Capital stocks and fixed capital consumption, UK: 2015

Annual estimates of the value and types of non-financial assets used in the production of goods or services within the UK economy and their loss in value over time.

This is not the latest release. View latest release

This is an accredited National Statistic. Click for information about types of official statistics.

Contact:
Email Jayne Olney

Release date:
1 December 2015

Next release:
To be announced

1. Main points

  • The United Kingdom’s (UK) net capital stock was estimated at £4.1 trillion at the end of 2014. Between 1998 and 2007, average annual growth was 2.2% but this fell between 2008 and 2014 to 1.1%. Since 1997, estimates of net capital stocks increased by £1.0 trillion (33.1%)

  • Consumption of fixed capital for the UK was estimated at £236 billion in 2014, an increase of £86 billion (57.7%) compared with 1997. The 2014 estimate is now £6 billion (2.4%) higher than the 2007 peak for this series, despite the fall in the economic downturn period between 2008 and 2009

  • Services industries held an estimated 77.4% of total net capital stocks at the end of 2014. The construction industry held an estimated 8.3% and manufacturing held 6.1%. Other production industries accounted for the remaining 8.2% of net stocks held

  • A number of methodological improvements have been introduced in this release; these were detailed in 2 articles titled ‘methodological changes to estimation of capital stocks’ and ‘impact of the methodological changes to capital stocks’ published on 14 September and 1 December respectively. The links to these articles are contained within the background notes

Back to table of contents

2. Overview

A number of methodological improvements were introduced for this publication, therefore the timeseries has been revised back to 1997. Users are advised to consult the background notes and articles detailing methodological improvements (269.4 Kb Pdf) and impact of the methodological improvements (235.4 Kb Pdf) for details on these changes.

Also published as part of this release are estimates of gross capital stocks, net capital stocks and consumption of fixed capital by institutional sector, asset and industry level. Further information on the detailed classification of these 3 areas can be found in our publication of concepts, sources and methods.

All data referred to in this bulletin are annual estimates of chained volume measures (CVM) unless otherwise specified. These are time series put in real terms by computing the volume of each year in the prices of the preceding year. The data are then chained together to obtain a time series of production figures from which the effect of price changes have been removed. The CVMs in this publication are referenced to 2012.

Back to table of contents

3. Important terms

  • Capital stocks represent the value of all fixed assets used in production in the economy that are still in use, such as machinery, dwellings and intellectual property products, formerly intangible fixed assets, such as software.

  • Economic assets are a store of value representing the benefits the economic owner will get by holding or using the asset over a period of time.

  • Fixed assets are non-financial items which are used repeatedly in the process of production for more than one year. For example, a machine on a production line or software used in production.

  • Gross capital stocks tell us how much the economy’s assets would cost to buy again as new, or their replacement cost. All of the fixed assets in the economy, that are still productive and in use, are added up to calculate this, regardless of how old they are or how much they may have deteriorated since they were first used. This measure shows the value at the end of the year. This is mainly calculated as an intermediate step towards net capital stocks but individually provides a broad indicator of the productive capacity of an economy.

  • Net capital stocks show the market value of fixed assets. The market value is the amount that the assets could be sold for, which will be lower than the value of gross capital stocks. This reflects the fact that the assets will have had some wear and tear compared to a new asset. This measure shows the value at the end of the year. This measure is used in preference to gross capital stocks as it provides a valuation of assets in the economy after the removal of depreciation.

  • The consumption of fixed capital is the decline in the value, or depreciation, of fixed assets in the economy over a time period. The decline in value can be due to wear and tear, assets no longer being used, or normal accidental damage. It can also be described as the quantity (or value) of the capital stocks which is used up in that period. Whilst these data are interesting, their primary purpose is to move from various gross measures of economic flows to the corresponding “net” variable, in particular for production and income (net domestic product, net value added) and a number of demand variables such as net investment.

  • Gross fixed capital formation (GFCF) is the acquisition less disposals of produced fixed assets; that is assets intended for use in the production of other goods and services for a period of more than a year. Acquisition includes both purchases of assets (new or second-hand) and the construction of assets by producers for their own use. New buildings and dwellings, and major improvements to buildings and dwellings are included in GFCF, but the acquisition and disposal of existing buildings and dwellings are not.

Back to table of contents

4. Calculating capital stocks and consumption of fixed capital

Figure 1 shows how gross capital stocks are calculated using the Perpetual Inventory Method (PIM). The cumulative sum of net investment in assets (GFCF), that is, the capital stock, is calculated by adding investment this period to the capital stock in the previous period and subtracting the value of assets which have reached the end of their useful life or that have been scrapped as a result of bankruptcy. The PIM replicates this process for each industry and asset combination for every year of data in the model.

Figure 2 shows how net capital stocks are calculated. This is also a ‘stock’ measure and estimates the value at the end of the year. The same process that is used for the estimation of gross stocks is used; however, an additional component for depreciation (for example, from wear and tear) is subtracted from the gross value. This can be thought of as the quantity of assets ‘used up’ in a year. At the end of an asset’s service life, its whole value has been ‘used up’, and it no longer contributes to the net (or gross) stock level.

