Table of contents
- Executive summary
- Background and definitions
- Statistics collated relating to e-commerce
- Possible approaches for monitoring e-commerce
- Further enhancements
- Annex 1: Analysis of the indicators - section A
- Annex 1: Analysis of the indicators - section B
- Annex 1: Analysis of the indicators - section C
- Annex 2: E-commerce purchases question
- Annex 3: Eurostat indicators
- Annex 4: EC digital agenda scoreboard
- Supporting information
- References
- Background notes
1. Executive summary
In 2014, as the consequence of a 2013 European Union e-commerce taskforce, chaired by Lord Young of Graffham, the Office for National Statistics began a project to consider how it could contribute to the monitoring of e-commerce nationally.
The initial phase of the work considered, but rejected, the idea of creating a national “e-commerce index”. Instead, this report proposes the use of an indicator set, combining a dashboard approach alongside an interactive graphic, to enable a user to take an over-arching and flexible view of the topic.
The report also sets out some future developments which could be carried out, depending on the views of stakeholders and subject to resource constraints. The two key recommendations are:
ONS, with users, to look into expanding the e-commerce survey to broaden its coverage and whether the sample size of the survey could be increased,
the indicator list should be flexible and reflect changes to the e-commerce landscape. ONS with input from stakeholders to keep the proposed list of indicators under review.
What do you think?
On 7 August 2014, ONS launched a public consultation on the future publication of statistics to measure the digital economy to ensure these statistics continue to meet user needs. As part of this consultation ONS are seeking user views on some of the recommendations raised in this article such as whether an indicator set of the type proposed would be of value to users and what the value of expanding the e-commerce survey to broaden the coverage would be to users. The publication will close on 30 October. The full consultation document can be found on the ONS Consultation pages. To respond to this consultation please complete the online questionnaire or send the questionnaire in Annex A of the consultation document to: esociety@ons.gov.uk
Back to table of contents2. Background and definitions
One of the important economic trends over the last 10 years has been the increased use of e-commerce, by consumers and businesses alike. The take-up of Information and Communication Technologies (ICT) by businesses is seen as crucial to the raising of productivity potential and future growth prospects. Eurostat (2009), for example, reported that the main difference between the United States and the European Union productivity differentials in the period 1995-2005 rests in the adoption of ICT. In terms of the proportion of total turnover derived from e-commerce, when compared with other European countries, the latest Eurostat data places the UK in fifth position.
What is e-commerce?
Electronic or digital commerce, commonly known as e-commerce, is where the buying and selling of products or services is conducted over electronic systems such as the Internet and other computer networks. Modern e-commerce typically uses the Internet at least at one point in a transaction's life-cycle, although it may encompass a wider range of technologies such as e-mail, mobile devices and social media.
The Organisation for Economic Co-operation and Development (OECD) defines an e-commerce transaction as “the sale or purchase of goods or services, conducted over computer networks by methods specifically designed for the purpose of receiving or replacing orders”. Under this definition “the goods or services are ordered by those methods, but the payment and the ultimate delivery of the goods or services do not have to be conducted online”. This definition is used by the ONS to measure e-commerce in the UK.
Back to table of contents3. Statistics collated relating to e-commerce
ONS first collected statistics on Internet access in 1998 and currently produces two reports:
an annual bulletin ‘E-Commerce and ICT Activity’ which measures the adoption and use of Information and Communication Technologies (ICTs) and electronic trading by UK businesses with 10 or more employees1,
an annual ‘Internet Access - Households and Individuals’ publication which explores the use of the Internet by adults and households in Great Britain.
Information estimating the numbers of people in the United Kingdom who have ‘ever’ and ‘never’ used the Internet is also available through the Labour Force Survey and was, until recently, released in the publication Internet Access Quarterly Update2.
It is important that these publications are not viewed in isolation, especially when one is seeking to take an over-arching look at e-commerce. The ability, for example, of individuals to partake in e-commerce activities is inextricably linked to their ability to access and make use of the Internet.
Complementary information is also available from other sources - the retail sales inquiry holds Internet sales information and the Department for Business, Innovation and Skills’ ‘Small Business Survey’ collects data on Internet and website usage by small businesses (including those with no employees). The latter is particularly important as currently ONS’ e-commerce survey only includes businesses with 10 or more employees.
E-commerce and ICT activity
On an annual basis ONS publishes estimates of the values of e-commerce and the adoption of the use of ICT, taken from the annual UK e-commerce survey. The latest available estimates are for 20123.
In this release, total e-commerce sales are estimated by aggregating sales received over a website and sales received over Electronic Data Interchange (EDI). EDI is the computer-to-computer exchange of documents in a standard electronic format, and accounts for around twice the volume of sales over a website.
