | other | c. 1980 | Number of houses and average rent used to impute urban home-ownership value added. |
| United Nations System of National Accounts 1953 and 1968; UN document ST/ESA/STAT 77 (1975) | commission or official report | 1953-1975 | Methodological standard for classification, dualism treatment and own-account construction valuation. |
| Reserve Bank / Department of Public Finance estimates of Rand circulation; Customs Union Contract; South African budget estimates | legislation gazette | 1974-1981 | Used to analyse South African transfers (Rand compensation, customs share, budgetary and project aid) and compare to provinces. |
New data¶
Author constructed 1980/81 national accounts (GDP, GNP, national income, capital formation) for Transkei from a 1981 business survey of 870 usable enterprises across 28 districts, government accounts and district agricultural questionnaires, and combined them with earlier estimates into a 1954-80 GDP series, plus GNP, real series and domestic disposable income for 1970-80.
Inference¶
Quantification: Descriptive tables.
National accounting reconstruction by the output (value-added) method, mixing sampled business-survey grossing-up, government accounts, imputation and interpolation, then time-series comparison of nominal and real aggregates (CPI-deflated) and shares; interpretation is descriptive and theoretical, with a critical comparison of competing official estimates.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Transkei's 1980/81 GDP was about R547.5m (subsistence about 31%, market 69%), with GNP of about R1,115m including migrant earnings of R574m. (p. 116) | Sector-by-sector value-added estimates from the business survey, Auditor-General accounts and agricultural questionnaires (Tables 38, 39, 41). | Partial data |
| Roughly 44% of 1980 GDP was externally originated, since government budget items (including SA budgetary assistance, Rand compensation and customs) fund public administration, health, education and part of other sectors. (p. 117) | Auditor-General revenue accounts (Table 30) compared with GDP components. | Partial data |
| Existing estimates by BEPA, Department of Statistics and BENSO are conceptually flawed and inconsistent (racial distinction, underestimated depreciation, conflicting population and migrant figures). (p. 75) | Comparison of BENSO Statistical Surveys 1979/1980, Department of Statistics reports and Transkei government figures (depreciation ratios, population columns, migrant counts). | Systematic data |
| Migrant earnings and numbers are highly uncertain (R402m versus R574m, a 43% discrepancy), making the GNP estimate fragile. (p. 125) | Department of Interior of Transkei labour survey versus recruiting agency data and BENSO (Tables 9, 55, 56). | Systematic data |
| Real GNP and real GNP per capita of Transkei declined after 1977, and per capita GDP in 1980 (R220) was a small fraction of South Africa's. (p. 140) | Nominal series deflated by South African CPI with interpolated 1978-79 values and adjusted population (Tables 52-58). | Partial data |
| Manufacturing is marginal and largely a product of South African decentralisation policy, dependent on subsidies, with 93.5% of industries decentralised in 1978. (p. 167) | Transkei GDP sector series and incentive details from Jackson (TDR); no establishment-level data shown beyond series. | Illustrative or anecdotal |
| South African transfers to Transkei are erratic and arbitrary and low relative to provincial administrations (per capita R48 vs R108-148 in 1980). (p. 151) | Budget and Statistical Newsletter figures compared across provinces and Transkei (Tables 60-64). | Partial data |
The author's caveats¶
The author states that the Transkei lacks statistics, that agricultural data from extension officers are miscalculated, that business survey answers were underestimated (adjusted, margin of error 5-7%), that depreciation and capital formation are not rigorously estimated, that migrant data are unverified, and that overall error is 4-5%. The 1978-79 GNP values rest on interpolation. Subsistence estimates rely on assumptions (hut life, household production averages).
Gaps in the evidence¶
Many key figures rest on assumed coefficients (profit ratios, 40% construction value added, per-household output, milk and egg yields) with no independent validation, and the 1980 estimate cannot be benchmarked against a census or household expenditure survey. Credible migrant earnings and remittance data (e.g. Chamber of Mines/recruiting agency or TEBA records, and a proper household income survey) would test the GNP and domestic disposable income claims; the 1980 population census and later surveys exist now. The explanatory claims about decentralisation, agricultural stagnation and dependence are largely asserted from secondary sources rather than data.
Replicability: Data described but not public.