The consumption of fixed capital is an estimate of a ‘flow’. It represents the change in the value of assets during the year. Figure 3 shows that it is made up of the sum of transfer costs (costs associated with purchasing or disposing of an asset) from GFCF, the depreciation or loss in value of assets due to usual wear and tear as well as the value of assets lost when companies go bankrupt. This value is calculated for each year within the model.

Back to table of contents

5. What do you think?

We would welcome any feedback you may have, and would be particularly interested in knowing how you make use of these data to inform your work. Please contact us via email: gcf@ons.gov.uk or telephone Jayne Olney on +44 (0)1633 455250.

Back to table of contents

6. Capital stocks and consumption of fixed capital in detail

Estimates of capital stocks and the consumption of fixed capital tend to follow a relatively smooth path over time, because they are recorded as accumulated balances. Nevertheless, in cases where the asset price or Gross Fixed Capital Formation (GFCF) have changed significantly, changes in the rate of increase or decrease in the capital stock can be observed.

The level of fixed GFCF investment fell sharply during the economic downturn by 5.9% in 2008 and 14.4% in 2009. Despite this, growth in capital stock remained above the level of depreciation and continued to grow in 2008 and 2009, albeit at slower rates compared with the decade prior to the downturn.

Gross Capital Stocks

Gross capital stocks was estimated at £7.0 trillion at the end of 2014, an increase of 41.3% since 1997 and equivalent to 2.1% growth per annum. In the decade prior to the economic downturn, gross capital stocks rose on average by 2.4% per annum, however this slowed to 1.7% per annum between 2008 and 2009. Gross capital stocks growth increased by 1.5% during the period 2010 to 2014, remaining below rates seen prior to the downturn (Table 1).

Net Capital Stocks

Net capital stocks account for the depreciation in assets, so both the level and the rate of increase in the net stock will be lower compared with gross stock (Figure 4).

Net capital stocks were estimated at £4.1 trillion at the end of 2014, an increase of 33.1% since 1997 and equivalent to 1.7% growth per annum. The average annual growth in net capital stocks slowed during 2008 and 2009 to 1.2%, compared with 2.2% prior to the downturn. Growth in subsequent years has remained comparatively subdued at 1.0% on average.

Consumption of fixed capital

In 2014, the consumption of fixed capital was estimated at £236 billion, an increase of 57.7% since 1997 (equivalent to an average of 2.7% growth per annum); the largest contribution came from the costs associated with the transfer of ownership of non-produced assets, which increased by 127% between 1997 and 2014. In contrast with the gross and net capital stock estimates, capital consumption is a flow variable and showed negative growth during the economic downturn. The level of consumption of fixed capital peaked in 2007 at £230 billion, but fell by 6.5% and 2.3% in 2008 and 2009 respectively to £210 billion in 2009. This may have been caused by a sharp fall in household transfer costs (costs associated with purchasing or disposing of an asset), which can be attributed to the adverse impact of the financial market shock on the housing market. Growth in the consumption of fixed capital subsequently picked up, albeit at a slower rate compared with pre-downturn rates. In 2014, the level of the consumption of fixed capital was 2.4% (£6 billion) higher than the 2007 peak (Figure 5).

Back to table of contents

7. Analysis by institutional sector

Net capital stocks estimates have been used for this analysis as they are a measure of the market value of fixed assets (what these assets were worth at the time).

The percentage share of net capital stocks owned by the institutional sectors remained relatively stable between 1997 and 2014. At the end of 2014, ‘non-financial corporations’ (NFCs) were estimated to have held the largest share of total net capital stock at £1.7 trillion (41.1%), followed by households and NPISH sectors which held an estimated £1.6 trillion (38.7%) of assets. NFCs held the largest number of assets in ‘other buildings and structures’, at £0.8 trillion (49.8% of total net capital stock held by NFCs). Households and NPISH primarily held assets in dwellings (excluding land) at £1.5 trillion (94.1%) of total net capital stock held by households and NPISH).

Figure 6 and Table 2 show that all institutional sectors experienced growth in net capital stocks between 1997 and 2014, albeit at different rates. In percentage terms, general government was the headline sector that experienced the strongest growth at 63.6% (equivalent to 2.9% on average per annum), followed by NFCs at 35.0% (equivalent to 1.8% on average per annum), and financial corporations at 33.9% (equivalent to 1.7% on average per annum). Households and NPISH increased by 19.3% (equivalent to 1.0% on average per annum).

Table 2 shows that the economic downturn during 2008 and 2009 affected the growth of net capital stock to varying degrees by institutional sectors. Growth in net capital stock held by NFCs was adversely affected by the downturn, falling from 2.4% on average per annum in the pre-downturn decade to 0.6% in 2008 and 2009. In contrast, growth in net capital stock held by general government rose during the downturn to 4.8% on average per annum, returning to a level just below its pre-downturn trend rate in the following years at 2.5%.

