1. Main points

Change in gross domestic product (GDP) is the main indicator of economic growth. GDP is estimated to have increased by 0.4% in Quarter 1 (Jan to Mar) 2016 compared with growth of 0.6% in Quarter 4 (Oct to Dec) 2015.

Output increased in services by 0.6% in Quarter 1 (Jan to Mar) 2016. The other 3 main industrial groupings within the economy decreased, with production falling by 0.4%, construction output by 0.9% and agriculture by 0.1%.

GDP was 2.1% higher in Quarter 1 (Jan to Mar) 2016 compared with the same quarter a year ago.

In Quarter 1 (Jan to Mar) 2016, GDP was estimated to have been 7.3% higher than the pre-economic downturn peak of Quarter 1 (Jan to Mar) 2008. From the peak in Quarter 1 (Jan to Mar) 2008 to the trough in Quarter 2 (Apr to June) 2009, the economy shrank by 6.1%.

The preliminary estimate of GDP is produced using the output approach to measuring GDP. At this stage, data content is less than half of the total required for the final output estimate. The estimate is subject to revision as more data become available, but these revisions are typically small between the preliminary and third estimates of GDP, with no upward or downward bias to these revisions.

All figures in this release are seasonally adjusted. In line with the national accounts revision policy, no earlier periods have been revised.

Back to table of contents

2. Understanding the preliminary estimate of GDP

About the preliminary estimate of GDP

Change in GDP is the main indicator of economic growth. The preliminary estimate of GDP is based solely on the output approach to measuring GDP and uses the same data that feed into the Index of Services, Index of Production and Output in the Construction Industry datasets. The growth estimates within this release are created from short-term measures of output and should be considered alongside medium and long-term patterns in the series to give a more comprehensive picture of the main movements (further information on longer-term patterns of GDP, including a comparison with other countries, can be found in the Economic context section).

The output approach measures gross value added (GVA) at a detailed industry level before aggregating to produce an estimate for the whole economy. GDP (as measured by the output approach) can then be calculated by adding taxes and subtracting subsidies (both only available at whole economy level) to this estimate of total GVA. However, as there is no information available on taxes and subsidies at this stage, the quarterly growth for output GVA is taken as a proxy for GDP growth (more information on creating the preliminary estimate of GDP is available on the methods and sources page of our website).

In the second estimate of GDP and the quarterly national accounts, the output GVA and GDP estimates are balanced with the equivalent income and expenditure approaches to produce headline estimates of GVA and GDP. Further information on all 3 approaches to measuring GDP can be found in the national accounts.

All data in this bulletin are seasonally adjusted estimates and have had the effect of price changes removed (in other words, the data are deflated). Further information on some of the main concepts (including seasonal adjustment and deflation) underlying the estimates can be found in background note 8.

The quality of the estimate of GDP

The national accounts are drawn together using data from many different sources. This ensures that the national accounts are comprehensive and provide different perspectives on the economy, for example, sales by retailers and purchases by households. One source of information is from business surveys which use information provided directly from UK businesses. These data are subject to many layers of vigorous quality assurance by highly trained personnel, from clarity and confirmation of individual unit data direct from the business contact to scrutiny of data at the macro level. By comparing and contrasting these different sources, the national accounts produce a single picture of the economy which is consistent, coherent and fully integrated.

The production and publication of each GDP release is managed by a highly skilled team with a strong emphasis on statistical, analytical and economic debate throughout the production process to publish the headline GDP estimate and components. Although a limited audience have access to GDP data ahead of publication, those involved in the process are selected to ensure each GDP estimate receives a rigorous statistical and economic challenge. A “balancing meeting” is held during each production round where presentations assess GDP and components against a swathe of external indicators and a focus on GDP headline components. This is attended by senior managers within the Office for National Statistics (ONS) who challenge the data to ensure consistency and plausibility of the GDP estimate.

The preliminary estimate of GDP is produced around 25 days after the end of the quarter based on data from the output measure only, to provide a timely estimate of GDP, at this stage the data content of this estimate is around 44% of the total required for the final output-based estimate. The methods for producing the preliminary GDP estimate use monthly data for the first 2 months in the quarter (January and February) and forecasts for estimating the third month (March), which incorporate early survey responses where available. More information about the data content for this release can be found in the Assumptions made for March 2016 section and the background notes.

