OPEN ASSUMPTIONS / VERSION 0.1-demo
How the model works
This page is a transparent record of the simulation's simplified economic relationships. A change in this model version changes the meaning of results; it is not an economic forecast, official poverty projection, or government policy recommendation.
Inputs and results
| Element | Definition | Prototype rule |
|---|---|---|
| Oil scenario | $70 / $140 / $200 per barrel | Facilitator-fixed for ten years. Changes oil fiscal revenue, not real oil production directly. |
| Real interest rate | 2%, 4.5%, or 8% | Illustrative real financing cost on accumulated public debt. |
| Government decisions | Spending shares, sector focus, taxation, corridor acceleration | One policy package lasts two years. Social + investment ≤90% of primary spending. |
| Public spending envelope | 14.5% of current real GDP, plus 45% of oil windfall above $70 | An assumed primary-spending rule. Higher oil revenue not spent goes toward debt or deficits. |
| Natural assets | Oil and minerals depletable; water renewable | All measured as indexed stocks, not physical reserve estimates. Oil −3.5 points annually; minerals −0.5; water renews while used. |
| National roots | Electricity, logistics, skills, institutions, water | 65% of productive investment (after corridor cost), equally allocated; outcomes delayed 1–4 years. |
| Industry-specific roots | Know-how, equipment, networks in each economic branch | 35% of productive investment goes to the chosen industry and improves capabilities after two years. Experience also compounds gradually. |
| Lobito Corridor | Extra domestic links, agrilogistics and related capabilities | Costs at most 0.6% GDP for two years within the existing investment envelope; benefits after three years. Existing rail plans are the baseline. |
| Supply | Potential production in each sector | Annual growth trend + sensitivity to national-root stocks + sensitivity to industry-root stocks. Sector sensitivities are illustrative. |
| Demand | Domestic absorption and market access | A simple annual ceiling on realized non-oil sector growth responds to social spending, investment, tax effort and corridor access. |
| GDP | Sum of five branches' value added | Oil and mining are one branch but have separate internal production components; the oil shock affects only the oil component's fiscal receipts. |
| Real income per inhabitant | Index 2026=100 | Real GDP per capita proxy, not yet true GNI / PNB. Net foreign primary income will be added in a later layer. |
| Employment | People employed / working-age population | Sector jobs grow with production minus assumed productivity trend; demographics changes the denominator. |
| Poverty | Percentage of residents below a fixed threshold | Change in rate (percentage points) = −0.34×income-per-person growth (%) −0.27×change in employment rate (pp) −0.018×(social-spending share−35). Unvalidated illustrative elasticities. |
| Debt | Public debt outstanding | Previous debt + primary spending + interest − taxes and oil revenues; debt ratio computed relative to real GDP under fixed-price approximation. |
Demography: three age groups
Children (0–14), working age (15–64), and older people (65+) evolve each year. The model adds births based on fertility, moves cohorts as they age, and applies approximate age-specific mortality scaled by life expectancy.
Starting values: population 40.58 million, children 43.8%, working age 53.2%, older people 3.0%, fertility 4.9, life expectancy 65.2. The initial employment rate is 63.7%, used as a proxy for ages 15–64 even though the source survey uses ages 15+.
Higher social spending gradually affects life expectancy and fertility in this toy model; these effects are not empirically estimated.
Institutional quality, concretely
Administrative capacity is the fraction of expenditure that becomes usable, functioning assets on time. It might reflect transparent procurement, permits, reliable courts, customs, project execution and corruption leakage.
In the starter engine, this is one national-root index applied as an efficiency factor to newly scheduled capital investments. Later, you can model those six mechanisms independently.
Calibration and data limitations
Starting values combine some sourced macroeconomic benchmarks with demonstration assumptions. Do not treat a 2036 poverty value as a forecast. The poverty measure is a legacy monetary benchmark from 2018/19; no official 2026 poverty estimate is implied. GDP is initialized to an IMF 2026 nominal estimate then used as a fixed-value base unit to construct illustrative real paths. This is a simplification.
Annual oil prices of $140 and $200 are stress tests, not forecasts. Angola's measured population differs by source. Initial sector GDP shares, sector employment shares, input-output link sensitivities, natural-capital stock indices, project multipliers, all poverty elasticities, and project completion impacts are exploratory.