Figure 7 highlights the contribution that each institutional sector made to annual net capital stock growth between 1998 and 2014. It shows that during the downturn, a strong rise in general government net capital stock growth partially offset slower net stock growth for households, NFCs and financial corporations, which accounted for a relatively higher share of the total stock over the 1997 to 2014 period. Due to methodological improvements made in this release, the deflators were re-aligned to those consistent with the latest in GFCF. This improvement to the deflator resulted in negative growth to general government in 2013. The increase in current prices (CP) was not enough to outweigh the decrease of the price effect at CVM.

Back to table of contents

8. Analysis by type of asset

Net capital stocks were estimated at £4.1 trillion at the end of 2014. Since 1997, estimates of net capital stocks increased by £1.0 trillion. Figure 8 shows ‘dwellings’ accounted for the largest share of assets in 2014, at £1.7 trillion (42.1%) of the total net capital stocks in 2014. ‘Other buildings and structures’ accounted for the second largest share, £1.4 trillion (34.2%), while ‘Information and Communications Technologies (ICT), other machinery, equipment and weapons systems’, ‘intellectual property products’, and ‘transport equipment’ followed with £0.7 trillion (16.3%), £0.2 trillion (4.8%) and £0.1 trillion (2.4%) respectively.

During the period 1997 to 2014, the asset ‘ICT, other machinery, equipment and weapons systems’ experienced the strongest growth, increasing by 51.2% (2.5% per annum), followed by ‘intellectual property products’ which increased by 37.4% (1.9% per annum).

By comparing net stocks with consumption of fixed capital, the data suggest that there may be proportionally more investment in assets with a shorter life-length (thus having a faster rate of depreciation) and hence the rate of depreciation is increasing. The consumption of fixed capital for the asset ‘intellectual property products’ increased by 66.2% and grew by 3.0% on average per annum. Within this asset, the largest average contribution to annual growth came from ‘software and databases’ at 6.3% on average per annum, followed by ‘research and development’ at 1.5%. The assets ‘software and databases’ and ‘research and development’ contain some of the shortest asset life-lengths, at 5 years for ‘software and databases’ and between 4 and 12 years for ‘research and development’ (206.7 Kb Pdf). ‘ICT, other machinery, equipment and weapons systems’ increased by 66.2% between 1997 and 2014, equivalent to 3.0% on average per annum. ‘Other machinery, equipment and weapons systems’ asset has a life-length of between 10 to 30 years depending on its industry and ‘ICT’ asset has a life-length of between 4 and 12 years.

Table 3 shows the economic downturn between 2008 and 2009 affected the pace of net capital stock accumulation by varying amounts across the assets. The asset ‘dwellings’ experienced the weakest average annual growth at 0.6% following the economic downturn. The average annual growth rate of ‘intellectual property products’ fell from 2.9% pre-downturn to a fall of 0.3% in the downturn period, post-downturn growth remained subdued at 0.8%. Growth in ‘ICT, other machinery, equipment and weapons systems’ fell from 3.7% in the pre-downturn decade to a decrease of 0.1% during the downturn and returned to growth of 1.1% post-downturn.

Figure 9 presents annual net capital stock growth between 1998 and 2014, according to the contribution to growth made from the headline assets (Table 3). The largest contributions to net capital stock growth between 1997 and 2014 came from ‘other buildings and structures’ and ‘dwellings’. This reflected the strong growth of the ‘other buildings and structures’ asset and the large share of ‘dwellings’. During the economic downturn, ‘other buildings and structures’ continued to make positive contributions to annual growth, while the contributions from ‘other machinery, equipment and weapons systems’ and ‘dwellings’ decreased and has remained low since. Due to methodological improvements made in this release, deflators were re-aligned to those consistent with the latest in GFCF. This change to the deflator for ‘other buildings and structures’ caused its contribution to growth in 2013 to be negative. The increase in current prices (CP) was not enough to outweigh the decrease of the price effect at CVM.

Back to table of contents

9. Analysis by industry

Estimates of capital stocks and the consumption of fixed capital can be analysed by industry using the UK Standard Industrial Classification 2007.

The share of the net capital stock held by the 4 broad industry groupings (other production; manufacturing; construction and services) reflects the relative shares of these industries in Gross Value Added (GVA) terms. Table 4 shows that the services industries (sections G to T) held an estimated £3.1 trillion (77.4%) of total net capital stocks at the end of 2014. Other production industries (sections A, B, D and E) held £332 billion (8.2%); construction (section F) £339 billion (8.3%), and manufacturing (section C) £246 billion (6.1%).

Over time, the percentage share of these industry groupings has been relatively stable. Services industries increased their percentage shares by 2.7% since 1997, while the ‘manufacturing’ and ‘other production’ have reduced their percentage share by 2.2% and 0.1% respectively (Figure 10).

