Consolidated Projections (Base: USD 60/barrel)
| Indicator | Today | Year 5 | Year 10 | Year 15 |
|---|---|---|---|---|
| Production | ~1M bpd | 1.5β2M | 2β2.5M | 2.5β3M |
| GDP | USD 83B | 120β160B | 200β300B | 250β350B |
| Sovereign Fund | USD 0 | 15β25B | 70β120B | 160β220B |
| Dividend/person | USD 0 | 5β10 | 15β25 | 22β30 |
| Fund return (5.5%) | USD 0 | 825β1,375M | 3,850β6,600M | 8,800β12,100M |
Calculation: fund x 5.5% return x 10% payout / 40M people. See stress test below for the explicit model. At USD 70-80/barrel, figures multiply β see The Dream.
The fund would grow to USD 500B+ and dividends would double. At USD 60 the plan works. Anything above is a bonus.
Sensitivity Analysis: Stress Test by Oil Priceβ
The plan uses USD 60/barrel as a conservative base (EIA STEO, Mar. 2026). But price is the most volatile variable. This table shows the impact under each scenario.
Model Assumptionsβ
Year 5 Production: 1.75M bpd (midpoint of 1.5β2M range)
Year 10 Production: 2.25M bpd (midpoint of 2β2.5M range)
Year 15 Production: 2.75M bpd (midpoint of 2.5β3M range)
Cost per barrel: USD 37.50 (average USD 35β40, heavy crude)
Fund contribution: 30% of net oil revenue
Fund return: 5.5% annual (conservative average of 4β7% range)
Dividend: 10% of fund net income to 40M people
Source for costs and returns: data-standards.md β barrel cost USD 35β40 (extraction + blending + transport + processing); fund return 4β7% annual (conservative vs. Norway 15.1% in 2025).
Explicit Calculation (Base Scenario example, Year 5)β
Gross revenue = 1,750,000 bpd Γ 365 days Γ USD 60 = USD 38,325M
Operating cost = 1,750,000 bpd Γ 365 days Γ USD 37.50 = USD 23,953M
Net revenue = USD 38,325M β USD 23,953M = USD 14,372M
Fund contribution = 30% Γ USD 14,372M = USD 4,312M/year
| Scenario | Price | Margin/barrel | Net Revenue Year 5 | Net Revenue Year 10 | Accumulated Fund Year 15 | Dividend/person/year |
|---|---|---|---|---|---|---|
| Crisis | USD 40 | USD 2.50 | USD 1,597M | USD 2,054M | USD 15,000β20,000M | USD 2β5 |
| Low | USD 50 | USD 12.50 | USD 7,984M | USD 10,269M | USD 80,000β110,000M | USD 11β15 |
| Base | USD 60 | USD 22.50 | USD 14,372M | USD 18,484M | USD 160,000β220,000M | USD 22β30 |
| Favorable | USD 70 | USD 32.50 | USD 20,759M | USD 26,699M | USD 250,000β340,000M | USD 34β47 |
| Boom | USD 80 | USD 42.50 | USD 27,147M | USD 34,914M | USD 350,000β470,000M | USD 48β65 |
At USD 40/barrel, the net margin collapses to USD 2.50/barrel β barely covering operating costs. The sovereign fund grows minimally and citizen dividends are symbolic. At this price:
- No capital for tech infrastructure investment
- Debt restructuring becomes unsustainable
- Stabilization fund + price floor in forward contracts required
Mitigation: Forward contracts include a guaranteed minimum price (floor) of USD 55. If Brent falls below USD 50 for more than 2 quarters, the stabilization fund activates.
At USD 70+, each additional dollar generates ~USD 639M/year in extra net revenue (at 1.75M bpd). At USD 80, the fund reaches levels that enable:
- Citizen dividends of USD 48β65/person/year (family of 4: USD 192β260)
- 100% financing of tech infrastructure with no additional debt
- Reaching Norway's population-adjusted fund level in ~20 years instead of 30
Every USD 1 above USD 60 is direct upside to the sovereign fund.
Methodological Noteβ
The accumulated fund at Year 15 includes: (1) annual contributions of 30% of net oil revenue over 15 years, with production growing per Rystad Energy; (2) compound return at 5.5% annually; (3) does NOT include diversification revenue (tech, tourism, agro) which is added separately. Ranges reflect variations in the production ramp-up pace.