The composition of GDP can be as important to the fiscal forecast as the headline total, because some components are more ‘tax rich’ than others. A forecast that alters the composition of GDP can therefore have a significant impact on the public finances, even if the path of GDP itself is unchanged.

The income approach to measuring GDP, known as GDP(I), estimates the money earnt – predominantly by households and companies – from total output in the economy. Various elements of household income are key drivers of public sector receipts such as income tax and national insurance contributions (NICs). The single most important element is total wages and salaries earned by employees. Corporate profits are the key driver of corporation tax receipts.

  • Average earnings

    Our forecast uses an implied measure of average earnings constructed by dividing the National Accounts measure of wages and salaries by the number of employees (rather than the official ONS measure of average weekly earnings, although the two measures are conceptually similar). This allows us to fit the earnings forecast directly into the National Accounts framework on which our economy forecast is based – and, in particular, the National Accounts measure of wages and salaries, which is an important determinant of tax receipts.

    Our short-term forecast for whole economy average earnings growth is informed by available indicators of labour market slack and pay pressure, including relevant indicators from business surveys. Over the medium term, the outlook for productivity (on an output-per-worker basis) and whole economy inflation are the main determinants of our forecast for earnings growth. We also make adjustments for policies that we expect to have material effects on earnings, including those that imply a cost for employers that we would expect to be passed on to employees via wages (for example, the apprenticeship levy and auto-enrolment).

    Our main focus is whole economy average earnings growth, but when producing that forecast we ensure that it is consistent with the weighted combination of separate forecasts for market sector and general government average earnings (since the government sector may be affected by centrally imposed pay policies).

    Our forecast for government sector wage growth takes into account recent data, relevant Government policies (such as overall departmental spending plans and any limits on pay growth, such as those that applied in recent years), historical rates of pay drift and whole economy earnings growth. In the absence of public sector pay policy, we would expect government sector earnings growth to broadly match market sector earnings growth, given both sectors are competing for employees in the same labour market.

    Having produced forecasts for whole economy and general government average earnings, we can derive an implied market sector earnings forecast. We can use this to sense check whether the whole economy forecast is reasonable given public sector pay policy – and will adjust our forecast further until it looks reasonable.

    Back to top

  • Wages and salaries

    Wages and salaries is a measure of the employment income earned by employees, excluding the social contributions (e.g. pension contributions) that employers make on behalf of employees. Our forecast of wages and salaries is derived from our forecasts of whole economy average earnings and employment growth. Wages and salaries are the main determinant of the biggest sources of tax revenue: income tax and national insurance contributions.

    Back to top

  • Household disposable income

    Our forecast for household disposable income is built up by forecasting its components, which can be split into three main groups: labour income, taxes and benefits, and non-labour income:

    • Labour income is made up of wages and salaries, and ‘mixed income’. Our forecast for wages and salaries is determined by our forecasts for average earnings growth and employees. Mixed income is largely composed of self-employment income, which is derived from our forecasts of self-employment and whole economy average earnings.
    • Most of our forecasts of net benefits and taxes – such as PAYE income tax, tax from self-assessment and social benefits – are taken directly from the relevant components of our fiscal forecast. These forecasts are affected by changes in policy and in many cases will depend on other elements of the economic forecast – our forecast of income tax, for example, will depend on our forecast for wage growth. Net benefits and taxes also includes a number of employee social contributions, such as our forecast of employee NICs that is also drawn from our fiscal forecast.
    • Non-labour income includes interest receipts, interest payments, dividend income, household operating surplus, employee pension contributions, withdrawals of income from quasi-corporations (such as income withdrawn from the profits of partnerships by their owners) and miscellaneous transfers. These elements are forecast separately using a variety of approaches:
      • Our forecasts for interest receipts and payments are determined by our forecasts for deposit rates, mortgage rates and the stocks of household assets and liabilities.
      • Our forecast of employee pension contributions is largely determined by combining our forecasts of the gilt rate and closing pension liabilities, consistent with the measurement of this variable in the National Accounts, plus adjustments for the effects of relevant policies (such as auto-enrolment).
      • Household operating surplus relates to the National Accounts concept of imputed rental, which represents an estimate of the housing services consumed by owner occupiers. As this spending is imputed, the income side of the National Accounts also includes the imputed rental income. (Including both imputed flows means that the size of GDP is not affected by changes in the proportion of dwellings that are owner-occupied.) Household operating surplus is equal to imputed rental on owner occupied dwellings less the current expenses that go into the up-keep of those dwellings (such as certain repairs and interest costs). As this is largely estimated using data on actual rents, our forecast of this component is partly informed by the expected growth of actual rental income.
      • Other elements, such as dividend income, withdrawals of income from ‘quasi-corporations’ and miscellaneous transfers are typically assumed to grow in line with other elements of income, such as wages and salaries or profits, with relevant items in our fiscal forecasts, or with nominal GDP. In recent years, dividend income has risen as a share of household income, largely reflecting a significant increase in the number of people setting themselves up as single-director companies, rather than working as an unincorporated self-employed worker or an employee. We do not directly adjust our economy forecast for the effect of these incorporations on dividend income, wages and salaries or mixed income. Instead, we factor this into our fiscal forecast, which allows us to capture the effects of incorporations on individual tax receipts more accurately – for example, the incentive to incorporate changes along the income distribution, which affects the size of the adjustment that needs to be made to our tax forecasts.

