Chapter 4 The Supply and Use Tables Framework

Sanjiv Mahajan and Graeme Chamberlin

What is the Supply and Use Tables framework and why is it an important source of information?

How are the three approaches to measuring GDP brought together in a single statistic?

What products does the UK import from, and export to, the rest of the world?

How does the nature of production differ across industries?

How can data from the Supply and Use Tables be used to analyse a specific industry or institutional sector?

4.1 Introduction

In recent years the UK has been hit by two large shocks.

1. The decision to leave the European Union (EU) altered the nature of trading relationships with the EU and the rest of the world.
2. The COVID-19 pandemic severely disrupted the physical movement of goods and provision of services at home and abroad.

Both these events highlighted that the UK economy, like many economies, consists of many linkages and interactions. This may not be fully appreciated or understood when looking at just the headline economic statistics such as GDP.

The products we consume use inputs produced in the UK, but also the rest of the world. The products we sell are often inputs into the production of other goods and services, consumed both in the UK and internationally. Production of goods and services is increasingly organised into long supply chains that often stretch across borders.

The Brexit and COVID-19 shocks have highlighted the complexity of production in a modern economy and the challenges for statisticians who want to measure it. The national accounts estimate headline figures like GDP, but they also ensure that the flows of goods, services, capital, labour, and the financial payments associated with them add up. The job of a national accountant is to provide a balanced set of statistics that record all these activities in the economy and how they change over time. It is not easy, especially in times of significant economic change, as in recent years.

National accounts data is remarkably consistent. The tables add up; aggregates equal the sum of their parts; identities hold, so the total supply of each product equals the total uses of each product; for each industry total inputs equal total outputs. This ensures that the three different approaches to measuring GDP – production (also referred to as output), income, and expenditure – provide the same answer.

Supply and Use Tables (SUTs) framework

Maps the flows of products and services in the economy across industries and institutional sectors, and links to the rest of the world. Can be used to balance the components of the production, income, and expenditure approaches of GDP to create a single estimate of GDP.

This consistency gives the national accounts their potency. Users of data can not only view the top-level numbers but also how all the parts of the economy add up and relate to each other. This consistency reflects the work of statisticians to reconcile and balance data from a wide range of sources. The application of the Supply and Use Tables (SUTs) framework is one of the main tools used to meet this goal.

In practice, the source data feeding into the national accounts often do not correspond perfectly with their uses. The data are compiled differently, reflecting different respondents, different concepts, methodologies, and classifications. The production, income, and expenditure approaches to measuring GDP are obtained from different statistical sources such as business surveys, household surveys and administrative data. We cannot expect them to add up to the same total without intervention: the compilation and balancing of the SUTs achieves that goal.

The SUTs reconcile the national accounts using a single detailed framework, within which different data sources are confronted in a coherent way with the aim of achieving a single estimate of GDP. The SUTs provide a picture of the flows of products and services in the economy across industries, products, and institutional sectors such as households, firms, and government, as well as the rest of the world. This framework can be used to set the level of GDP for each year, in current prices and in volume terms (which reflects the removal of price inflation).

Producing a balanced set of SUTs not only provides a suitable tool for balancing the accounts, but also displays detailed information on the production processes, the interdependencies in production, the use of goods and services and the income generated in the production process. As such, the SUTs framework is an important and incredibly detailed source of information for those analysing and modelling the economy. It is used in two ways:

• Economic analyses. These study the changing structure of the economy relating to the production and flow of products and services in the domestic economy and with the rest of the world. For example, they may investigate how domestic industries are embedded in international supply chains and markets, and what the composition is of uses in the economy across households, government and corporations, as well as the rest of the world?
• Industrial and products analyses. These study changes to specific industries and products across the economy. Interest might focus on new and emerging activities such as the industries forming the digital economy, or traditional industries such as food and energy. The SUTs are typically an important data source in the formation of satellite accounts, allowing attention to be focused on a certain theme or aspect of economic and social life in the context of national accounts. Common examples are satellite accounts for the environment, tourism and unpaid household work.

This chapter provides several examples of how data from the SUTs framework can be used to analyse the UK economy, including:

• The UK’s international trade in products and services.
• Understanding the nature of production processes in different industries.
• The impact of taxes and subsidies on purchasers’ prices.
• Measuring the size of the UK’s creative sector.

Input-Output Tables (IOTs)

Describe how products, services, and other primary inputs such as labour are used to produce further products and services and satisfy final uses. They are derived from the SUTs and are typically used for impact and policy analyses relating to the economy, international trade, and the environment.

As the SUTs link the interconnectivities between industries, products and institutional sectors within the economy and relate the flow of products to final use(s), they can also be directly integrated into macroeconomic models of the economy. One of these modelling approaches involves the creation of Input-Output Tables (IOTs) which are derived from the SUTs and allow users to study relationships between final users and industrial output.

The close relationships, links, and dependencies between SUTs and IOTs needs to be understood, as they are two separate products under the same umbrella.

