GDP Quarterly National Accounts: Quality and Methodology Information

Information for quarterly gross domestic product (GDP) statistics, introducing the data sources and methods used and the strengths and limitations of the data.

The Expenditure Approach

The expenditure approach for GDP is often shortened to GDP(E), and is the measure of GDP widely used by economists for modelling and forecasting the economy based on aspects of supply and demand.

GDP(E) is often presented using the equation GDP = C + G + I + Exports – Imports, or even more simply as C + G + I + (net trade). That is, the total value of consumer expenditure, government expenditure, capital investment and exports, minus the value of imports.

More formally, GDP(E) is equal to the total value of final demand for goods and services produced, minus the value of imports. Final demand refers to expenditure on goods and services which are not subsequently used up to produce other things; or in other words, the final stage of the supply chain for each good or service within Scotland.

By focussing on final demand, the expenditure approach avoids double counting the value of goods and services which are re-used or re-processed at many different stages of a supply chain. For example,  where there is a sale of ingredients to a restaurant, which then sells meals to the public, the value of the ingredients is included in the price of the meals so does not need to be counted twice. This is why spending by businesses, other than for capital investment, does not need to be included in GDP(E).

The main data and methods used for each expenditure component are summarised below.

Household Final Consumption Expenditure (HHFCE)

Estimates for HHFCE are based on a top-down methodology applied to the quarterly UK statistics published by ONS in consumer trends.

National concept HHFCE is defined as all relevant spending by residents of Scotland, whether that spending is in Scotland or elsewhere. In practical terms, we use HHFCE data from the ONS to estimate HHFCE as spending by Scottish residents in the UK plus spending by Scottish residents in the rest of the world (ROW).

The main data source for spending in the UK is the Living Costs and Food Survey (LCFS) run by the ONS, which provides estimates of how much households spend on a detailed range of goods and services within the UK, and can be used to derive the proportion of the UK total spending done by Scottish households. Outputs from the LCFS are published in the annual Family Spending publication by ONS, and quarterly LCFS results are also supplied directly to the Scottish Government. Sample sizes are small and levels of uncertainty are high, so LCFS data is pooled over several years to provide a smoother estimate of the share of UK spending in Scotland. We use the LCFS to estimate Scottish residents' share of UK spending, split into COICOP categories, from Consumer Trends, and then use data from the UK Supply and Use tables to map this spending to the product breakdown used in the QSU system.

The main data source for spending in the rest of the world, i.e. tourist expenditure, is the International Passenger Survey (IPS) run by ONS. Results from this are mapped to products for the QSU, similar to spending in the UK.

Quarterly estimates of UK and overseas expenditure are then combined together to give a total HHFCE value for each product, which is benchmarked to the annual supply and use tables.

Final Consumption of Non-Profit Institutions Serving Households (NPISH)

Estimates for NPISH final consumption expenditure are based on a simple top-down methodology applied to the equivalent series in the UK quarterly national accounts. There are limited data sources for the NPISH sector, so short term estimates are initially made by taking a fixed share of UK NPISH expenditure, and spread across products based on the latest supply and use tables. These estimates are then benchmarked to the annual supply and use tables. In the simplified representation of the expenditure approach, NPISH final expenditure is usually included with households as part of consumption.

General Government Final Consumption Expenditure (GGFCE)

The General Government sector is split into two subsectors: central government (CG) which includes the devolved Scottish Government and reserved UK Government activities in Scotland; and local government (LG). In both cases, final consumption expenditure is estimated using a top-down methodology applied to the equivalent UK series published in the UK quarterly national accounts.

The data source used to estimate the Scottish share of CG FCE is the detailed spending database linked to the Government Expenditure and Revenue Scotland (GERS) statistics, which is filtered to remove items of spend which are not part of final consumption expenditure.

The data source used to estimate the Scottish share of LG FCE is supplied to the Scottish Government by the ONS and represents the underlying Scottish data within their estimates for UK LG FCE.

Total spending is split into product groups based on the latest year in the supply and use tables, and these quarterly estimates are then benchmarked to the annual supply and use tables.

Gross capital formation

Gross capital formation is widely referred to as investment. It covers three subcategories: gross fixed capital formation, which is capital investment in fixed assets; changes in inventories, which is the building or depletion of stocks of finished goods or work in progress; and acquisitions less disposals of valuables, which are produced (non-financial) items which are not used up and held primarily as a store of value.

Gross Fixed Capiral Formation (GFCF)

Estimates of GFCF are made using a mixture of top-down and bottom-up methods for its various components. GFCF can be split into several distinct sub-categories, and the estimates for these sub-categories are made with reference to the underlying splits in the compilation of the annual supply and use tables, and to the equivalent splits in the UK supply and use tables:

• Business investment is mostly estimated using a bottom-up methodology. The main data source is an extract of the Quarterly Acquisitions and Disposals of Capital Assets Survey (QCAS) run by ONS, which gives us estimates of GFCF for several broad asset classes across most private sector industries. These values are mapped to product groups based on the latest supply and use table. For industries not covered by the QCAS, GFCF is provisionally estimated to change in line with output of those industries.
• Government investment is estimated using a top-down methodology based on the equivalent series from the UK quarterly national accounts. The data source used to estimate the Scottish share of government investment is the detailed spending database linked to the Government Expenditure and Revenue Scotland (GERS) statistics, which is filtered to remove items of spend which are not relevant.
• Dwellings investment is estimated using a bottom up methodology. The data source for provisional estimates of dwellings investment is the regional results from the ONS construction output statistics for Great Britain.
• Costs of ownership transfer for fixed assets are estimated using a top down methodology using a fixed share of the UK series based on the latest supply and use table year.