In current prices, the types of fixed assets held as stock varies across the industries of the economy in 2014 (Table 5). The services industries held the majority of their total net stocks in ‘dwellings’ (52%) and in ‘other buildings and structures’ (29%).

The construction industry held 72% of its assets in ‘other buildings and structures’ and 25% in ‘dwellings’. ‘Other production’ industries A,B,D,E held the majority of their total capital stocks as ‘other buildings and structures’ (65%) and ‘ICT, other machinery and equipment and weapons systems’ (29%).

In contrast, the manufacturing industry was the only industry group which held the majority of their capital stocks in ‘ICT, other machinery, equipment and weapons systems’ (53%), followed by ‘other buildings and structures’ (33%).

The growth in net capital stock showed variation across the industries (Table 6). Net capital stocks held by manufacturing industries fell during the economic downturn, by 2.2% on average per annum in 2008 and 2009, before continuing to fall by 1.3% per annum between 2010 and 2014. In contrast, growth in the net capital stock held by ‘other production’ industries rose by 2.9% per annum during the downturn from 0.2% per annum in the pre-downturn decade and has risen further in the most recent years to 4.2%.

Figure 11 shows that services industries consistently made the largest positive contribution to annual net capital stock growth between 1997 and 2014. The services industries hold the majority of their total net stocks as ‘dwellings’ (52%) and in ‘other buildings and structures’ (28%) which tend to be high value assets with longer life-lengths than other assets.

Back to table of contents

10. Capital output ratio analysis

The capital output ratio is calculated by dividing capital stocks estimates by gross value added (GVA), to obtain a measure of the capital intensity of each industry. GVA is the value of output less the value of intermediate consumption. Intermediate consumption consists of the value of those goods and services consumed as inputs by the process of production, excluding fixed assets whose consumption are recorded as the consumption of fixed capital.

All else being equal, if there is a higher level of capital stock in the economy, firms should theoretically be able to produce a greater quantity or quality of output using the same quantity of labour input. This higher level of productivity arises because workers have more or better tools or facilities at their disposal. Figure 12 shows that using this measure, the most ‘capital intensive’ industry in 2014 was real estate – holding the majority of assets in dwellings, though this interpretation should be used with caution. The output associated with owner-occupied dwellings has no labour input associated with it, and therefore is fully capital intensive.

The industries traditionally associated with infrastructure, such as ‘electricity, gas, steam and air conditioning supply’ are the most capital intensive excluding ‘real estate activities’. This fits in with prior expectations, as the majority of assets are buildings and heavy machinery in these industries, which are typically highly valuable and long lived. In contrast, service and high tech industries, such as ‘professional, scientific and technical activities’, are below average on this measure. Due to their nature, these industries typically require the intensive use of labour relatively more than capital, and as such, they surface as less capital intensive than the ‘heavy’ industries.

Figure 13 shows that industries have different net capital stock to output ratios over time. It indexes ratios for broad industry groupings to 2010=100. An increase in the line shows that the ratio has increased, and relates to either a decrease in GVA or an increase in net capital stocks. The economic downturn has proved a mixed picture for all industries. Construction saw a spike in 2012, given both an increase in net capital stocks and a noticeably large decrease in GVA. The general trend since the economic downturn is a steady decline in the net capital stock to GVA ratio. It is evident that manufacturing was hit hardest by the downturn, peaking in 2009.

Back to table of contents

11. Net capital stocks per employee

It is possible to analyse the level of capital stock per employee. Net capital stocks are used in this calculation because they provide a better estimate of the value of capital stocks at a point in time compared with gross capital stocks. Estimates of employees are taken from our Labour Market Statistics publication.

Figure 14 shows the annual growth in net capital stock per employee for the period 1998 to 2014. The sharp rise identified between 2007 and 2009 came as a result of a sudden fall in employees in order to offset the slowing growth in capital stocks during this period. The 3 periods of visible negative growth implies that businesses are beginning to shift their resources away from capital and more towards labour inputs.

The relationship between the number of workers and the size of the aggregate net capital stock grows steadily until 2008 (Figure 14). In 2009, the relationship spikes upwards in response to the sharp fall in employment which occurred over the early part of the economic downturn (Figure 15). As employment recovers in the years following 2009, the ratio stabilises and begins to fall as employment starts to grow from 2010, and subdued investment leads to lower than average growth to the aggregate capital stock.

Back to table of contents

.Background notes

  1. What’s new?

    Methodological changes in this release

    These estimates are consistent with the 2015 United Kingdom National Accounts Blue Book and estimates include European System of Accounts (ESA) 2010 changes.

    A number of improvements and changes to the methodology for estimating capital stocks and consumption of fixed capital have been implemented in this publication.