Two important and very topical applications of IOTs are:

• Inter-connectedness: Connections between industries, and how shocks feed through global supply chains.
• Environmental impacts of economic activity: For example, carbon footprints.

Input-output modelling involves some basic knowledge of matrix algebra. We have placed this theory and some interesting applications in Appendix 7.

4.2 The structure of the SUTs

Leontief was recognised for his contributions to input-output analysis, receiving the Nobel Prize in Economics in 1973.

The formation of IOTs and their use in economics was pioneered by Wassily Leontief, an American economist of Russian descent. In his 1936 article on ‘Quantitative input and output relations in the economic system of the United States’¹ he discussed the construction of an economic transactions table to represent the interdependencies between different sectors of a national economy. His work was partly based on the Tableau Economique, proposed by the French economist and physician François Quesnay in 1758, to track the flow of agricultural and manufactured products through the economy. The SUTs evolved much later through the System of National Accounts developments.

This article provides a brief history of how the UK SUTs have evolved since the first official tables were published in 1961:

Mahajan S (2007), ‘Development, Compilation and use of Input-Output Supply and Use Tables in the UK National Accounts’, Paper prepared for the 16th International Input-Output Conference, Istanbul

The first IOTs for the UK were produced by the University of Cambridge for the years 1948 and 1950. They consisted of eight industry groups only, with the results published in 1952 and 1953 respectively. It was not until 1961 that the first official UK IOTs were published. These were produced for the year 1954 by the Board of Trade and Central Statistical Office, using forty-six industry groups. Presently, the Office for National Statistics (ONS) constructs SUTs for the UK at the level of 112 industries and 112 products, although for disclosure reasons the results are published at the level of 105 industries and 105 products.

If you are interested in a more detailed treatment of the compilation and balancing of the SUTs, as well as their applications, this is the key source for conceptual, methodological and practical guidance:

UN Department of Economic and Social Affairs (2018), ‘Handbook on Supply and Use Tables and Input-Output Tables with extensions and analyses’

Those National Statistical Institutes (NSIs) that produce SUTs do so using different numbers of industries and products. For example, some countries have rectangular tables with several hundred industries and over 2,000 products. However, all NSIs will operate within the confines of the accepted international industrial classification (in the UK, this is the Standard Industrial Classification or SIC) and product classification (in the UK, the Classification of Products by Activity or CPA).

There are two fundamental identities within the SUTs that must hold. These are:

1. For each product: Total Supply of each product ≡ Total Use of each product
2. For each industry: Sum of the industry outputs ≡ sum of the industry inputs

4.2.1 The supply and use of products

For each product, the fundamental relationship displayed in the SUTs is:

\[\begin{align} &\text{Total Supply } ≡ \text{Total Use} \end{align}\]

The “supply” of each product to the economy is:

\[\begin{align} &\text{Total Supply} ≡ \text{domestic output } + \text{imports} \end{align}\]

“Use” describes the nature of demand for each product:

\[\begin{align} \text{Total Use } ≡ &\text{ intermediate consumption } +\\ &\text{ final consumption } +\\ &\text{ gross fixed capital formation } +\\ &\text{ changes in inventories } +\\ &\text{ net acquisition of valuables } +\\ &\text{ exports } \end{align}\]

To demonstrate, let us consider one product – motor vehicles, trailers, and semi-trailers – which is product 29 (CPA 29) in the current SUTs framework in the UK.

Domestic output relates to production within the UK, for example the Nissan plant in Sunderland, the Vauxhall plant in Luton and the Honda plant in Swindon, amongst others, and only of the product 29. There may be other products produced by these plants, but they will be classified as other products such as distribution services and repair services. Imports include those vehicles imported directly from abroad, such as BMW from Munich in Germany, and Renault from the Boulogne-Billancourt plant near Paris in France.

Correspondingly, on the use side for this product:

• Intermediate use relates to items like body work, parts or industrial services such as sub-contracting and spraying that are used in the creation of output of the motor vehicle and other industries.
• Final consumption is the use of motor vehicles that satisfies final uses, for example, the personal vehicles purchased by households.
• Gross fixed capital formation (GFCF) refers to the consumption of goods that help or support the production process but are not themselves used up in the process except for depreciating over time. In this case, GFCF would refer to vehicles that are purchased by businesses, for example, if a taxi company or an insurance company purchased a fleet of cars for the use by its staff.
• Changes in inventories refers to inventories (sometimes referred to as stocks) of unsold or semi-produced vehicles, for example, vehicles held on the forecourts of car dealers and showrooms.
• Acquisition less disposals of valuables relates to non-financial assets that essentially provide a store of value. In this case, for example, classic or vintage cars.
• Exports are sales of vehicles overseas. For example, many cars produced at the Nissan plant in Sunderland are designated for exports, primarily to the European Union, rather than for domestic sales in the UK.

