Using Portuguese Sources to Build an Angolan Social Table

Angola is the first Portuguese colonial case prepared for AFLIT’s social-table series, introducing a Lusophone settler economy into a comparison that previously focused on British and French Africa (Aboagye & Bolt, 2021; Alfani & Tadei, 2019; Bigsten, 1986; Bolt & Hillbom, 2016; De Haas, 2022; Hillbom et al. 2024). The effort raises questions about which relevant statistics can be retrieved from Portuguese sources, what they enable us to measure, and how to go about constructing an Angolan social table.

The source corpus

In the sources, the corpus includes five series. The first comprises the general censuses of 1940, 1950, 1960, and 1970. The others are the Anuário Estatístico de Angola (Statistical Yearbook of Angola), the Orçamento Geral da Província de Angola (General Budget of the Province of Angola), the trade publications later grouped under Comércio Externo (Foreign Trade), and the imperial Anuário Estatístico do Ultramar (Statistical Yearbook of the Overseas Territories). Together, they provide population counts, labour categories, wages, prices, production, trade, and public finance data. These form the basis for constructing a social table.

What the censuses provide

The 1940 census, published in several volumes, records population details by race, sex, age, religion, schooling, nationality, ethnic group, household structure, and profession. Volume VII is particularly important for social-table analysis. It categorises settlers, assimilados (assimilated persons), and mestiços (mixed-race persons) by profession, sex, race, and their position within the profession, providing an occupational map of the urban and formal-sector economy. The occupational tables conclude with this population. The rural African majority, classified in the census under the Indigenato (Indigenate system), appears in other volumes mainly by age, sex, religion, ethnic group, and locality.

The 1950 census maintains much of this structure and links 1940 to 1960. The 1960 census adopts a different classification. Its labour volume categorises the resident population by relation to work, major occupational group, branch of activity, occupation, and status in the profession, using ILO and UN nomenclatures. This makes 1960 a benchmark because its categories are closer to the class structures used in AFLIT. The 1970 census still reports some racial categories for population totals, while its occupational classification follows the ILO nomenclature introduced in 1960.

Income, prices, and the yearbooks

The censuses identify population groups. The yearbooks help assign incomes to them. The Anuário Estatístico de Angola (Statistical Yearbook of Angola) covers labour and salaries, agriculture, livestock, fisheries, mining, manufacturing, transport, trade, prices, banking, public finance, education, health, and justice. This matters for social tables because the series links population categories to crop output, production values, wages, prices, and fiscal aggregates, often by district. The rural sections are important because the African majority cannot be directly read from the 1940 occupational tables. Subsistence must therefore be measured rather than ignored. For Uganda, De Haas (2022) shows that excluding non-monetised self-provisioning raises the Gini coefficient by nearly 0.2 points. Angola, with its smallholder majority, is unlikely to be an exception.

Budgets and trade

The budget and trade series include wage schedules and export structures. The Orçamento Geral da Província de Angola (General Budget of the Province of Angola) details offices, departments, localities, and salary schedules, enabling the reconstruction of public-sector wages. The trade publications, later consolidated as Comércio Externo (Foreign Trade), identify commodity booms, export specialisation, and the geography of commercial activity. These materials link social groups to the export economy rather than to wage labour alone. The imperial Anuário Estatístico do Ultramar (Statistical Yearbook of the Overseas Territories) provides cross-colony comparisons with Mozambique, Cabo Verde, São Tomé, and Guiné, and fills gaps in the Angola series.

What the sources support

The source base supports benchmark years anchored in 1940, 1960, and eventually 1970, with 1950 serving as a bridge. The 1940 census provides occupational details for the urban and formal-sector economy. The 1960 census offers more standardised categories. The 1970 census captures the late colonial economy after the abolition of the Indigenato (Indigenate system). The yearbooks, budgets, and trade series can then be utilised to estimate incomes for the groups identified in the censuses.

The same materials enable regional decomposition. The censuses tabulate populations at district, concelho (municipality), and sometimes circunscrição (administrative circumscription) levels. The yearbooks report output by district, and the budgets identify public-sector posts. Angola can therefore generate both a colony-level table and district comparisons between Luanda, the northern coffee districts, cotton zones, mining regions, and the pastoral south.

Obstacles

The main challenges involve coverage and classification. The Indigenato (Indigenate system) division means Portuguese statistics collected different data for urban and rural populations, so the African rural majority must be reconstructed indirectly. Coerced and semi-coerced labour also needs separate handling. The contratados (contract workers) cannot be combined into an agricultural wage class without losing the structure of the colonial labour regime. Much of the collection is image-only and requires OCR and cleaning. The categories used in 1940, 1950, 1960, and 1970 need a documented crosswalk. The table will also require a pricing and currency strategy that transitions between réis, escudos, and angolares while valuing subsistence at Angolan rather than metropolitan prices.

The way forward

The 1897/98 budget has been entered into a spreadsheet with over four hundred occupation-locality-wage entries, showing the detail that the Orçamento Geral (General Budget) can provide across benchmark years. The next stage will produce a social table for 1960, where the ILO-aligned categories align most clearly with existing AFLIT cases. The work will then extend back to 1940 and 1950 using a documented crosswalk and move forward to 1970. Simultaneously, the OCR and cleaning pipeline for the statistical yearbooks and Comércio Externo (Foreign Trade) volumes are being developed. A rural price series is also being compiled from the yearbooks for subsistence valuation and Inequality Extraction Ratio calculations.

These tables link Angola to AFLIT’s comparative questions. Did the rise in inequality documented by Hillbom et al. (2024) persist in a Lusophone settler and tree-crop economy? What shifts occurred in the distribution around the 1961 abolition of the Indigenato (Indigenate system)? How did contrato (contract) labour, a later and more extensive coerced-labour regime than those used in French or British Africa, influence wages in a commercialising colonial economy? The Portuguese sources do not answer these questions alone. Instead, they serve as comparative evidence when Angola is examined alongside the existing AFLIT cases.

References

Aboagye, P. Y., & Bolt, J. (2021). Long-term trends in income inequality: winners and losers of economic change in Ghana, 1891-1960. Explorations in Economic History, vol. 82, article 101405

Alfani, G., & Tadei, F. (2019). Income inequality in French West Africa: building social tables for pre-independence Senegal and Ivory Coast. UB School of Economics Working Papers, no. 2019/396, University of Barcelona School of Economics

Bigsten, A. (1986). Welfare and economic growth in Kenya, 1914-76. World Development, vol. 14, no. 9, pp.1151-1160

Bolt, J., & Hillbom, E. (2016). Long-term trends in economic inequality: lessons from colonial Botswana, 1921-74. The Economic History Review, vol. 69, no. 4, pp.1255-1284

De Haas, M. (2022). Reconstructing income inequality in a colonial cash crop economy: five social tables for Uganda, 1925-1965. European Review of Economic History, vol. 26, no. 2, pp.255-283Hillbom, E., Bolt, J., De Haas, M., & Tadei, F. (2024). Income inequality and export-oriented commercialization in colonial Africa: Evidence from six countries. The Economic History Review, vol. 77, no. 3, pp.975-1004