    The changes were presented on 11 September 2015 in methodological changes to the estimation of capital stocks and consumption of fixed capital (269.4 Kb Pdf) and can be summarised as:

    • review of the linking of pre-1997 (‘historic’) GFCF estimates
    • review of deflator linking between 1996 and 1997
    • implementing improved deflators for assets ‘Software and Databases’ and ‘Mineral Exploration’
    • change in the treatment of roads, in this publication the life length was changed to 55 years instead of 75 years as previously used in the model
    • re-classification of Network Rail from private non-financial corporation to central government
    • estimates for public corporations and NPISH institutional sectors are forecast from 2013 onwards, as opposed to coming from the PIM model
    • improvements to the deflation methodology for Central Government and Local Government, these institutional sectors are now deflated by asset in line with the other institutional sectors in this publication.

    Methods changes feeding through from GFCF:

    • Improvements to the estimation of major repairs to dwellings
    • Improvements made for survey undercoverage and other GFCF changes

    Data changes:

    • Alignment with GFCF – using GFCF consistent with Blue Book 2015.
    • Restoring consistency between capital stocks and the consumption of fixed capital estimates.

    An assessment of the impact of these changes are presented in ‘impact of the methodological changes to the estimation of capital stocks and consumption of fixed capital’ (235.4 Kb Pdf) published alongside this release.

    The series in this release cover the period 1997 to 2014. Estimates for 2015 will be published on 5 August 2016 along with estimates for periods 1995 and 1996.

  2. Revisions

    There are revisions to the estimates for 1997 to 2013 in line with the National Accounts revisions policy (41.6 Kb Pdf). This is the result of implementing methodological changes, detailed above in background note 1, along with updating GFCF estimates in line with Blue Book 2015 in the PIM model. The impact of the revisions is described in detail in the ‘impact of the methodological changes to the estimation of capital stocks and consumption of fixed capital’ (235.4 Kb Pdf) accompanying this release.

  3. Use of the data

    Capital stocks and the consumption of fixed capital estimates are produced annually and we use these internally in the UK National and Economic Accounts and Public Sector Finances. They are also used externally by the Bank of England, the Office for Budget Responsibility, Her Majesty’s Treasury, the Statistical Office of the European Communities (EUROSTAT), business and research communities, educational communities, the media and the general public. These estimates are used to monitor economic performance and inform monetary and fiscal policy decisions, as well as for international comparisons.

    The consumption of fixed capital is used in various ways in the National Accounts, in particular:

    • for all sectors, consumption of fixed capital is used to convert the gross-based estimates, such as gross domestic product or gross operating surplus, into the net estimates such as net national income
    • for the non-market sectors (central and local government, and non-profit institutions serving households) consumption of fixed capital forms part of the sectors’ contributions to the economy as measured using gross value added, and is the only component of gross operating surplus for these sectors, leaving the net operating surplus as zero

    Additionally, capital stocks and capital consumption estimates are also used in the calculation of quarterly profitability estimates of UK companies.

  4. Methods

    The aims of the Quality and Methodology Information (QMI) report are to provide users with a greater understanding of our statistics, their quality and the methods that are used to create them. A Quality and Methodology Information report (88.9 Kb Pdf) for this publication can be found on our website; however the methods used to compile these estimates have changed significantly since it was last updated. This bulletin provides an updated description of the methods and sources, along with articles published in September and November 2015.

    A revised QMI report is included in the forward work plan and will be published on our website in 2016.

    Capital stocks and the consumption of fixed capital are estimated using the Perpetual Inventory Method (PIM). The PIM models capital stocks and capital consumption from estimates of gross fixed capital formation (GFCF). Estimates of GFCF and its breakdowns by sector and asset are published separately in the quarterly Business Investment Statistical Bulletin.

    Understanding the data: perpetual inventory method (PIM)

    A PIM is an economic model that enables balance sheets (or stocks) to be calculated from the associated flows. In this case, the PIM uses GFCF estimates to model the value of capital stocks in use in the UK. Assumptions about the life of these capital stocks are used to ensure that they are withdrawn from the model when they are no longer economically useful.

    For estimates of consumption of fixed capital and net capital stocks, assets are written down (that is, decrease in value) over their lifetime. For gross capital stocks, the asset is valued at its new replacement cost until it is retired.

    We assume straight line depreciation. This is a depreciation profile based on a constant rate for the consumption of fixed capital over the service life of the asset. This service life is the total period during which the asset remains in use or ready to be used, in a productive process, even if the asset has more than one owner.

    The main measures produced by the PIM are gross capital stocks, net capital stocks and consumption of fixed capital by asset, industry and sector. The PIM calculates all the series at constant prices and then uses appropriate price indices to produce (reflate) current price estimates.

    The PIM uses GFCF at constant prices for low-level assets and industries across the whole economy. It estimates the capital stocks and consumption of fixed capital series at constant prices at this level. It then applies proportions to sectorise these across the economy based on the historic pattern of capital stocks. These series are then reflated using deflators to give current price estimates.