The key principle underlying the SUTs is that for motor vehicles and all other products, the total supply equals total use. However, one important issue is that both sides of the equation are not carried out at the same prices. Therefore, the flows described above making up supply and use are unlikely to balance. This is because:

• Supply is valued intrinsically at basic prices – the amount a producer can obtain for a unit of output. Users, however, face market prices or purchasers’ prices.
• The difference between basic and purchasers’ prices explicitly includes trade margins and taxes and subsidies.
  • Trade margin is the difference between the actual price realised on a good purchased for resale (either wholesale or retail) and the price that would have to be paid by the distributor to replace the good at the time it is sold or otherwise disposed of. For example, a retail margin would be the difference in price that a car retailer sells a vehicle at relative to the price they paid the producer or distributor for the vehicle.
  • Taxes and subsidies can also result in differences between basic and purchasers’ prices. For example, sales of cars are subject to value added tax (VAT) which is added to get to the market price of the vehicle.

Therefore, to balance supply and use of motor vehicles we need to adjust the left hand-side of the equation so that it is also is expressed in purchasers’ prices:

Supply at purchasers’ prices ≡ output at basic prices + imports at basic prices + taxes on products – subsidies on products + trade margins

We can now show some actual numbers to demonstrate how the SUTs framework works for a particular product. Table 4.1 shows the key parts of the supply and use for product 29 (CPA29), Motor vehicles, trailers, and semi-trailers, for 2019 for the UK.

Figure 4.1a shows the total supply of motor vehicles in the UK in 2019 to be valued at £157.5 billion at purchasers’ prices. The value of domestic production and imports at basic prices are similar at £60.1 billion and £58.3 billion respectively. Trade margins of £30 billion and taxes and subsidies of £9.1 billion are needed to move these values to purchasers’ prices.

Figure 4.1b shows how this is matched to the demand of this product.

Intermediate use is £37.5 billion. Note that the full set of SUTs published by the ONS would also show how this total intermediate use was allocated across the 105 industries, that is the inputs of this product used in the production of the respective outputs of each industry.

Final use of motor vehicles in 2019 is £120 billion, just over half (£61.2 billion) of which is from final consumption expenditure. Likewise, the components of final use can also be broken down further. For example, final consumption expenditure can be separated into household, government and non-profit institutions serving households (NPISH). Gross capital formation of £16.8 billion can be further allocated into fixed investment or gross fixed capital formation (GFCF), valuables and changes in inventories. Exports of £42.1 billion can also be further broken down into goods and services with an EU and non-EU split.

The important result is that when balanced, as shown in Figure 4.1, total supply equals total use.

4.2.2 The inputs and outputs of industries

For each industry, the fundamental relationship displayed in the SUTs is:

\[\begin{align} &\text{Total Outputs } ≡ \text{Total Inputs} \end{align}\]

The output of an industry relates to the value of all products created by that industry. Note that this might consist of more products than just the primary product associated with that industry.

The inputs for each industry can be sorted into two main categories:

• Intermediate consumption: The products used up in the creation of industry output.
• Gross value added (GVA): the generation of income by each industry, which is split between taxes less subsidies on production, compensation of employees, and gross operating surplus. These can be thought of as payments made by the industry to those that provide labour and capital to the industry as well as to the government.

Again, it is helpful to demonstrate this relationship using a specific industry as an example. Industry 29 (SIC29) in the UK SUTs is Manufacture of motor vehicles, trailers, and semi-trailers. You will not be surprised to know that the principal output of this industry is product 29 (CPA29) used previously. The production account recording the value of output and inputs for this industry is presented in Figure 4.2.

The total value of output for the motor vehicle industry in 2019 was £64.8 billion. Again, it is important to realise that this will consist of the value of all products produced by the industry, not just its principal product of motor vehicles. For example, it might also consist of distribution, leasing and financial services, and parts provided for repair and maintenance.

The intermediate consumption of the industry is the value of the products used up during production to make its output, which in 2019 was £47.4 billion. Figure 4.2 lists the main product inputs, the largest with a value of £18.2 billion being inputs created by the industry itself. Think about the huge array of parts that the industry produces itself (engines, bodywork, chassis, electrics, and other) that are assembled into the final product.

The other intermediate consumption products are listed in Figure 4.2. For example, basic iron and steel, other basic metals and fabricated metal products sum to £9.2 billion; paints, petrochemicals and rubber and plastic products add up to £3.8 billion; machinery and electrical equipment combined was £3.4 billion.

The GVA for the industry is the difference between the values of output and intermediate consumption, which in 2019 was £17.4 billion (= £64.8 billion − £47.4 billion). The income account for the industry then describes how this was split between taxes less subsidies on product (£133 million), the compensation of employees (£10.3 billion) and a gross operating surplus (£7.0 billion).

4.2.3 The aggregate tables

So far, we have only considered the supply and use of one product and the inputs and outputs of one industry, forming one row and one column of the complete SUTs framework. The aggregate tables show the same relationships for all 105 products, each recorded as a separate row, and 105 industries are each recorded in a separate column. The basic structure of the SUTs framework for the entire economy is shown in Figure 4.3.