Unlike many short-term indicators published by ONS, there is no simple way of measuring the accuracy of GDP. All estimates, by definition, are subject to statistical uncertainty and for many well-established statistics we measure and publish the sampling error and non-sampling error associated with the estimate, using this as an indicator of accuracy. Since sampling is typically done to determine the characteristics of a whole population, the difference between the sample and population values is considered a sampling error.

Non-sampling errors are a result of deviations from the true value that are not a function of the sample chosen, including various systematic errors and any other errors that are not due to sampling. The estimate of GDP, however, is currently constructed from a wide variety of data sources, some of which are not based on random samples or do not have published sampling and non-sampling errors available and as such it is very difficult to measure both error aspects and their impact on GDP. While development work continues in this area, like all other G7 national statistical institutes, we don't publish a measure of the sampling error/non-sampling error associated with GDP (more information on the quality of the output approach to measuring GDP can be found on the Methods and sources page on our website). It should be noted that we are continually working on methodological changes to improve accuracy of the output approach to measuring GDP. As part of the GDP Continuous Improvement Programme, articles are regularly published on the statistical continuous improvement page, which provide detailed updates of the work carried out so far.

One dimension of measuring accuracy is reliability, which is measured using evidence from analyses of revision to assess the closeness of early estimates to subsequently estimated values. Many users try to minimise the impact of uncertainty through using the historical experience of revisions as a basis for estimating how confident they are in early releases and predicting how far and in what direction the early release might be revised. Revisions are an inevitable consequence of the trade-off between timeliness and accuracy. The estimate is subject to revisions as more data become available, but between the preliminary and third estimates of GDP, revisions are typically small (around 0.1 to 0.2 percentage points), with the frequency of upward and downward revisions broadly equal. Many different approaches can be used to summarise revisions; the Validation and Quality Assurance section in the Quality and Methodology Information report analyse the mean average revision and the mean absolute revision for GDP estimates over data publication iterations. In addition to this analysis, Section 14 of the Revisions to GDP and components in Blue Books 2014 and 2015 article updates the metrics used to test revisions performance in order to answer the question ‘Is GDP biased?

On 11 December 2014, the UK Statistics Authority announced its decision to suspend the designation of Construction price and cost indices (CPCIs) due to concerns about the quality of these deflators. As a result, the UK Statistics Authority also suspended the designation of Output and New orders as National Statistics in respect of the Code of Practice for Official Statistics.

We took over responsibility for the publication and development of the CPCIs from the Department for Business Innovation and Skills on 1 April 2015. On 8 May 2015, we published an article describing the proposed interim solution for construction price and cost indices (CPCIs) to replace the statistical models that had been used in the production of chained volume measures (CVMs) for output in the construction industry since Quarter 3 (July to Sept) 2014 and to provide an ongoing source of data. Since the publication of the Quarterly National Accounts, Quarter 2 (Apr to June) 2015, this interim solution has been used for data periods from Quarter 1 (Jan to Mar) 2014 onwards. This interim solution is used within this release.

Back to table of contents

3. Main information

The preliminary estimate of GDP focuses on the growth in output between 2 consecutive quarters (in this release Quarter 4 (Oct to Dec) 2015 and Quarter 1 (Jan to Mar) 2016). GDP increased by 0.4% in the first quarter of 2016.

The contribution an industry grouping makes to GDP quarterly growth is dependent on the change in that industry grouping and its weight within the output approach to measuring GDP. The current 2012 based weights are: services 78.6%; production 14.9%; construction 5.9%; and agriculture 0.7%.

Services increased by 0.6%, contributing 0.50 percentage points to Quarter 1 (Jan to Mar) 2016 GDP growth (as seen in Figure 1). This followed an increase of 0.8% in Quarter 4 (Oct to Dec) 2015. In the latest quarter there were increases in all 4 of the main services aggregates (distribution, hotels and restaurants; transport, storage and communication; business services and finance; government and other services). Growth in business services and finance slowed from 0.7% growth in Quarter 4 (Oct to Dec) 2015 to 0.3% in Quarter 1 (Jan to Mar) 2016. This was the main reason behind the reduction in services growth between the 2 quarters.