Constraints
The provision of data and aggregated estimates to Eurostat is a requirement set out in EU Regulation 808/2004 and the survey is conducted in order to meet this requirement. The following should therefore be noted:
estimates only relate to those parts of the economy where coverage is required by Eurostat4. The survey, as it is currently run, does not provide full coverage of all UK ICT and e-commerce business activity,
estimates do not cover businesses which have no employees,
estimates (from 2005 onwards) only cover businesses with 10 or more employees. It is not possible to estimate what the current survey results would be if businesses with fewer than 10 employees were covered. Until the 2004 survey, businesses with fewer than 10 employees were included in the survey; subsequently the survey was refocused on what was required under the EC regulation as funding could not be prioritised to maintain a wider coverage,
estimates are subject to change over time given the regulation may stipulate different collections or definitions in different years. This means for some variables it is not possible to undertake time-series analysis.
Two of the key outputs from this survey are the value of e-commerce sales and the proportion of business turnover they represent. The actual questions used to arrive at these estimates (shown in Image 1) do not directly ask for the value of e-commerce sales but ask, of all orders, what percentage was received via a website (Q30) or via electronic transmission methods (Q34). This approach is internationally agreed and endorsed.
These estimates are then matched to business sales from the Annual Business Survey in order to derive the actual value and the proportion of turnover (see the Quality and Methodology Information Report for e-commerce and ICT (99.7 Kb Pdf)).
Image 1: Questions from the ONS 2013 e-commerce survey relating to proportion of orders received by the business that were received either via their website or via electronic transmission methods
Internet access - household and individuals
ONS first collected detailed statistics on Internet access in 1998. Since then, a number of changes have been made to the survey, including the publication of annual results from 2006. The latest published results are for 20145.
The source of this information is the Opinions and Lifestyle survey – a multi-purpose survey developed by ONS for use by government departments, public bodies, charities, academics and the media. A random sample of about 1,800 adults, aged 16 and over, living in private households throughout Great Britain are selected each month for the survey. After accounting for refusals and where no contact can be made, approximately 1,000 interviews are conducted each month.
Constraints
The provision of the results to Eurostat is a requirement set out in EU Regulation 808/2004 and the survey is conducted in order to meet this requirement. Estimates are subject to change over time given the regulation may stipulate different collections or definitions in different years. This means for some variables it is not possible to undertake time-series analysis.
While there is the flexibility to amend and/or add questions to the Opinions and Lifestyle survey, the sample size makes it difficult to analyse data at a low level of disaggregation, for example, by age and sex.
Notes for statistics collated relating to e-commerce
This is in line with European reporting requirements.
A consultation looking at future publication of statistics to measure the digital economy (August 2014) can be found on the ONS consultation pages.
Information from previous e-commerce surveys is also available – 2013 data is provisionally scheduled for release in November 2014.
The parts of the Standard Industrial Classification 2007 not covered by the survey are as follows: Agriculture, Forestry and Fishing (A); Mining and Quarrying (B); Finance and Insurance Activities (K); Veterinary Activities (Division 75); Public Administration and Defence, Social Security (O); Education (P); Health and Social Work (Q); Arts, Entertainment and Recreation (R); Other Service Activities (S) except SIC 95.1, Repair of computers.
Information from previous Internet access - household and individual surveys is also available.
4. Possible approaches for monitoring e-commerce
Background
This section lists a series of measures which could be considered when monitoring and evaluating e-commerce performance in the UK. Although it is within the remit of ONS to suggest possible indicators to assist with the future measurement of e-commerce, suggesting targets for improvement is beyond its remit.
While it would be technically possible to produce a national e-commerce index, this is not the approach that has been taken forward. Such an index could be formed by calculating a weighted average of relevant indicators. The key argument for production of such an index would be the apparent simplicity of a one number approach to monitoring and evaluation. However, apart from the complexity (and subjectivity) of assigning weights to each of the indicators, a one-number approach can mask some of the underlying trends which can help interpret what is really happening and, to some extent, why. The approach outlined below enables one to take an over-arching look at e-commerce. A flexible approach, where indicators could be removed and or introduced is also seen as important given the fast changing nature of e-commerce and Internet usage.
Eurostat make use of a series of different indicators to help benchmark digital Europe from the perspectives of the individual, the household and the enterprise. Some of these relate directly to e-commerce, whereas others potentially impact on the ability of these entities to partake in e-commerce activities. Annex 3 shows Eurostat’s list of indicators. The Digital Agenda for Europe also has a set of indicators, shown in Annex 4.
This report proposes the use of an indicator set, combining a dashboard approach alongside an interactive graphic. The indicators can be organised into three groups.
A. “Key performance indicators” - the headline figures on e-commerce performance (monetary value of sales and proportion of turnover).
B. “Business engagement” - businesses activity associated with e-commerce (differences by size of business, industry, access to websites, broadband etc.)
C. “Engagement by households and individuals” - activity associated with e-commerce (including adoption of mobile devices and engagement with social media).