Our example looking at the supply and use of a specific product, such as product 29, motor vehicles, trailers, and semi-trailers, illustrates the fundamentals of the SUTs. The flow of products from supply through to uses can be tracked by reading across the rows of the tables. However, the full detail underpinning the set of SUTs consists of more tables yet to be introduced.

First, the domestic output part of the Supply Table shows the output of each product by each industry. In Figure 4.1, we only considered the total output of the product, essentially the total output of that product created by all industries in the domestic economy, and not just the main industry.

In the UK, the Classification of Products by Activity (CPA) is closely linked with industry classifications from the Standard Industrial Classification (SIC). We expect most products to be produced as the principal output of the industry to which the product classification is aligned, as we saw with CPA29 and SIC29 relating to motor vehicles, trailers, and semi-trailers. If it were the case that each product was only produced by that corresponding industry, then all the entries in the domestic supply matrix would be on the diagonal, that is, product CPA 01 (products of agriculture), is only produced by industry SIC 01 (Crop and Animal production), and so on.

In practice, it is possible for an industry to produce more than one product, in which case there would be off-diagonal entries in this matrix, such as secondary production and by-products. Consider a farmer who sets up a farm shop. He or she now also produces retail services as a secondary product alongside agricultural products (retail trade services are product CPA 47). These off-diagonal entries are usually small, so the ONS does not publish this matrix in full because some of the entries would publish information that could reveal confidential financial data.

The intermediate use part of the Use Table shows what products are used as an input into the production process of each industry. Reading down the columns, you will see the inputs used up in the production processes of each industry, as we saw in Figure 4.2 for the manufacture of motor vehicles. For example, the three main inputs into the production of industry SIC 51, Air transport, are products CPA 19 coke and petroleum products, CPA 33.16 repair and maintenance of aircraft and spacecraft, and CPA 51 air transport services. There are of course many other smaller inputs in addition to these.

Reading across the rows, you will see that some products are used extensively by most industries, for example, product CPA 35.1, electricity. Some products are more sparsely used across industries. For example, product CPA 25.4 which is weapons and ammunition, is only used in 6 of the 105 industries. Of these, 98% is accounted for by industry SIC 25.4 (manufacture of weapons and ammunition) and industry SIC 84 (public administration and defence).

The full set of the SUTs provide more of a breakdown in the components of final use, especially into final consumption and gross capital formation categories. In Figure 4.4 you will also notice a set of extra tables below the intermediate use part of the Use Table.

• Production accounts by industry: By reading down the columns, you will see that for each industry, total output is equal to total inputs (that is, the sum of intermediate consumption and GVA).
• Generation of income account by industry: This shows how GVA in each industry is split by the three uses (also known as the income components): compensation of employees, taxes less subsidies on production, and the gross operating surplus. Figure 4.2 showed an example column for SIC29, manufacture of motor vehicles, trailers, and semi-trailers.

4.2.4 Summary Supply and Use Tables for the UK

The published SUTs for the UK consist of 105 products and 105 industries, so presenting the full set of tables here is impractical. To illustrate how the tables work, in Figure 4.4 we use a summary of the SUTs where products and industries have been aggregated to four basic categories: agriculture, production, construction, and services. These figures correspond to the year 2019,which was the latest year published in the 2021 Blue Book.

The Blue Book is an annual publication providing the full set of economic accounts (national accounts) for the UK.

Figure 4.4a shows the supply side of the UK economy. In terms of total supply, services are the largest component. However, imports of produced goods greatly exceed those for services, so although services constitute most of the output, international trade is still dominated by flows of physical products.

One aspect that needs a little bit more of an explanation is the negative entry for trading margins for services. This is a national accounts convention designed to prevent the double counting of retail and wholesale services – which would otherwise enter the supply table as both domestic output and a margin. The output of these sectors equals total trade margins, so is subtracted from the distributors’ trading margins for the wholesale and retail products, which lie within the services products.

The first part of the Use Table, Figure 4.4b(i), shows output of the four broad industry groups. The top part shows the intermediate use for products, used as inputs into each industry’s production process. At the bottom is the total value (in basic prices) of each industry’s total output.

The production accounts, for each industry, show how total output reflects the sum of intermediate consumption and GVA. Total UK GVA, in basic prices, in 2019 is just over £2 trillion (£2,017,344 million), of which £1.7 trillion is accounted for by services. Figure 4.5 shows GVA by industry, with more detail in the service industries as well as the proportion of total GVA accounted by each industry in percentages. The sum of the service industries accounts for 79% of total GVA, whereas in comparison, agriculture is less than 1% of total GVA.

Figure 4.4 also shows the generation of income accounts, where for each industry, GVA is broken down into three main sources: taxes less subsidies on production, compensation of employees, and gross operating surplus.