There was a downward contribution (0.05 percentage points) from the production industries; these industries fell by 0.4%, with mining and quarrying decreasing by 2.2% following the same decrease in Quarter 4 (Oct to Dec) 2015 and manufacturing decreasing by 0.4% following a rise of 0.1% in Quarter 4 (Oct to Dec) 2015. Partially offsetting these decreases was a rise of 2.0% in water and waste management following an increase of 0.9% in Quarter 4 (Oct to Dec) 2015 and a rise of 0.4% in energy supply following a decrease of 2.2% in Quarter 4 (Oct to Dec) 2015.

There was a downward contribution (0.05 percentage points) from construction; this industry fell by 0.9%. This follows an increase of 0.3% in Quarter 4 (Oct to Dec) 2015.

Back to table of contents

4. Economic context

As seen in Figure 2, GDP in the UK grew consistently during the 2000s until a financial market shock affected UK and global economic growth in 2008 and 2009. Economic growth resumed towards the end of 2009, but generally at a slower rate than the period prior to 2008 (Figure 2). This growth was also erratic, with several quarters between 2010 and 2012 recording low or declining GDP growth. This 2-year period coincided with special events (for example, severe winter weather in Quarter 4 (Oct to Dec) 2010 and the Diamond Jubilee in Quarter 2 (Apr to June) 2012) that are likely to have affected growth. Since 2013, GDP has grown steadily, passing its pre-downturn peak in Quarter 2 (Apr to June) 2013.

GDP and all of its components are referenced to 2012, making the average index in 2012 equal to 100. It is for this reason that Figure 3 shows all components converging in 2012.

Figure 3 shows the industry breakdown of GDP from 2003. Up until the downturn, services in the UK grew steadily, while production output was broadly flat over the same period. Construction activity grew strongly between 2003 and 2004 and although there was a temporary decline in the mid-2000s, this was reversed in 2006 when construction started growing again.

Industries have shown differing trends following the recent economic downturn. This is illustrated in Figure 4, which shows the path of GDP and its components (excluding agriculture, but including manufacturing which is a sub-component of production), relative to their level in Quarter 1 (Jan to Mar) 2008. The construction and production industries were clearly more acutely affected by the deterioration in economic conditions. Following the downturn, the services industries generally grew steadily with output exceeding its pre-downturn peak in Quarter 1 (Jan to Mar) 2012.

Production and construction activity began to grow in 2010 - with manufacturing showing particular strength – but neither industry sustained this growth. Production output fell between 2011 and 2013 to below levels seen at the height of the downturn in 2009. Construction output sharply decreased in 2012 but started growing again in 2013. Construction output in 2015 as a whole was 3.4% higher than 2014, but much lower than the rate of growth for 2014 (7.5%). Although there has generally been growth across all major components of GDP since the start of 2013, the services industries remain the largest and steadiest contributor to economic growth (Table 1) and the only major component of GDP where output has exceeded its pre-downturn peak.

Our preliminary estimate of GDP is one of the earliest GDP releases to be published internationally. As a result, comprehensive cross- country GDP comparisons cannot yet be made for Quarter 1 (Jan to Mar) 2016.

However, GDP data are widely available for most major economies up to Quarter 4 (Oct to Dec) 2015, and a comparison of this information is shown in Figure 5. The level of GDP in each country has been indexed to Quarter 1 (Jan to Mar) 2008 so that a comparison of recoveries since the global downturn can be made. Cross-country GDP data are publicly available from the Organisation for Economic Co-operation and Development (OECD).

The level of GDP in the UK surpassed its pre-downturn peak in Quarter 2 (Apr to June) 2013. Figure 5 indicates that the UK recovery took longer than some other countries in the G7 (Germany, Canada, France and the US). This is in part due to the nature of the downturn in the UK; GDP fell to a greater extent and as a result has taken longer to recover. Since 2013, the UK has had the fastest growth relative to the rest of the G7 economies.

