Zwölf stilisierte Wirtschaftsräume mit synthetischer Handelsmatrix. Wählen Sie Koalition, Zielregion und Sanktionsumfang je Sektor, stellen Sie Substitution und Umgehung über Drittländer ein und lassen Sie den Schock rundenweise durchs Netz laufen. Achten Sie auf das Verhältnis von Wirkung im Ziel zu Kosten der Koalition.
Twelve stylised economic regions with a synthetic trade matrix. Choose coalition, target and sanction scope per sector, set substitution and third-country leakage, and let the shock propagate round by round. Note the ratio of impact on the target to cost for the coalition.
12 stylised regions · seeded synthetic trade matrix · direct cuts, substitution, leakage, secondary sanctions and input-output spill-overs, round by round
The world is reduced to twelve stylised regions with fixed GDP weights (world GDP = 1,000 synthetic units). A seeded random generator (mulberry32) builds a 12 × 12 trade matrix — exports from region i to region j as a share of i's GDP — from a gravity rule (partner size × proximity × noise), scaled to a per-region openness, plus a four-sector split of every link (energy, intermediates, machinery, consumer goods) derived from a stylised exporter profile. None of this is real trade data; the matrix merely has the texture of one.
Each round the model does four things. Direct cut: on every link between the coalition and the target, the sanctioned share of each sector
is removed; of that volume, a part is substituted (both sides find alternative partners, growing with the substitution elasticity as
σ·(1 − 0.6ʳ)), a part leaks through non-coalition conduits (the leakage share, eroded each round by secondary-sanctions pressure as
λ·(1 − p·(1 − 0.6^(r−1)))), and the rest is blocked. Output loss: blocked exports reduce the exporter's GDP by their domestic
value-added share (0.7); blocked imports hit the importer through a sector-specific input shock (energy 0.5, intermediates 0.35, machinery 0.2,
consumer 0.05). Propagation: a region that loses output imports less, in proportion to its import intensity and a damping factor of 0.5; the
shortfall is distributed to its suppliers by their existing shares, input-output style, so losses — and gains — ripple outward one round at a time.
Gains: conduits earn a transit margin on leaked trade; non-coalition exporters pick up the substituted demand of the coalition and the target.
The effectiveness ratio divides the target's GDP loss by the coalition's GDP-weighted loss; the steady-state figure is the first round in which no
region moves by more than 0.02 percentage points.
Limitations are deliberate: a static, linear, single-period-per-round structure; no prices, exchange rates, inventories, financial channels or retaliation; and — most importantly — a synthetic matrix that is not calibrated to any published trade statistics. The point is the mechanics, not the magnitudes.
Second-order effects are where intuition fails and where models earn their keep: a cut that looks decisive on the first link can be diluted by leakage and substitution, or amplified back onto the coalition through its own suppliers. Mature organisations do not ask an AI system for a single number; they ask it to expose the chain of assumptions — elasticities, leakage, damping — and to show how the answer moves when each one does. The discipline of stating what the model does not contain, as prominently as what it does, is the difference between decision support and false precision.
Research and teaching demo. All regions, GDP weights, trade flows and sector shares are synthetic and stylised — not real trade statistics, not a forecast, and not policy or investment advice. Computed entirely in your browser; nothing is stored or transmitted. Scenario settings are kept in the URL so you can share a view.