The final part of the Use Table, Figure 4.4b(ii), shows the final uses for each of the four main product categories, as well as total use, which is the sum of intermediate use and final use. In this table, final use has been further broken down showing the contributions of each product to final consumption expenditure, gross capital formation and exports. This shows that services products are the largest component of final consumption, construction products the largest component of gross capital formation, and production products the largest component of exports.

Figure 4.6 shows the overall composition of final use in the UK for 2019. The components of final consumption expenditure have been split out further into households, general government, and non-profit institutions serving households (NPISHs). Together these account for just under two-thirds of total final use.

Gross capital formation, which consists of gross fixed capital formation, valuables, and the change in inventories, represents 14% of final use. Exports account for 23% of total final use, of which exports of goods and services are broadly equal in value.

4.2.5 The three approaches to measuring GDP

The SUTs provide a coherent picture of the UK economy, whereby the Total Supply of products ≡ Total Use of products. This is not just in the aggregate, but for each of the 105 products. Likewise, for each of the 105 industries, total output ≡ total inputs, and GVA is the difference between total output and intermediate consumption. Therefore, the tables are balanced across the supply and use for products, and the inputs and outputs of industries.

When balanced, the SUTs demonstrate that the estimate of GDP is the same, whether calculated using the production, income, or expenditure approaches. In 2019, UK GDP at market prices was £2,255,283 million. Looking at Figure 4.1 and using the data from Figure 4.4, we can see that each of the three approaches to estimating GDP gives the same answer.

Production approach

\[\begin{align} \text{Output at basic prices: £}3,842,866 \text{ million} \\ − \text{ intermediate consumption: £}1,825,542 \text{ million} \\ = \text{GVA at basic prices: } 2,017,344 \text{ million} \\ + \text{ Taxes} \textit{ less } \text{subsidies on products: £}237,939 \text{ million} \\ = \text{GDP at market prices:} \textbf{ £2,255,283} \text{ million} \end{align}\]

Income approach

\[\begin{align} \text{Compensation of employees: £}1,097,245 \text{ million} \\ + \text{ Gross Operating Surplus: £}893,197 \text{ million} \\ + \text{ Taxes} \textit{ less } \text{subsidies on production: £}26,902 \text{ million} \\ = \text{GVA at basic prices:} 2,017,344 \text{ million} \\ + \text{ Taxes} \textit{ less } \text{subsidies on products: £}237,939 \text{ million} \\ = \text{GDP at market prices:} \textbf{ £2,255,283} \text{ million} \end{align}\]

Expenditure approach

\[\begin{align} \text{Final consumption: £}1,873,083 \text{ million} \\ + \text{ Gross Capital Formation: £}402,888 \text{ million} \\ + \text{ Exports: £}699,250 \text{ million} \\ − \text{ Imports: £}719,938 \text{ million} \\ = \text{GDP at market prices:} \textbf{ £2,255,283} \text{ million} \end{align}\]

ONS Resource

Summary SUTs for the UK can be found in Chapter 3 of the ONS Blue Book.

The full set of SUTs for the UK for the years 1997 to 2019 and the associated IOTs can be found in the Supply and Use Tables on the ONS website.

The next part of this chapter describes how this consistency is achieved and how the SUTs are used in the balancing of the UK national accounts.

4.3 How are the SUTs used to balance the national accounts?

The attractiveness of balanced national accounts is that they provide a coherent view of the many interactions that make up the economy, resulting in the three approaches to measuring GDP being the same. However, the data sources that underlie the national accounts are often different in terms of when, how and from whom the data is collected. Although the SUTs present a balanced picture, it is actually through the construction of these tables that balancing the national accounts takes place.

The process of SUTs balancing at the ONS aims to fully reconcile production, income, and expenditure data at a very detailed level to the nearest million pounds. Due to the large amount of data needed, the SUTs are usually published approximately 18 months after the end of the latest balanced year. For example, the 2019 SUTs were first published in the 2021 Blue Book. The supply and use estimates underpin all but the most recent estimates of GDP, for which a separate quarterly approach to balancing is used.

There are various ways to carry out balancing, each with its own merits and problems, but the essential differences are whether balancing is carried out simultaneously or sequentially, and whether the process is manual or automated. Regardless of the approach undertaken, before balancing begins, time is spent assessing the raw statistical sources and making any necessary quality adjustments to the data.

4.3.1 The sequential approach

The sequential approach aims to establish the strongest elements of GDP first, and balance the remaining elements to those levels. For example, GDP by the income approach might be set initially, after which the levels for GDP Expenditure and GDP Production would be brought into line.

Alternatively, the strongest elements of each approach to measuring GDP can be established first. For example, the levels of industrial output (Production), compensation of employees (Income) and Household Final Consumption (Expenditure) might be set, and the remaining elements balanced around these afterwards.

In the past, the ONS had balanced the SUTs using a centralised team of supply and use experts, fixing the level of output and income GVA first by industry, and then adjusting the product balances afterwards, on an iterative basis. Virtually all the balancing was done manually, but simple methods of automated balancing were used at the very end of the process to deal with small balancing differences of less than £10 million and/or 5% of the product supply.