In Quarter 4 (Oct to Dec) 2015, economic growth in France was 0.3% following from a growth of 0.4% in Quarter 3 (July to Sep) 2015. GDP in Germany and Italy increased by 0.3% and 0.1% on the quarter respectively. GDP in Italy still remains 8.8% below the level observed in Quarter 1 (Jan to Mar) 2008.

Back to table of contents

5. Industry analysis

Agriculture

Agriculture output decreased by 0.1% in Quarter 1 (Jan to Mar) 2016, following an increase of 0.3% in the previous quarter. Between Quarter 1 (Jan to Mar) 2015 and Quarter 1 (Jan to Mar) 2016, agriculture output increased by 1.1%.

Production

The index of production decreased by 0.4% in Quarter 1 (Jan to Mar) 2016, following a decrease of 0.4% in the previous quarter. Mining and quarrying contributed the most to the decrease, contracting by 2.2%. Between Quarter 1 (Jan to Mar) 2015 and Quarter 1 (Jan to Mar) 2016, production output increased by 0.1%.

Construction

Construction output decreased by 0.9% in Quarter 1 (Jan to Mar) 2016, following a increase of 0.3% in the previous quarter. Between Quarter 1 (Jan to Mar) 2015 and Quarter 1 (Jan to Mar) 2016, construction output decreased by 1.7%.

Distribution, hotels and restaurants

The index for distribution, hotels and restaurants increased by 1.3% in Quarter 1 (Jan to Mar) 2016, following an increase of 1.4% in the previous quarter. Wholesale and retail trade and repair of motor vehicles and motorcycles made the largest positive contribution to the increase. Between Quarter 1 (Jan to Mar) 2015 and Quarter 1 (Jan to Mar) 2016, distribution, hotels and restaurants output increased by 4.9%.

Transport, storage and communication

The index for transport, storage and communication increased by 1.0% in Quarter 1 (Jan to Mar) 2016, following an increase of 1.2% in the previous quarter. Computer programming, consultancy and related activities made the largest contribution to the increase. Between Quarter 1 (Jan to Mar) 2015 and Quarter 1 (Jan to Mar) 2016, transport, storage and communication output increased by 4.1%.

Business services and finance

The index for business services and finance increased by 0.3% in Quarter 1 (Jan to Mar) 2016, following an increase of 0.7% in the previous quarter. Real estate activities made the largest positive contribution to the increase. Between Quarter 1 (Jan to Mar) 2015 and Quarter 1 (Jan to Mar) 2016, business services and finance output increased by 2.1%.

Government and other services

The index for government and other services increased by 0.5% in Quarter 1 (Jan to Mar) 2016, following an increase of 0.4% in the previous quarter. Human health activities made the largest positive contribution to the increase. Between Quarter 1 (Jan to Mar) 2015 and Quarter 1 (Jan to Mar) 2016, government and other services output increased by 1.6%.

Back to table of contents

6. Assumptions made for March 2016 in the Quarter 1 (Jan to Mar) 2016 GDP preliminary estimate

Background

The methods for producing the preliminary GDP estimate use monthly data for the first 2 months in the quarter and forecasts for estimating the third month. The forecasts are reinforced by early responses to our Monthly Business Survey (MBS), but the monthly response rate are generally lower at this stage (typically between 30% and 50% at this point in time).

Each of the first 2 months includes monthly data from MBS with the 44,000 businesses sampled, covering the production, manufacturing, services, and retail and construction industries.

The forecasts for March use our standard method of fitting an autoregressive integrated moving average (ARIMA) model with adjustments made for Easter, trading days and outliers. The forecasts are calculated for each individual industry level series (for example, food and beverage services). More information on creating the preliminary estimate of GDP is available on the methods and sources page.

Purpose of this section

This section provides details of the assumptions made for March 2016 for each of the main components of the output approach to measuring GDP: services, production and construction.

It was estimated that there was a 0.2% rise in the output of the services industries between February and March 2016.