List of proposed indicators:
A. Key performance indicators
A1. Volume of e-commerce sales via EDI or via website
A2. E-commerce sales as a percentage of total turnover
B. Business engagement
(Note - currently measured using data from businesses with 10 or more employees)
B3. E-commerce sales as a percentage of total turnover by size of business
B4. Businesses making sales online by industry
B5. Businesses having purchased online (further detail by industry)
B6. Proportion of businesses with a website
B7. Speed of Internet connection (businesses)
B8. Businesses with fixed broadband access (further detail by other types of access)
C. Engagement by households and individuals
C9. Use of the Internet for ordering goods or services from UK, EU, rest of world (individuals)
C10. Daily use of the Internet (individuals)
C11. Use of selected mobile devices to access the Internet (individuals)
C12. Use of the Internet for participating in social networks (individuals)
C13. Reasons for not having Internet access (households)
Analysis of the indicators is included in Annex 1.
Presentation of the indicators
The indicator set has been presented on the ONS website in the form of a dashboard (similar to Image 2). An interactive graphic has also been created, linked from the dashboard.
Image 2: The dashboard approach (screengrab from the ONS website)
Back to table of contents5. Further enhancements
Data collection
As discussed in the section “Statistics collated relating to e-commerce”, there are some constraints associated with the current data collection. The following recommendations may improve the accuracy of the estimates. As with any change to a survey question or methodology, adoption of these recommendations will cause a break in the time series thus making monitoring of progress more complex. Adoption of these recommendations is also subject to the usual resource constraints and prioritisation within ONS. Support from interested parties, including provision of funding, may overcome some of these constraints.
Recommendations:
ONS, with appropriate support from partners, to look into the feasibility, from a cost-benefit, burden and quality perspective, of expanding the e-commerce survey to:
cover businesses with fewer than 10 employees,
widen the range of industries1, and
explore whether the sample size could be increased.
ONS to consider the relationship between these estimates and those from the Annual Business Survey and explore whether there are any improvements that can be made - including asking for the actual monetary value of e-commerce sales rather than the proportion of sales.
Indicator set
Discussions with interested parties have identified several potential subject areas that are not currently represented in the indicator set. In some cases no data exists, in others, there is information available, but there are either concerns over data quality or issues with comparability/availability of the information over time.
Business engagement
Cross border e-commerce – in addition to indicator C9 which looks at household purchases, a further indicator on cross border engagement by businesses could be considered – this (or C9) could also in some way incorporate barriers to cross border e-commerce.
‘Micro and non-employing’ businesses (businesses with zero to nine employees) or SMEs (small to medium sized enterprises, i.e. 0-249 employees) – information from the Small Business Survey could be considered for inclusion alongside the other indicators for the larger employers and/or (as mentioned under Data Collection) the e-commerce survey coverage could be expanded to cover micro businesses.
Value of orders placed online by businesses – a related question, asked in 2013, is shown in Annex 2. Subject to an assessment of data quality and comparability over time this could be included as an indicator - or a further question could be considered for inclusion in the survey in future.
Big data and data analytics – an area of policy interest which as yet is not clearly defined and is therefore currently impractical to collect /release information on.
Retail sales inquiry information could be considered as a separate indicator – currently this is not incorporated into the indicator set.
Regional information – there is potential to look at developing regional estimates from the e-commerce survey.
Indicator B8 arguably may lose relevance as the proportions of businesses using fixed broadband becomes more prevalent - this could be removed in future.
Engagement by household and individuals
Mobile ICT – indicator C11 looks at use of mobile devices to access the Internet but there are other aspects such as m-commerce which could be reflected in the indicator set in future, should there be an appropriate data source.
Cloud computing – data collected for 2012 and 2014, could be included as an indicator in the next update, subject to assessment of data quality
Social media as a communications tool – information regarding use of social media was collected from the household survey (and published) in 2012 with a partial collection in 2013. An indicator relating to social media could be incorporated in the next update.
Recommendation:
The indicator list should be flexible and should reflect changes to the e-commerce landscape. ONS, with input from stakeholders, to keep the proposed list of indicators under review.
To summarise, this paper does not recommend the creation and maintenance of an ‘e- commerce index’. Instead, it proposes the use of an indicator set, combining a dashboard approach alongside an interactive graphic, to enable a user to take an over-arching and flexible view of the topic.
Notes for further enhancements
- Subject to the coverage of the Annual Business Survey. The value of e-commerce sales in Section K (Finance and Insurance Activities), apart from being problematic for businesses to report (issues with definitions), may also skew the analysis for the rest of the economy (due to high levels of electronic cash transfers). Section K is not covered by ABS. The value of e-commerce sales within the other sectors currently excluded may be too small to warrant surveying but could be considered. These are Agriculture, Forestry and Fishing (A); Mining and Quarrying (B); Veterinary Activities (Division 75); Public Administration and Defence, Social Security (O); Education (P); Health and Social Work (Q); Arts, Entertainment and Recreation (R); Other Service Activities (S) except SIC 95.1, Repair of computers.