4.3.2 The current ONS process for balancing

The current process is iterative, using alternating rounds of product supply and demand and industry input and output balancing to arrive at final estimates of the three measures of GDP at the end of the process:

1. Row balancing. This challenges the data in each row with the aim of achieving a balance across the row, in other words, so that the accounting relationship that the supply of a product is equal to the use for that product.
2. Column balancing. This challenges the data in each column with the aim of achieving a balance down the column, so that for each industry, the inputs to the process of production equals outputs.

The basis of this approach is that the supply and use of products and industry inputs and outputs are intertwined through the output and intermediate consumption of products, so each product balance unbalances the industry picture and vice‐versa. However, this iterative process of successive row and column balancing effectively hones in on a position of balance by way of narrowing the degree of imbalance remaining in the supply and use framework after each balancing cycle. As the number of iterations rises, the product and industry imbalances gradually reduce towards zero.

While the description above may seem to indicate fairly mechanistic balancing, it is still a largely manual approach. A significant amount of knowledge of the methods and quality of the basic data is used as part of the process. This confrontation process identifies areas of inconsistency between the various sources, which can then be investigated. The evolving balance is reviewed at each stage to see how the economic picture is developing and make sure that aggregate and detailed time-series estimates are credible. It is only once the annual imbalances are economically insignificant, for example at less than 0.1 per cent of total supply, that an automatic approach is applied to apportion these across the economy and reach the final balanced position.

Quarterly balancing for more recent data takes place outside of the supply and use system. In the UK, there are far more data available on output than in the other two approaches in the near term. The rate of change in output is, for this reason, used to set the rate of change of UK GDP. Income and expenditure levels and growth are aligned to this. As more data from all three approaches become available, the GDP estimate will be improved, and potentially revised.

4.3.3 Automated balancing

Sir Richard Stone is a British economist who is widely recognised as the “father of national income accounting”. In 1984 he won the Nobel Prize for his work in developing accounting models for tracking activity in the national economy.

Automated balancing of the national accounts is not a new concept, but has been around since the 1940s, following the initial work of Sir Richard Stone and others. Many National Statistical Institutes conduct some form of automated balancing, but, similarly to the ONS, usually only to deal with small rounding issues towards the end of the process. Most automated balancing procedures follow a basic approach where a balance is achieved by first apportioning the supply and use differences for each product across the various industries, and then scaling the industry totals to ensure consistency of output and income GVA. However, in some countries, greater use is made of automated balancing techniques.

More sophisticated automated approaches can determine how much balancing affects each component series based on quality factors. For instance, those series that exhibit lower variance or considered more reliable are less prone to being adjusted.

4.4 The UK’s international trade in products and services

The SUTs framework tracks the flow of products through the economy, including those that enter via imports and those that leave via exports to the rest of the world. The tables provide a detailed and very useful source of information on the UK’s international trade in products and services and how it has changed over time.

Figure 4.7 shows imports as a percentage of total supply for several different product types. In 2019, imports accounted for 15.8% of the value of total supply of products in the UK, but this proportion was significantly higher in the manufacturing (48.4%) than the services (6.4%) products. This is demonstrated in Figure 4.8, which shows the top five products and bottom five products for the UK in terms of import penetration. For leather and clothing products, around 90% of the UK supply was accounted for by imports in 2019. However, for products such as retail services, sewerage and veterinary services, imports were negligible in total supply.

Figure 4.7 also shows how import penetration into the UK has changed between 1997 and 2019. In total, imports as a proportion of supply increased from 12.9% to 15.8%, but the respective increases for manufacturing (33.1% to 48.4%) and agricultural products (22.4% to 29.6%) were far larger than for services (4.3% to 6.4%). However, it should also be noted that there were significant increases in import intensity for some services, notably finance and insurance (3.7% to 7.2%) and professional and support activities (7.0% to 12.9%). The growing share of imports in total supply, particularly for manufactures, shows the growing importance of international supply chains and outsourcing in the production of UK goods and services.

Similar changes can be viewed when looking at exports as a proportion of total demand (Figure 4.9). In total, there was an increase from 12.5% in 1997 to 14.6% in 2019. However, it is noticeable how important certain services are in accounting for total UK exports, and in contributing to the change over recent decades.

While exports of manufacturing account for 23.5% of total demand for these products in 2019, this is a relatively small increase of 2.5 percentage points since 1997. Over the same period, the share of exports in total demand for certain services has increased far more substantially, including information and communication (6.8 percentage points), financial and insurance (6.3 percentage points) and professional and support activities (9.2 percentage points).

The top five and bottom five products by export shares for 2019 are shown in Figure 4.10. This shows that the relative importance of foreign demand relative to domestic demand varies considerably by product, from over 70% of air and spacecraft related machinery to negligible shares for water, sewerage and waste management products.