At the more detailed level, it was estimated that distribution, hotels and restaurants increased by 0.5%, business services and finance increased by 0.2% transport, storage and communication increased by 0.2% and government and other services increased by 0.1%.

The services data for January and February 2016 used in the calculation of the Quarter 1 (Jan to Mar) 2016 GDP preliminary estimate are consistent with the data contained in the February 2016 Index of Services release published on 27 April 2016.

It was estimated that there was a 1.9% increase in the output of the production industries between February and March 2016. At the more detailed level, it was estimated that mining and quarrying increased by 10.8%, manufacturing increased by 0.4%, energy supply increased by 1.6% and water and waste management increased by 0.2%.

In this release a more significant revision has been made to January than usual reflecting upwards revisions to data for mining and quarrying supplied by the Department of Energy and Climate Change. In line with the usual practice outlined below, these revisions are reflected in the March assumption for total production and mining and quarrying. For this reason some caution should be given to the interpretation of the March growth assumption for production and its components. These and any subsequent revisions will be reflected in the published monthly path for Quarter 1 (Jan to Mar) 2016 in the March Index of Production due for publication on 11 May 2016.

Revisions to the January and February 2016 estimates, published in the latest Index of Production (IoP) release on 8 April 2016, have been used in the calculation of the Quarter 1 (Jan to Mar) 2016 GDP preliminary estimate. To retain coherence between the published monthly and quarterly indices for Quarter 1 (Jan to Mar) 2016, adjustments have been made to the monthly growth rates for March 2016 for total production and its components. This ensures that if the monthly growth rates for March are applied to the published February 2016 indices for total production and the main components (and then an average taken of the January, February and March 2016 indices), the results are consistent with the published quarterly indices.

Monthly data for the construction industries are only available from January 2010.

The forecast for construction is calculated slightly differently to production and services due to the shorter time span of monthly turnover data. More weight is placed on early responses to the monthly business survey for March 2016. Responses from businesses were the starting point to inform the forecasts; this was then adjusted (using information collected in previous months) in recognition that these early responses from businesses tend to be lower than later responses. This approach led to an estimated fall of 0.9% in the output of the construction industries between Quarter 4 (Oct to Dec) 2015 and Quarter 1 (Jan to Mar) 2016.

In this release a more significant revision has been made to January and February than usual reflecting both late survey returns and the effect of seasonal adjustment. In line with the usual practice outlined above, these downward revisions are reflected in the March assumption for construction. For this reason some caution should be given to the interpretation of the March growth assumption for construction. These and any subsequent revisions will be reflected in the published monthly path for Quarter 1 (Jan to Mar) 2016 in the Output in the Construction Industry – March 2016 due for publication on 13 May 2016.

Some revisions (due to receipt of additional survey data and revised seasonal factors allowing for the addition of March 2016 data) to the January and February 2016 estimates, published in the latest Output in the Construction Industry – February 2016 release, on 15 April 2016, have been used in the calculation of the Quarter 1 (Jan to Mar) 2016 GDP preliminary estimate. To retain coherence between the published monthly and quarterly indices for Quarter 1 (Jan to Mar) 2016, adjustments have been made, in line with our normal practice, to the monthly growth rates for March 2016 for construction output. This ensures that if the monthly growth rates for March 2016 are applied to the published February 2016 indices for construction output (and then an average taken of the January, February and March indices), the results are consistent with the published quarterly indices.

Back to table of contents

7 .Background notes

1. What’s new?

Coverage of Standard Business Survey Population extended

The coverage of our Standard Business Survey Population has been extended to include a population of solely PAYE based businesses as described in Improving the Coverage of the Standard Business Survey Population published on 21 December 2015. This change will make our business survey universe more representative.

This is the first preliminary GDP release to contain Monthly Business Survey data based on the extended population of solely PAYE based businesses. Across National Accounts outputs adjustments are made to account for survey under-coverage, meaning that theoretically any increase to GDP caused as a result of improved survey coverage should be offset by a reduction in the under-coverage adjustment. However, it is only possible to reassess the under-coverage adjustment when all surveys used in the National Accounts are based on the new register population, this is not currently the case due to the periodicity and lag of some ONS business survey sources.