    Further information on PIM can be found in the following articles:

    UK capital stocks developments in coverage and methodology (446.3 Kb Pdf)

    Improving non financial balance sheets and capital stocks (609.9 Kb Pdf)

    Changes to Methodology

    Below is a summary to the changes to methodology, for more detail and impact see ‘impact of the methodological changes to the estimation of capital stocks and consumption of fixed capital’ (235.4 Kb Pdf)

    • review of the linking of pre-1997 (‘historic’) GFCF estimates

    For GFCF, the low level industry by asset data were originally linked using a factor based on the difference between the 1997 values in the old and new series. The factor is equal to the ratio of the new to the old series in 1997. The linking factor is intended to account for methods changes between the series.

    Since the previous publication, we reviewed the linking of the GFCF series and made improvements. This only affects the GFCF data before 1997. Even though the changes are before 1997, the impact will be seen in 1997, and later years, as the PIM is a cumulative model where the GFCF of previous periods have an impact on the outputs for the current period.

    • review of deflator linking between 1996 and 1997

    Since the previous publication, we reviewed the linking of deflators to provide improved estimates. This affects the constant price (KP) GFCF series before 1997. Even though the changes are before 1997, the impact will be seen in 1997 and later years. The PIM is a cumulative model and the GFCF of previous periods have an impact on the outputs for the current period. For more information see the background section of this bulletin.

    • implementing improved deflators for assets ‘Software and Databases’ and ‘Mineral Exploration’

    During a review of deflation methods, there were significant improvements made to 2 deflators:

    • Software and Databases

    The deflator for the asset ‘computer software and databases’ changed following a National Accounts consultation. The changes include an additional price index to deflate the pre-packaged component of software investment. This change was explained in the ‘improvement to deflation of investment in software’ article.

    • Mineral Exploration

    The asset mineral exploration was deflated using petroleum prices hence a more volatile series in the 2014 publication. The deflator has been changed to ‘other buildings and structures’ as this is appropriate for deflation of these assets.

    • change in the treatment of roads, in this publication the life length was changed to 55 years instead of 75 years as previously used in the model

    In response to a Eurostat reservation, we changed the average life length of roads in the UK from 75 years to 55 years, in line with European guidance and other EU member states. Average life length is used in the estimation of the consumption of fixed capital. It is an estimate of the useful service life of an asset. It can therefore be thought of as the period over which an asset deteriorates until it is no longer able to be used in the production of goods and services.

    Roads are contained with the asset ‘other buildings and structures’ and affect the institutional sectors of ‘central government’ and ‘local government’. A shorter life length leads to a higher estimate of the consumption of fixed capital for each year that the asset is productive. Gross capital stock and net capital stock estimates both reduce as a result of the shorter life length.

    The impact of changing the life length of roads on total fixed assets is approximately 4% of gross and net capital stock and 2% of consumption fixed capital. Therefore the impact of this change is relatively small at the total economy level.

    • re-classification of Network Rail from private non-financial corporation to central government

    Network Rail has been reclassified from the ‘private non-financial corporations’ sector to the ‘central government’ sector with effect from April 2004. This change resulted from the inclusion in ESA 2010 of new rules on assessing public control over non-profit institutions. Additionally, the inclusion of net interest charges in the assessment led to the conclusion that Network Rail was not a ‘market producer’. As a government-controlled, non-market body, Network Rail was reclassified to the ‘central government’ sector.

    This change will have no impact on totals for the whole economy. However, it will move the data in the industry ‘warehousing and support activities for transportation’ for all relevant assets from ‘private non-financial corporations’ to ‘central government’. It will firstly move the GFCF from April 2004 onwards into ‘central government’. Secondly, it moves the gross and net capital stock and the consumption of fixed capital of Network Rail’s fixed assets in April 2004 into ‘central government’.

    More information is available in the article ‘impact of the methodological changes to the estimation of capital stocks and consumption of fixed capital’ (235.4 Kb Pdf)

    • changes to the deflation methodology for Central Government and Local Government

    In this release, the central and local government institutional sectors are now deflated by asset, in line with the other institutional sectors in this publication. Previously they were deflated by one estimated deflator. This only has an impact on CVM measures for these sectors and the impact is detailed in ‘impact of the methodological changes to the estimation of capital stocks and consumption of fixed capital’ (235.4 Kb Pdf) published alongside this release.

    Future work plan:

    • the production of estimates for general, central and local Government using a new methodology

    As a consequence of timetable constraints, we have been unable to fully implement the new methodology to produce estimates using the PIM for the central and local government sectors. As a result, the sector totals used in the previous publication in 2010 have been maintained and the estimates for 2010 to 2014 have been forecast for these sectors based on historic trends. A main input to the estimates of capital stocks and the consumption of fixed capital are gross fixed capital formation (GFCF) investment estimates which come from sample surveys and administrative sources. This means that any updates to gross fixed capital formation (GFCF) for these sectors are not reflected in this release. All other sectors have been produced using the new methods. Due to other development priorities within the Non-Financial Assets Branch at ONS, it has not been possible to improve the methods for calculation of these sectors. We have included these improvements in its future development plan for Non-Financial Assets, and will keep users informed as this work progresses.