There are further analyses of the UK’s trade with the rest of the world in Appendix 7 on the Input-Output Tables, including Trade in Value Added (TiVA), which emphasises the role played by UK producers of goods and services as part of global supply chains.

Analysis of the SUTs shows the flow of products and services into and out of the UK economy and how import intensities and export shares by products and services can differ significantly. Looking at the tables over time also highlights fundamental shifts in the structure of the economy towards imports of manufactured goods and exports of financial and business-related services, often considered to be the UK’s comparative advantage.

4.5 Understanding the nature of production processes in different industries

One of the fundamental relationships in the SUTs framework is that for each industry, output is equal to inputs. Inputs consist of the intermediate consumption of products and services used in the creation of output, and gross value added which relates to three income measures: taxes minus subsidies on production, compensation of employees, and gross operating surplus.

Figure 4.11 shows a summary of how inputs into each industry were proportioned across these categories in 2019. The full SUTs provide a detailed breakdown of the inputs feeding into the output of all industries, and are therefore a very useful source on the nature of production processes across the economy.

4.5.1 Intermediate consumption

Figure 4.11 shows that intermediate consumption of products is a larger share of inputs into agriculture, production, and construction than for the services industries. This is apparent from Figure 4.12, which shows the top five and bottom five UK industries for intermediate products as a share of total inputs. The top five list is dominated by industries that essentially process primary products, whether it be energy- or agriculture-based industries. In contrast, the bottom five industries are those where GVA is a very high proportion on inputs, either because they are labour intensive industries such as domestic household workers, accountancy, and legal professions (likely to have relatively high compensation of employees), or because they are capital intensive such as owner-occupiers’ housing or sewerage (likely to have relatively high gross operating surplus)

4.5.2 Compensation of employees

Labour-intensive industries are likely to be those where the compensation of employees accounts for a high proportion of total inputs. Figure 4.13 shows the respective top five and bottom five industries in 2019. Labour intensive industries are domestic household employers, education, residential care and social work, accountancy, and employment activities. Industries where labour costs are a small proportion of inputs tend to be those that are capital-intensive, such as owner-occupiers’ housing (where gross operating surplus is likely to be large), and those that process primary products, particularly energy-based, where intermediate inputs likely account for a high proportion of inputs.

4.5.3 Energy intensity

The SUTs provide detailed information on the inputs to each industry, so a vast array of analyses is possible. One topic that might be of particular interest, given the growing importance of environmental concerns and targets, is the energy intensity of production. For each industry, this can be calculated as the share of energy inputs in total inputs, where energy inputs consist of the following five products:

• CPA_B05: Coal and lignite
• CPA_B06 & B07: Crude petroleum; natural gas and metal ores
• CPA_C19: Coke and refined petroleum products
• CPA_D351: Electricity; transmission and distribution
• CPA_D352_3: Gas; distribution of gaseous fuels through mains; steam and air conditioning supply

Figure 4.14 shows energy inputs as a percentage of total inputs across all UK industries between 1997 and 2019. Because these have been measured in purchasers’ prices, the trend will strongly reflect the movement in energy prices relative to total prices. For example, the peak in oil prices in 2008 followed by the collapse during the global financial crisis the following year can be seen in the data. Also, the energy price super-cycle between 2010 and 2016, reflecting expectations of growth in emerging market economies such as China, is clear in the data.

The variation in energy intensities across industries can be observed by looking at energy inputs as a percentage of total inputs each year. Figure 4.15 presents the top five and bottom five industries in terms of energy intensity in 2019. In terms of the top five energy intensive industries, there are few surprises, consisting of the energy industries, air transport and petrochemicals. The least energy-intensive industries are based in the services sector, and are predominantly labour intensive.

Appendix 7 shows how consumption-based measures of energy use and environmental footprints can be measured in comparison to the standard production-based approach typically used to measure progress against environmental targets.

Differences in energy intensities across industries imply that changes in the industry composition of the economy, in other words a shift from production to services, would lower energy use and the associated emissions of greenhouse gases. In effect, when lower domestic production of energy-intensive products is replaced by imports, the economy is outsourcing its energy use overseas, which can be handy in helping to meet environmental targets, at least within the UK.

4.6 The impact of taxes and subsidies on purchasers’ prices

Part of the difference between basic and purchasers’ prices is accounted for by the impact of taxes and subsidies. Taxes increase purchasers’ prices relative to basic prices, whereas subsidies reduce them. Figure 4.16 shows the value of taxes minus subsidies on total supply at purchasers’ prices between 1997 and 2019. This shows a stable path, close to 5%, although this dipped towards 4% following the large but temporary reductions in value-added tax (VAT) in 2009 as part of the government’s fiscal stimulus plan in response to the global financial crisis.

Although the overall tax rate (net of subsidies) in the UK economy appears stable, there are very large differences by product, as shown in Figure 4.17. In 2019, net taxes on alcoholic beverages and tobacco products, coke and refined petroleum products, and gambling and betting services were significantly higher than the UK average, reflecting the impact of specific excise duties on these products. However, for other products such as scientific research and rail transport services, the net tax rate was negative, reflecting the impact of government subsidies on these products. When using SUTs it is always worth remembering that differences in valuations at basic prices and purchasers’ prices can be significant for certain products.