In order to mitigate the effect of any potential step change in short term economic indicators as a result of the increased universe, the impact of this improvement was investigated by running turnover results both including and excluding these units at a sub-industry level.

This work has shown that the impact was negligible (less than 0.01 effect on the current price index) in the majority of sub-industries. However, larger impacts were seen in 11 services industries and 2 production industries. In these industries a link factor was applied to remove the step effect. Therefore, the impact on the published preliminary GDP estimate, Index of Production and Index of Services estimates for Quarter 1 (Jan to Mar) 2016 and future estimates will be negligible.

The December IoS publication noted work to revise the back series as part of the Blue Book 2016 round, this work will now be taken forward at a later date, once data for the new universe has been taken on across all National Accounts survey sources and the survey under-coverage adjustment can be reassessed.

Leap year adjustments

A methodological note on leap year adjustments was published on 29 February 2016, explaining how leap years might affect ONS time series and the methods used to adjust for them as part of seasonal adjustment. Economic Review March 2016 was published on 2 March 2016, providing further commentary on the economy, GDP and leap year effects.

In this release a seasonal adjustment review has been undertaken of all series that show statistically significant leap year effects in light of new data for February 2016.

VAT Project

An article entitled HMRC VAT project update was published on 4 April 2016, the fourth in a series of articles. It outlined plans to use HMRC VAT turnover data as a pilot to replace MBS in summer 2016 for parts of the Index of Services and the Output approach to measuring GDP.

Three previous articles have been published in this series:

Feasibility study into the use of HMRC turnover data within Short-term Output Indicators and National Accounts, 14 August 2015.

Exploitation of HMRC VAT data, 7 October 2015.

"HMRC VAT project update" 21 December 2015.

2. What do you think?

As a user of our statistics we would welcome your feedback on this publication. If you would like to get in touch please contact us via email: ios.enquiries@ons.gsi.gov.uk

3. Continuous improvement of GDP: sources, methods and communication

The GDP Output Improvement Report, published on 30 September 2015, provides a detailed update of the implementation of improvements for Blue Book 2015, progress on industry reviews and wider cross-cutting improvements, a comprehensive timetable for the industry review project, progress on experimental statistics, an update of industry quality ratings and progress on experimental statistics. It also features sections on deflation and annual coherence adjustments to improve the understanding and transparency of the methods involved in producing Index of Production, Index of Services, and GDP(O).

Assessment reports by the UK Statistics Authority are available for the output approach to measuring GDP and the short-term indicators that feed into it. Furthermore, the priorities for national accounts production and development over a 5 year period (financial year ending 2014 to financial year ending 2018) are highlighted in the National Accounts and Related Statistics Work Plan and an independent review of the UK's national accounts and balance of payments has been produced as part of our programme of National Statistics Quality Reviews (NSQRs).

4. Special events

We maintain a list of candidate special events in the Special events calendar. As explained in our Special events policy, it is not possible to separate the effects of special events from other changes in the series.

We are considering the impact of the high rainfall and flooding in December 2015 in line with the special events policy. Previous incidents of flooding in January to February 2014, December 2013 and November 2012 were not considered special events.

5. Understanding the data

Short guide to GDP

Gross domestic product (GDP) is an integral part of the UK national accounts and provides a measure of the total economic activity in the UK. GDP is often referred to as one of the main “summary indicators” of economic activity and references to “growth in the economy” invariably refer to the growth in GDP during the latest quarter.

In the UK 3 different, but equivalent, approaches are used in the estimation of GDP:

  • the output or production approach — GDP(O) measures the sum of the value added created through the production of goods and services within the economy (our production or output as an economy); this approach provides the first estimate of GDP and can be used to show how much different industries (for example, services) contribute within the economy
  • the income approach — GDP(I) measures the total income generated by the production of goods and services within the economy; the figures breakdown income into, for example, income earned by companies (corporations), employees and the self employed
  • the expenditure approach — GDP(E) measures the total expenditures on all finished goods and services produced within the economy

How our statistics explain the economy

The Changing Shape of UK Manufacturing, an event coordinated jointly with the Department for Business, Innovation and Skills, took place on 22 October 2014. The event featured a range of talks from users, producers and suppliers of manufacturing statistics, not just from government, but also business representatives and academics. To view the content of the day, please visit Storify.

