How does the Office for Budget Responsibility produce economic and fiscal forecasts?

What is the role of economic and fiscal forecasts in the budget?

Economic and fiscal forecasts are a critical part of the government budget process. They provide the government with useful information about the prospects for the economy and public finances, to help inform policy decisions. They are needed to judge the likely impact of the government’s plans and to assess its compliance with its fiscal rules. The independent Office for Budget Responsibility (OBR) is responsible for producing these forecasts which cover the next five years (separately, the OBR also produces longer-term assessments of the sustainability of the public finances).

The OBR must produce at least two forecasts each fiscal year (between April and March), but the timing of these is determined by the government and is usually aligned with government fiscal events. The government is required to give the OBR 10 weeks’ notice to prepare its economic and fiscal forecasts and the OBR will present several draft versions to the chancellor over those 10 weeks.

On some occasions, the period for the OBR to prepare its forecast has been shortened in response to events, such as during the coronavirus pandemic.

How does the OBR produce its economic and fiscal forecasts?

The OBR produces both an economic and a fiscal forecast, which interact and are developed iteratively. During this process, the government will provide details of any proposed policy changes so the OBR can incorporate these into its forecast and the OBR will commission analysis from government departments to feed into its forecasts.

The OBR engages with government, academics and others to develop and test its approach to producing forecasts. This includes publishing details of the models and assumptions it uses and much of the evidence for how they have made their judgements. 

What data and assumptions go into the forecasts?

The OBR creates an initial forecast for the economy which is largely developed using a macroeconomic model which is jointly owned with the Treasury. However, the OBR independently selects the data and assumptions which it uses within the model. Data and judgements which feed into the model include:

• Gross domestic product (GDP): The growth path of the economy depends on two key factors, how much spare capacity there is in the economy, and how quickly the economy’s capacity will grow (which depends on forecasts for elements such as population size and age structure, employment, the capital stock and productivity).  
• Population and employment rates: The OBR takes the Office for National Statistics' (ONS) population projections and makes assumptions about future age/sex-specific employment trends, for example factoring in rising employment at older ages as the state pension age increases.
• Migration: In recent years, the OBR has used the ONS’ regular migration projections to set its assumption for levels of migration in the medium term but has adjusted the ONS figures in the short-term to take account of the latest data and policy announcements to produce a more realistic profile for short-term changes in net migration.
• Productivity growth: The OBR makes a judgement about future productivity growth based on historical trends, current data and research. Productivity is one of the most challenging and uncertain metrics to forecast but is critically important to forecasting economic growth. OBR forecasts of productivity since its inception in 2010 have generally been overoptimistic.   
• Inflation: The OBR typically assumes that consumer price inflation will return to its 2% target by the end of the five year forecast period. In the short run, the OBR forecasts inflation by considering the likely path of prices in different sectors e.g. food, fuel, housing rents etc. The OBR produces forecasts for three key measures of inflation: the consumer prices index (CPI, the best measure of prices faced by households and used to uprate many benefit and tax rates and thresholds); the retail prices index (RPI, an outdated and flawed measure of prices faced by households but which is still used for some index-linked government bonds and some tax rates); the GDP deflator (which captures the price of goods and services produced in the UK and is used to convert nominal GDP figures to real terms).
• Borrowing costs: The OBR uses data from the market for government bonds to forecast future borrowing costs. As prices in markets can be volatile, the OBR takes average market prices over a two week period in the run-up to the fiscal event, and can adjust the results if they think they are unrepresentative. 

In addition to these components, the OBR must forecast many other, more granular components of the economy – that most other macro-forecasters would exclude – in order to provide the necessary inputs into models for tax revenues and spending. For example, the OBR has to forecast the volume of property transactions to forecast stamp duty land tax revenues.

Based on its initial economy forecast, the OBR will update its forecast of government spending and revenue.

• Tax revenues: The OBR commissions HMRC to update its forecast for revenues from each individual tax stream based on the economy forecast. The OBR can examine the assumptions included in these forecasts and challenge any that they feel are unrealistic.
• Welfare spending and other demand-led spending (annually managed expenditure, AME): The OBR commissions DWP to produce forecasts for individual components of welfare spending, much as it does HMRC for tax revenues. The OBR produces its own forecast for debt interest spending, based on the borrowing cost assumptions set out above. Forecasts for other components of AME are commissioned from other relevant government departments (for example, the Department for Education produces the forecast for spending on student loans).
• Other government spending: For departmental resource and capital spending, the OBR uses the government’s announced budgets for the next five years. Historically, the OBR has always accepted that the government will stay within its planned spending assumption, sometimes adjusting it to account for a small underspend. However, in 2024 the new Labour government accused its predecessor of not sharing the full extent of spending pressures with the OBR. It empowered the OBR to access more data from the Treasury to understand spending pressures and, if appropriate, to forecast that the government will overspend allocated budgets.

All these forecasts can then be adjusted based on the OBR’s assessment of the impact of new government policies. 

• New government policies: The government must provide an estimate of the cost or revenue of all new policies for the OBR to scrutinise. The OBR work with the Treasury and relevant departments to make any changes until they are happy with the costings provided (or the OBR agree to use a different costing to the government). If the government’s plans are insufficiently detailed or uncertain to support costings, the OBR will not include it in their forecasts.
• Wider effects: Some policies may be significant enough to affect the future trajectory of economic growth. The OBR is formalising its approach to “supply-side scoring ”, which ensures that these impacts are captured within the forecast processes. The OBR also applies fiscal multipliers to estimate the overall impact of the package of measures on the economy. In both cases, these are difficult technical judgements and the OBR publishes detail of when and how it has done this.