4.7 Measuring the size of the UK’s creative sector

One of the main applications of data created through the SUTs framework is on the analysis of specific industries or sectors in the economy, including satellite accounts. An example of this is the measurement of the size of the UK’s creative industries, currently published by the Department for Culture, Media, and Sport (DCMS).

4.7.1 What are the creative industries?

The creative industries are defined by DCMS as “those industries which have their origin in individual creativity, skill and talent and which have a potential for wealth and job creation through the generation and exploitation of intellectual property”. These were identified based on creative intensity using the following three steps:

1. Through consultation a list of “creative occupations” was identified.
2. The proportion of creative jobs in each industry was calculated (the creative intensity).
3. Industries with creative intensity above a specified threshold are considered creative industries.

Figure 4.18 presents the list of the industries (using SIC 2007) identified by DCMS as making up the UK creative sector.

4.7.2 Using data from the SUTs framework to measure the size of the UK’s creative industries

The Department for Digital, Culture, Media and Sport (DCMS) deems the most reliable estimates of GVA (measured in current prices) to be the annual SUTs, which contain balanced data for each of the 112 industries represented. However, the majority of the industries in the SUTs are published at the 2-digit SIC level, whereas the DCMS industries identified in Figure 4.18 are defined at the 3- or 4-digit SIC levels. This means a method for breaking down the SUT data to a more detailed industry level must be applied.

This was achieved by using approximate gross value added (aGVA) data from the UK non-financial business economy from the Annual Business Survey (ABS), by:

• Extracting aGVA from the ABS at a 3- or 4-digit industry level (for example, SIC 32.12).
• Calculating aGVA from the ABS at 2-digit industry level (for example, SIC 32, by aggregating industries in the division together).
• Calculating the proportion of the division aGVA that each industry accounts for (for example, aGVA for SIC 32.12 as a proportion of SIC 32).
• Applying the proportion for each industry to the division GVA in the SUT, to get a national accounts consistent estimate of GVA for each industry at the detailed level.

The methodology describing in full how DCMS estimates the creative industries can be found in the Creative Industries Economic Estimates Methodology Guide.

This method, using the national accounts consistent SUT data, is preferable to only using aGVA from the ABS, because the data from the SUTs has been balanced and judged to be of higher quality.

4.7.3 The contribution of the creative industries to UK GVA

Figure 4.19 provides a summary of the creative industries contribution to UK GVA. According to DCMS estimates, the creative industries contributed £115.9bn in 2019, accounting for 5.9% of total UK GVA. However, between 2010 and 2019, the GVA of the creative industries has increased by 43.6% in real terms. As Figure 4.20 shows, over the same period, creative industries GVA has been growing significantly faster than the UK economy.

• IT, software and computer services contributed the most towards creative industries GVA in 2019. The contribution was £47.0bn, accounting for 40.6%. This subsector also accounted for 53% of growth in creative industries GVA between 2010 and 2019.
• Film, TV, video, radio and photography was the next biggest contributor to GVA in this sector. This was £21.6 billion, accounting for 18.6% of the total.
• Advertising and marketing were the second biggest contributor to growth between 2010 and 2019. They accounted for 19% of growth in creative industries GVA over the period.
• Design and designer fashion were the fastest growing of the creative industries between 2010 and 2019. GVA more than doubled in real terms.

4.8 How policymakers use the Supply and Use Tables

The SUTs domain has also become increasingly important to users of economics statistics. The SUTs provide detailed information on the interactions in the economy between industries, products, and institutional sectors as well as the rest of the world. They are a valuable source of information to analyse structural changes in the economy, and specific industries and institutional sectors. For example, SUTs are often the main source of data for satellite accounts, allowing a focus on key economic and social changes, such as the environment and home-production, within the context of the overall national accounts.

The development of the SUTs framework has meant many analytical opportunities have materialised, and the use of the framework is reaching out into new and exciting areas serving policy needs, nationally and sub-nationally. With the increasing demand for SUTs, this has in turn reinvigorated the need for IOTs and new products based on IOTs. Appendix 7 demonstrates several of these interesting applications with respect to international trade and the environment.

4.9 Summary

This chapter has set out to explain the SUTs framework and its many uses in analysing the economy. The derivation of the SUTs is at the heart of the national accounts, providing the framework in which national accounts data are balanced, and enabling a single estimate of GDP using the production, income, and expenditure approaches to be established and published. This maintains the circular flow of income within the economy at a detailed level and provides the national accounts with the consistency that makes them valuable to key users of economic statistics including central banks, finance ministries and the private sector.

4.10 Further reading

• Leontief W (1936), ‘Quantitative input and output relations in the economic systems of the United States’, The Review of Economics and Statistics, Volume 18, Issue 3, pages 105 to 125

Notes