    Real household disposable income is derived by deflating nominal household disposable income by the consumption deflator.

    Back to top

  • Corporate profits

    There are three elements to our forecast for corporate profits:

    • North Sea profits. These are assumed to grow in line with our forecasts for nominal North Sea output (derived from our forecasts for North Sea production and the price of oil, which are determinants of our oil and gas revenues forecast).
    • Non-North Sea private non-financial corporation (PNFC) profits. Empirical evidence suggests that profit margins over marginal costs are positively correlated with the economic cycle. At a whole economy level, this means using our output gap forecast to inform the path for profits as a share of nominal GDP. We also factor in the effects of any relevant policy measures. Non-North Sea PNFC profits are a key determinant of our onshore corporation tax forecast.
    • Financial company gross trading profits. These are forecast by residual having taken into account our expenditure-driven forecast for total nominal GDP and our forecasts for all other components of income. It is worth noting that the National Accounts concept of ‘financial company gross trading profits’ is difficult to interpret: for example, the series is consistently negative. It is not used for the purposes of the fiscal forecast, where a separate assumption is made about future financial company profits on a measure consistent with that used in companies’ tax returns.

    While the role of corporate profits as a residual in the income forecast is required by identity so that the income and expenditure measures of GDP are equal, it also serves as a useful diagnostic on the nominal GDP forecast in general. If the profile for financial company profits determined by residual were not considered plausible, it would prompt us to revisit the forecast judgements on income or nominal GDP in subsequent rounds of the forecast.

    Back to top

  • Whole economy income

    The preceding sections of this page describe how we forecast the two biggest income components of GDP – household income and corporate profits. Together these make up around 80 per cent of GDP by income. The full GDP(I) forecast comprises:

    • Labour income (wages and salaries and mixed income).
    •  Employer social contributions are made up of employers’ national insurance contributions (NICs) and employers’ pension contributions. Our forecast of employers’ NICs is taken directly from our forecast of the public finances. Employers’ pension contributions are generally assumed to grow in line with wages and salaries, plus any adjustment for the effect of relevant policies – such as the coverage and rates associated with auto-enrolment.
    • Household operating surplus (largely imputed rental income for owner-occupiers).
    • Corporate profits (onshore PNFCs, North Sea, and financial companies).
    • Operating surplus of general government and public corporations, basic price adjustment and net taxes on products. These are all derived from our forecasts of the public finances.
    • Other smaller elements of income include corporate rental income, which is assumed to grow in line with nominal GDP; stock appreciation, which is determined by our forecast for nominal inventories; and financial intermediation services indirectly measured (FISIM), which is equal to the sum of FISIM flows in the household, corporate, government and rest of world sectors and is related to interest rate spreads and stocks of assets and liabilities.

    Back to top