How much is being borrowed, and what can a model add?

The OBR emulator — the OBR macroeconometric emulator — is the model closest to this topic, and the first thing to say about it is what it is not. Its question types in the registry are economic_shock, translated_policy_scenario: forecast is not among them. It answers "what would this shock do to the March 2026 EFO baseline", never "what will borrowing be".

• Outputs: gdp, consumption, investment. Neither number above is one of them.
• It cannot answer arbitrary statutory reform incidence and borrowing through the current adapter. The second of those is decisive here: the emulator cannot report borrowing at all.
• Status: validated for selected scenarios. Uncertainty: not comprehensive.
• Baseline: March 2026 EFO baseline, so every scenario is a deviation from that vintage, not from today's outturn.

Predictive validation: weak. Free-running GDP and consumption MAPE are 5.75% and 9.56%; the anchored fit is by construction. Policy counterfactuals: weak. One income-tax costing is independently compared with HMRC; trade, labour, prices and parts of household income remain constrained. Read the validation page →

So the borrowing and debt figures above are outturns with no model path beside them. What the emulator adds is the counterfactual: run a spending or corporation-tax change through it and read the GDP, consumption and investment response — then take the fiscal arithmetic from tax and benefit reform, which is where the costing actually lives.

Public-sector finance series are revised heavily and often; the snapshot column is the difference between a reproducible number and a moving one.

The snapshot files are append-only and never edited, so a number published here can be reproduced against the data as it stood — browse the store, its release calendar, and the as-of recipe →