    • providing an asset breakdown for institutional sectors S.11001 and S.11PR

    We have been unable to provide an asset breakdown of non-financial corporations (S.11), into public corporations (S.11001) and private non-financial corporations (S.11PR) due to the requirement for additional quality assurance of the asset breakdowns. We have included these improvements in its future development plan for Non-Financial Assets, and will keep users informed as this work progresses.

    • fully introducing new data for Non-profit Institutions Serving Households (NPISH) and Public Non-Financial Corporations

    NPISH data from the PIM model were fixed at an early stage of the production process between 1997 and 2012 due to deliveries required to meet the Blue Book 2014 timetable. NPISH estimates have not been re-calculated by the PIM since this time, apart from the year 2013, they have been forecast since. Public Non-financial Corporations data from the PIM model was fixed following the Blue Book 2014 publication and forecast since this time. Therefore these institutional sectors are inconsistent with outputs for other non-government sectors and updates to gross fixed capital formation (GFCF) for these sectors are not reflected in this release.

    We have included these improvements in its future development plan for non-financial assets, and will keep users informed as this work progresses.

    Data sources

    Estimates of capital stocks are calculated using the Perpetual Inventory Method (PIM).

    The input data for the PIM are at a low level of aggregation with industries classified using the 2007 version of the Standard Industrial Classification (SIC2007). The main input data sets entering the PIM are given below.

    i. GFCF (investment) estimates

    The PIM uses a long run constant price time series of GFCF estimates (investment in capital assets) classified by SIC2007 industries. The series extend as far back as the nineteenth century for some long-lived assets such as buildings and dwellings. The PIM uses constant price investment estimates at 2012 prices.

    Estimates of GFCF and its breakdowns by sector and asset are published separately in the quarterly Business Investment Statistical Release.

    ii. Price indices (deflators)

    The capital stocks system uses, from 1997 onwards, the same deflators used to deflate GFCF; prior to 1997 consistent deflators are used. The GFCF asset deflators are constructed from product based Producer Price Indices (PPIs), Services Producer Price Indices (SPPIs), construction price indices and some Gross Domestic Product (GDP) implied deflators. The product-based deflators are weighted to assets using estimates from our annual Business Spending on Capital Items Survey (BCIS). BCIS collects detailed data on GFCF by product, which informs the weight of each product in each asset. The same deflator is used for each asset, regardless of the industry it is in. We are planning a full review of the GFCF deflators, as part of the National Accounts programme of continuous improvement.

    iii. Life lengths and premature scrapping

    Each asset is assumed to have a life length during which it is economically productive. This will differ between assets, with buildings typically having a longer life length than computer software for example.

    Bankruptcy series are used in the PIM to model firm closures, and it is assumed that half the ICT and other machinery and equipment assets of insolvent firms are prematurely scrapped. Bankruptcy data are provided to us by the UK’s Insolvency Service on an annual basis.

    A review on the life lengths of assets is planned during 2015. We will update users on progress and any changes to estimation when this review has been completed.

    iv. Other data

    Retirements of assets are assumed to be normally distributed around the mean asset life length. That is, not all of the stocks of an asset which lasts 5 years, for example, are assumed to leave the PIM after 5 years – rather, the retirements are spread around this date. To model this, the parameters for the distribution of asset retirements are entered into the PIM. This is called the coefficient of variation for each asset.

    International reporting standards

    The International Financial Reporting Standards (IFRS) were introduced from 2005 onwards in the UK. IFRS is the legally-required financial reporting framework for the consolidated accounts of EU listed groups of companies. IFRS differs in some respects from the UK financial reporting standards.

    The impact on the capital stocks and consumption of fixed capital estimates is difficult to assess as the impact of the transition to IFRS varies by company. Our work provided little evidence that material differences would occur as a result of the transition. Furthermore, the Quarterly Capital Expenditure Survey, which is the source for around 50% of GFCF estimates, asks respondents to record their investment data without any allowance for depreciation. On this basis, the transition to IFRS should not prevent time series analysis of the capital stocks and consumption of fixed capital dataset.

    International Comparisons

    The UK is legally required to produce the capital stocks and consumption of fixed capital estimates in line with the European System of Accounts (ESA). The PIM used to calculate capital stocks and consumption of fixed capital is utilised across the world and is in line with international guidance. The way in which the PIM and its assumptions are implemented will differ between countries. This will reflect the differing circumstances in different countries. This means that users can compare estimates between countries but should be aware that differences in methods may contribute to differences in estimates.

    Both Eurostat and the OECD hold internationally comparable estimates for capital stocks and the consumption of fixed capital. When comparing between countries, users should ensure that they are comparing figures in the same currency and that there are no definitional differences noted.