Total GFCF for each product group is added together and benchmarked to the latest supply and use table.

Business investment is defined in the Scottish national accounts as GFCF by industries excluding public administration, education, and health. This is an approximate definition and does not align with the ownership-based approach used in the UK national accounts. For example, this includes some investment by parts of government such as in the rail industry, and excludes private sector investment in the education and health sectors. The results for this subcategory of GFCF are published as official statistics in development (experimental statistics) and are not directly comparable to the series with the same name produced by the ONS, or to similarly named components for other countries. For this reason, we generally recommend that comparisons with other countries and with the UK as a whole should be made using the total GFCF value, not the business investment subcategory.

Changes in inventories and valuables

The estimates for both of these categories are initially estimated using a top-down methodology based on a fixed share of the equivalent UK series. The total value for each series is spread across product groups based on the proportions in the latest supply and use tables, and benchmarked to those tables.

Changes in inventories is one of the most uncertain parts of the accounts, and for a large number of product groups is further adjusted during the assessment of quarterly supply and use to ensure that supply and demand are balanced in each quarter.

Trade in goods and services is split into exports and imports, and for each of these into trade with the rest of the world (ROW) and trade with the rest of the UK (RUK).Trade is defined as transactions with other parties which are not residents of Scotland. This includes the buying or selling of goods and services from businesses which are located outside of Scotland, and also includes tourist expenditure in Scotland (exports, sales to non-residents) and spending abroad by Scottish residents as tourists (imports, purchases from non-residents).

Exports to the rest of the world (ROW Exports)

Exports to the rest of the world are estimated using a mixture of top-down and bottom-up methodologies for different groups of products and types of trade. These are summarised below.

ROW exports are unusual in the quarterly national accounts that the supply and use tables are not the only source for annual benchmarks. Instead, the reference level for annual exports can be supplemented beyond the years covered by the SUTs using data from the Exports Statistics Scotland (ESS) publication. For this we use detailed industry level estimates from ESS to produce additional reference year values for exports of each product. This is based on the principal product assumption that the goods or services exported by each business match the main product of the industry, for example that a metal products manufacturer will export metal products. This approach ensures that growth in export values in the most recent years is broadly consistent with the Exports Statistics Scotland results. In the compilation the annual SUTs a more detailed approach is used which can map exports from each industry across one of more product groups if needed. Having produced the extended annual benchmark values for exports of each product, we then proceed to use a mixture of top-down and bottom-up methods for estimates of quarterly exports.

For exports of manufactured products, the data source for the estimates is the HMRC Regional Trade Statistics (RTS). We use a bottom up methodology to split the values from RTS across the product groups used in GDP.

For exports of services, we use a top-down methodology based on taking a fixed share of the broad services export categories in the UK quarterly national accounts. The total value for each broad services category is spread across relevant product groups based on the fixed proportions in the latest supply and use tables.

For tourist exports, also known as non-resident expenditure in Scotland, we use a bottom-up estimate based on data from the International Passenger Survey, with expenditure split across products based on shares in the latest supply and use tables.

All types of exports for each product groups are then summed together and benchmarked to the supply and use tables.

Imports from the rest of the world (ROW imports)

Imports from the rest of the world are estimated using similar data sources and methods used for ROW exports.

For imports of manufactured products, the main data source for the estimates is the HMRC Regional Trade Statistics (RTS). We use a bottom up methodology to split the values from RTS across the product groups used in GDP.

For imports of services, while the annual International Trade In Services data is used for the production of the supply and use tables, for the quarterly national accounts we use a top-down methodology applied to the broad services import categories in the UK quarterly national accounts. The total value for each broad services category is spread across relevant product groups based on the fixed proportions in the latest supply and use tables.

For tourist imports, which are also part of HHFCE, we use a bottom-up estimate based on data from the International Passenger Survey, with expenditure split across products based on shares in the latest supply and use tables.

All types of imports for each product groups are then summed together and benchmarked to the supply and use tables.

Exports to the rest of the UK (RUK Exports)

For exports to the rest of the UK we use a simpler methodology to produce provisional results in the quarterly national accounts. This is a bottom-up approach where RUK exports for most products are projected as a fixed proportion of domestic output of each product derived from the GDP output by industry results. RUK tourist exports are estimated using results from the GB tourism survey, or provisionally forecast to move in line with the ROW tourism exports from the IPS.

Imports from the rest of the UK (RUK Imports)

RUK imports are the most uncertain component of the national accounts. There are no data sources for these estimates and therefore they must be derived as a residual in the supply and use framework, on the understanding that if there are estimates for all other components of supply and demand for each product, then the value of RUK imports must, by necessity, equal the value required to ensure that supply equals demand in all periods. In the quarterly national accounts, the value of RUK imports is first estimated to change from the base year in line with total domestic demand estimated from the other components in the framework, for each product. For many products, a further adjust is then made to the estimate to ensure that total supply and demand for each product balances.

Contact

economic.statistics@gov.scot