6. National accounts methodology

The national accounts provide an integrated description of all economic activity within the economic territory of the UK, including activity involving both domestic units (that is, individuals and institutions resident in the UK) and external units (those resident in other countries). In addition to being comprehensive, the accounts are fully integrated and internally consistent. More information can be found in the UK national accounts: a short guide.

7. Interpreting the data

Figures for the most recent quarter are provisional and subject to revision in light of:

  • late responses to surveys and administrative sources
  • forecasts being replaced by actual data
  • revisions to seasonal adjustment factors which are re-estimated every quarter and reviewed annually

Data for the retail industry are broadly comparable with the Retail Sales Index published on 21 April 2016. However, the 2 series operate under different revisions policies meaning there can be timing differences in the updating of the 2 series. Also, adjustments to the data within the Index of Services release are sometimes made at the time of the Blue Book release to improve the coherence of the 3 approaches to measuring GDP. Therefore, inconsistencies between the 2 series are not unusual but tend to be small. There are also conceptual and coverage differences between retail sales and retail output which can lead to apparent inconsistencies.

Sample sizes and data content

This is the first estimate of GDP, based on preliminary information for the quarter. Although based on a significant number of returns from businesses, there is still a lot of information to come in, particularly for March.

The amount of data available at this stage is about 44% of the total data that will be available in 1 year’s time. The estimates in this release are, however, based on a large amount of information returned by businesses across the whole of the economy. Information on activity (more specifically, turnover or sales) is available from about 44,000 businesses for each of the first 2 months of the quarter and from about 20,000 businesses for the third month. In addition, we collect price information on nearly 200,000 individual products each month from around 30,000 businesses. This information is used to remove the effect of price changes from the estimates.

Response rates

Approximately 43% of the data used in the preliminary estimate of GDP are based on data collected via ONS’s Monthly Business Survey (MBS) for production and services. In addition, approximately 6% of the data are collected via ONS's Retail Sales Inquiry (RSI) and approximately 6% are collected via ONS’s Monthly Business Survey for Construction. The remainder is based on data received from other ONS sources and external data sources. At this stage the estimate of GDP includes actual data for January, February and March for the RSI element, but only January and February for the production, services and construction elements. Forecasts are generated to estimate March growth rates which are then compared with early responses to the MBS surveys to assess their credibility. Response rates (for the percentage of sampled turnover returned and also the percentage of questionnaire forms returned) for the most recent month and the 3 months prior are available in the background notes of the Index of Services, Index of Production and Retail Sales statistical bulletins. The response rates for the historical periods are updated to reflect the current level of response, incorporating data from late returns. In addition, response rates for the most recent month are available in the latest Output in the Construction Industry release.

8. Definitions and explanations

Definitions found within the main statistical bulletin are listed.

Index number

An index number is a number which indicates the change in magnitude relative to the magnitude at a specified point, the latter usually taken as 100.

Seasonal adjustment

The index numbers in this statistical bulletin are all seasonally adjusted. This aids interpretation by removing annually recurring fluctuations, for example, due to holidays or other regular seasonal patterns. Unadjusted data are also available.

Seasonal adjustment removes regular variation from a time series. Regular variation includes effects due to month lengths, different activity near particular events, such as shopping activity before Christmas, and regular holidays, such as the May bank holiday.

Some features of the calendar are not regular each year, but are predictable if we have enough data – for example the number of certain days of the week in a month may have an effect, or the impact of the timing of Easter. As Easter changes between March and April we can estimate its effect on time series and allocate it between March and April depending on where Easter falls. Estimates of the effect of the day of the week and Easter are used respectively to make trading day and Easter adjustments prior to seasonal adjustment.

X-13-ARIMA-SEATS is the current seasonal adjustment software used for the short-term indicators that feed into the preliminary estimate of GDP.