    Comparisons with Other Data Sources

    National Balance Sheet (NBS)

    There are links between the net capital stocks estimates and the estimates shown for AN.11 Fixed assets in the non-financial section of the NBS. In the NBS, estimates are taken directly from the current price net capital stocks dataset for assets ‘machinery, equipment and weapons systems’ and ‘intellectual property products’ for all sectors.

    For all other assets, estimates are made based on a wide range of data sources of asset values in the economy. For this reason, these 2 data sources will show different values. The biggest difference is the valuation of dwellings. The capital stocks estimates use investment data over time in new dwellings and major improvements to dwellings only and not the land. The NBS is calculated using property market data and also includes the value of land. As a result, the valuation of fixed assets in the NBS are higher than in the net capital stocks dataset.

    Capital services

    Capital services are the flow of services into the production of output that are generated by the capital stock, as opposed to the stock of capital itself. Capital services are a measure of capital input that is more suitable for analysing and modelling productivity.

    The fundamental difference between capital stocks and capital services is that whereas capital stocks are a wealth-based measure of the total amount of capital within the economy, capital services are a measure of the flow of services derived from net stock. Capital services weight together the different assets by their relative productivity, measured by a derived rent function. The rent function, otherwise known as the user cost, is essentially how much the user is willing to pay to use the asset. As a user is willing to pay up to what an asset will give back, the rent function can be seen as measuring the difference in the additional returns between assets. The rent function is important because assets such as computers and software have become cheaper over the last decade, meaning a wealth measure like capital stocks does not truly reflect the return to those assets.

    We have not published a paper on the stand-alone capital service also known as VICS (Volume Index of Capital Services) since Appleton and Wallis 2011 (199.9 Kb Pdf). However capital service estimates have been produced for Multi Factor Productivity (MFP) the latest paper being Field & Franklin 2014. Users should note that these estimates are highly experimental. Volume Index of Capital Services latest publication was released on 23 January 2015.

  5. Quality assurance

    What quality assurance has taken place?

    Quality assurance checks have included detailed analysis at all levels of the data, to guarantee integrity and consistency. In terms of detail, the process has involved analysing all data being entered in to the system, and system outputs to ensure that results are correct according to our specified methodology.

    To summarise, the quality assurance process for the December 2015 publication included:

    • review of the linking of pre-1997 (‘historic’) GFCF estimates
    • review of deflator linking between 1996 and 1997
    • improving deflation for assets ‘Software and Databases’ and ‘Mineral Exploration’
    • improvements to the deflation methodology for Central Government and Local Government, these institutional sectors are now deflated by asset in line with the other institutional sectors in this publication.
    • thorough dataset analysis, ensuring that the data are correct at the lowest level of detail
    • publication analysis – detailed checking of all published articles and tables for accuracy and to better tailor the material to serve the citizen user
    • economic analysis, to review what the data are showing at a macroeconomic level and to see how the proposed estimates compare with existing economic figures
  6. What do you think?

    We are striving to improve this release and its associated commentary. We would welcome any feedback you may have, and would be particularly interested in knowing how you make use of these data to inform your work. Please contact us via email: gcf@ons.gov.uk or telephone Jayne Olney on +44 (0)1633 455250.

    There is a Business and Trade Statistics community on the StatsUserNet website. StatsUserNet is the Royal Statistical Society’s interactive site for users of official statistics. The community objectives are to promote dialogue between users and producers of official business and trade statistics about the structure, content and performance of businesses within the UK. Anyone can join the discussions by registering via either of the links above.

  7. Publication policy

    Details of the policy governing the release of new data are available from the Statistics Authority or from the Media Relations Office email: media.relations@ons.gov.uk. A list of the names of those given pre-publication access to the contents of this bulletin is also available.

  8. Social media

    Follow us on Twitter and receive up to date information about our statistics.

    Like us on Facebook to receive our updates in your newsfeed and to post comments on our page.

  9. Government Statistical Service (GSS) business statistics

    To find out about other official business statistics, and choose the right data for your needs, use the GSS Business Statistics Interactive User Guide. By selecting your topics of interest, the tool will pinpoint publications that should be of interest to you, and provide you with links to more detailed information and the relevant statistical releases. It also offers guidance on which statistics are appropriate for different uses.

  10. Special events

    We have published commentary, analysis and policy on 'Special Events' which may affect statistical outputs. For full details visit the Special Events page on the our website.

  11. Release policy

    All data in this release can be downloaded free of charge from our website. Here are the instructions to obtain a full time series of data from the statistical bulletin or release pages:

    select 'Data in this release'select 'View datasets associated with this release'
    select the latest release
    select 'Select series from this dataset'select the reference table of interest
    select 'View series'
    select the series of interest (Hint: for a custom download you can use SHIFT to select a range of
    series or CTRL to select multiple individual series)
    select 'View selection'
    select 'Download'

  12. Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: media.relations@ons.gov.uk

Back to table of contents

Contact details for this Statistical bulletin

Jayne Olney
gcf@ons.gov.uk
Telephone: +44 (0)1633 455250