Baseline methodology

Pressure separates baseline calibration from scenario simulation.

Cause and effect, not forecast

PolicyLever is an economic machine, but it is deliberately neutral about magnitudes. We are not predicting GDP, inflation or the future. We are showing the structural cause and effect of a decision: which way the money flows, who receives it, and which flows repeat every period.

Every pound has a path. A penny and a billion travel the same pipes. Direction and pressure are the truth; amounts are illustrative.

Pipe thickness and flow intensity scale with the size of the decision relative to the actor involved — a 5% Bank Rate move sends a heavier flow from mortgage holders to banks than a 0.25% move. The number on the pipe is illustrative; the heaviness of the pipe is the message.

Four layers

Baseline
The starting economic map of sector balance sheets and gauges.
Scenario
The policy decision or shock applied to that baseline.
Flow model
The ledger-backed movement of money between actors. Cause and effect, not prediction.
Gauge model
Non-cash pressure such as inflation, FX, market-value stress and counterfactual drag.

Current baseline

Reference UK Economy is currently stylised and manually calibrated. It is designed to make policy flows visible. It is not an official forecast or national accounts model.

Future versions may be updated from ONS, Bank of England, OBR and DMO sources through an admin-approved monthly process.

Honest disclosure

Current baseline: stylised / manual.

Future baseline: admin-approved official-data calibration.

Official data and stylised assumptions

Pressure separates baseline calibration from scenario simulation. Some values can be pulled from official sources. Others must be derived or stylised because the economy cannot be reduced honestly to a single table.

The rule is: Official where possible. Derived where transparent. Stylised where necessary. Always labelled.

Current proof: Bank of England Bank Rate is the first official-source pull.

Current simulator baseline: still stylised/manual unless a reviewed baseline is explicitly published.

How PolicyLever separates economic effects

Every visible effect is classified so you can see what is a real cash movement, what is a market-value change, and what is a pressure signal.

LEDGER
A cash or balance-sheet booking with counterparties. Both sides balance — money changes hands.
VALUATION
A market-value effect on an asset someone already owns. No cash today; someone feels richer or poorer.
GAUGE
A pressure, confidence, risk or capacity signal. Not a £ flow — a reading on the state of the system.
ASSET_SWAP
Money moves one way and an asset moves the other. The books balance at the moment of the swap.

Channel ownership

Each economic channel has a primary lever that owns its effect, plus optional secondary modifiers. Levers do not silently overlap — moving one lever does not book pressure that belongs to another.

  • Bank Rate owns existing-debt refinancing pressure, mortgage and deposit interest, and reserves/APF operating cost.
  • Gilt issuance owns new-issuance rollover cost (priced at the effective borrowing rate) and its share of market-absorption pressure.
  • Energy shock owns household and company energy bills, energy imports and energy CPI pressure.
  • FX shock owns sterling/FX effects — pass-through to imports, export competitiveness, revaluation and risk gauges.
  • QE/QT owns the gilt asset-swap legs, reserve-abundance conditions and gilt-price valuation pressure.

Government borrowing costs

Government debt-service is split into two distinct channels so Bank Rate and gilt issuance never double-book the same pressure.

  • Price side (Bank Rate / yields) — higher rates raise the cost of refinancing existing debt over time, as gilts mature and roll over. This is a gradual rollover effect, not an instant repricing of all government debt.
  • Quantity side (gilt issuance) — new gilts add future debt-service cost priced at the effective borrowing rate (Bank Rate plus term-premium tilt plus credit spread), not a flat coefficient.
  • Gilt valuation — market-value moves on existing gilts are shown as VALUATION on holders' balance sheets, not as government cash costs.
  • Market absorption / credibility — directional financing-cost pressure from issuance, QT or QE is shown as a GAUGE, never as a cash transaction.

QE / QT

Quantitative Easing and Tightening are asset swaps, not ordinary spending or taxation.

  • QE — the central bank credits reserves / pays cash to holders and receives gilts in return.
  • QT — the central bank receives reserves / cash and returns gilts to holders.
  • Each slice is one balanced cross-actor transaction. The pipes show two paired flows: a money leg and an asset leg.
  • Yield and gilt-price effects are shown separately on the valuation and gauge channels — they are not double-booked against the swap.

Why the £1 panel may not change when the scenario gets bigger

Two different questions are answered in two different places.

  • “Where did the £1 go?” shows the composition of pressure — the share of each pence going to households, banks, government and so on, rebased to 100p.
  • “Scenario pressure vs baseline” shows the magnitude — how big the total move is.
  • If only one lever moves, the composition can stay broadly the same while total pressure rises or falls. That is the expected behaviour, not a bug.

Back to the simulator.