Deflation

It is standard practice to present many economic statistics in terms of ‘‘constant prices’’. This means that changes or growth, are not affected by changes in price. The process of removing price changes is known as deflation and the resulting series is often described as volume (as opposed to value). The index numbers in this bulletin are volume measures.

Chained volume

The indices in this bulletin are “chained volume” measures. This means that successive volume estimates are linked (or chained) together. The process of annual chain-linking was introduced in 2003. More information on chain-linking can be found in the Tuke and Reed (2001) article, and an article on chain-linking weights in the output approach to measuring GDP can be found on the methods and sources page.

Gross value added industry weights dataset

An update to the annual weights used within the output approach of GDP has been included in our dataset. These weights have been used since the quarterly national accounts, published on 30 September 2015 and are consistent with the data used in the Blue Book 2015 dataset, published on 30 October 2015. All weights are given in parts per thousand.

9. Quality

Some general information on the quality of the estimate of GDP can be found in the Understanding the preliminary estimate of GDP section in the main part of this statistical bulletin. Further information is available on the methods and sources page of our website.

In addition, a quality and methodology report for estimates of GDP is provided on our website. This report describes, in detail, the intended uses of the statistics presented in this publication, their general quality and the methods used to produce them.

10. National accounts revisions policy

In accordance with the national accounts revision policy, there are no periods open for revision in this release. More information on revisions in the output approach to measuring GDP can be found on the Methods and sources page.

This release includes information available up to January 2016.

11. Revisions triangles

Spreadsheets giving revisions triangles (real time databases) of estimates from 1992 to date are available to download. They can be found under the section Revisions triangles for gross value added at basic prices, chained volume measure.

The revisions triangles for the components of GDP have been temporarily removed following the move to the new Standard Industrial Classification (SIC2007) in October 2011. The revisions triangles for total GDP are still available and the services industry analysis is separately available on a monthly basis via the Index of Services dataset.

Revisions to data provide one indication of the reliability of main indicators. Tables 6 and 7 show summary information on the size and direction of the revisions which have been made to data covering a 5 year period. A statistical test has been applied to the average revision to find out if it is statistically significantly different from zero. An average revision close to zero is desirable as it suggests that revisions are not predictable in any one direction. The result of the test is that the average revision is not statistically different from zero.

Table 6 shows the revisions between the early estimates of gross value added (GVA). The analysis of revisions between month 1 and month 2 uses month 2 estimates published from May 2011 (Quarter 1 (Jan to Mar) 2011) to February 2016 (Quarter 4 (Oct to Dec) 2015). The analysis of revisions between month 2 and month 3 uses month 3 estimates published from Jun 2011 (Quarter 1 (Jan to Mar) 2011) to March 2016 (Quarter 4 (Oct to Dec) 2015).

Table 7 shows the revisions to GVA growth between the estimates published 3 months after the end of the quarter and the equivalent estimate 3 years later. The analysis uses month 3 estimates first published from June 2008 (Quarter 1 (Jan to Mar) 2008) to March 2013 (Quarter 4 (Oct to Dec) 2012).

Understanding the quality of early estimates of Gross Domestic Product, which was first published in December 2009, is available on our website.

This article presents an analysis of revisions to the early estimates of GDP based on a long period database of real time GDP back to 1955. This database is regularly updated and is available on our website.

We published Revisions to GDP and components on 28 January 2014 which updates analysis undertaken previously on GDP revisions, as well as launching a real time £ million database for all the components of both the expenditure and income approaches to measuring GDP.

The Revisions to GDP and components in Blue Books 2014 and 2015 article updates the metrics used to test revisions performance in order to answer the question ‘Is GDP biased?

12. Accessing data

The data presented in the tables of this statistical bulletin are also available to download from the data section of this publication. A completed run of data is available as a time series dataset on our website.

13. Code of Practice for Official Statistics

National Statistics are produced to high professional standards set out in the Code of Practice for Official Statistics. They undergo regular quality assurance reviews to ensure that they meet customer needs. They are produced free from any political interference.

Back to table of contents

8 . Methodology

Contact details for this Statistical bulletin

Robert Smith
ios.enquiries@ons.gsi.gov.uk
Telephone: +44 (0)1633 651618