Skip to content
Futurisca A computational laboratory for the future
Reference

The model, block by block

Seven coupled blocks, stepped a year at a time, over a few hundred futures. Everything on this page is what the engine actually computes, in the order it computes it.

The step

A year, a month, a week, a day

The engine solves a year at a time. The finer units do not run a second simulation; they read the same answer arriving, each variable at its own pace, so a price moves before output does and output before jobs. The ruler opens from years to days and the readings land in the order they do in the engine.

stepThe ruler of the run, opening to finer steps. Readings land in the order the engine produces them.

What happens in one year

  1. Levers are applied

    Every lever active in this year is turned into effects on the countries it names. A lever with a start year and a duration phases in and out rather than switching, because policy does.

  2. Markets clear on price

    Energy, food and semiconductors each have supply, demand and a price. Supply answers the price on a lag, and a fab takes four years, which is the whole reason a chip shock hurts rather than clearing overnight.

  3. Trade moves through the network

    Bilateral intensity from partner size over distance, adjusted for agreements and frictions. Tariffs, export controls, sanctions and freight all bite through it, and governments answer in kind with a probability rather than automatically.

  4. Output is built as a sum

    Trend, technology, rates, fiscal stance, external demand, energy and food, input constraints, conflict, the global cycle, confidence and the unmodelled shock. They add up to the growth number they explain, every time.

  5. Prices and employment follow

    A Phillips curve for inflation, a Taylor rule for the policy rate, a currency that mean reverts, and Okun’s law for unemployment with an automation term.

  6. The state pays for it

    Primary balance, debt dynamics, a risk premium that rises with debt and sanctions, and a financing crisis hazard that can fire in any year.

  7. People move

    Births, ageing, migration driven by income gaps and stress, political stability, tension between blocs, and conflict hazard that rises with tension and falls where nuclear weapons are involved.

Why the causes page can be trusted

Growth, inflation and unemployment are not solved and then decomposed afterwards. They are built as a sum of named terms, which means the bars you click are the arithmetic the engine actually ran. They add up to the number they explain, every time, and each one opens into what drove it.

Why the comparison can be trusted

Random draws are addressed by scenario seed, future number, year, country and channel through a hash rather than pulled from a sequence. Two scenarios run on the same seed meet the same luck in the same years, so the difference between them is your lever and not the noise. That is what makes an A and B comparison meaningful at a few hundred futures instead of needing tens of thousands.

What it does not contain

No financial sector

Banking, credit and asset markets enter only through a risk premium and a debt crisis hazard. There is no balance sheet anywhere in this model.

No politics

It models capacity, trade, prices, energy, demography and alignment. It does not model what one leader decides on one afternoon, and it never claims to.

No sub-national economies

The map goes down to the county. The model runs at country level, and a selected county is a place rather than an economy of its own. The panel says so.

No transfer mechanism

A development transfer is modelled on the side that pays for it and not credited to the receiver, because there is no mechanism between two balance sheets. The lever says this rather than implying otherwise.

The limits page goes further than this

Every assumption, prior and known weakness, written down.

Method and limits
Reference 06 / 10