Research on the pub sector in Scotland phase 1: scoping study

3. Research Methodology

This section contains detail on the objectives of the research and the methodology employed across all the pub types/ sectors interviewed. It also provides a brief explanation on the difficulties encountered throughout the project.

Further detail on the methodology is provided through various annexes containing details of the chronological research process and questionnaires employed. These are referenced throughout this section.

Research objectives

The overall aim of this research was to provide a robust evidence base to assist Ministers in coming to a view as to whether legislation on the pub sector in Scotland is required, and where the parameters of that legislation should apply.

Outputs will aim to help inform future policy direction on better regulation for the Scottish pub sector whether using a voluntary or regulatory approach.

The original research design required the data collection to be made over two phases. The first step, the Scoping Study, aimed to use empirical evidence to assess if any part of the pub sector was unfairly disadvantaged based on case studies from all parts of the sector. This initial exercise also looked to inform whether there was a need for further investigation through follow on research. The second stage aimed to expand upon the key results of the initial case study through a robust quantitative assessment of the market via a wider sample survey of pubs.

This report provides findings following the Scoping Study. Specifically, the first phase aimed to:

• Scope out what was required and whether it was possible to gather empirical evidence, demonstrating whether any part of the sector is unfairly disadvantaged compared to another.
• Carry out case studies of operators/ tenants across all operating models and provide an initial assessment of whether any part of the market appears unfairly treated. Assessment to take into account the whole contractual package they are operating under, business accounts, contractual arrangements (including SCORFA ^[21] received and their value), beer charges, rents paid, etc.…
• Produce case study interviews with a selection of pub companies to understand the different contractual arrangements available. This included flexibility to amend initial contract, rents paid, beer charges, SCORFA arrangements offered and their value.

CGA designed questionnaires to gather both financial and non-financial information and to maximise comparability across different pub types. The questionnaires employed for each type of pub can be found in Annexes 3 to 6.

Methodology

The Scoping Study was designed to be representative of the different pub types across Scotland; i.e. Fully Tied, Partial Tied (Leased/ Tenanted), Managed and Independent Free Trade.

To provide as broad an evidence base as possible CGA used a combination of qualitative and quantitative analysis to help understand the scale of any issues within the Scottish pub tie model, and the rationales behind them.

The research undertaken included a literature review, semi-structured in-depth interviews, case study data collection and triangulation of data by contrasting internal CGA data, Companies House information (when available) and data collected through interviews.

In order to assist with the research, a Sounding Board was established. The Sounding Board comprised of key stakeholders from the Scottish Licensed Trade, relevant trade associations and related businesses. The group was instrumental in assisting CGA with defining both the requirements of the research and in providing access to key contacts. (A detailed remit of the Sounding Board is in Annex 1).

The paragraphs that follow contain information on the construction of the sample for the primary data collection, the recruitment process, the interviews conducted and the challenges faced throughout the whole process.

Case Study Sample Construction

To provide the research with as accurate data of the structure of the current Scottish on trade universe as possible, CGA used their market leading Outlet Index service. The only continuously updated on trade outlet universe database of ca.124,000 GB licensed premises (including individual information on tenure, ownership and operational type).

A number of pubs volunteered to participate in the case study research programme. These pubs were verified against Outlet Index and had been continuously present in the database for at least two years. The sample set was a random sample across tenure type and trading style of outlets to a pre-set quota provided in the project brief.

CGA retained control of the sample base at all times to maintain complete anonymity and made the final decision, regarding those outlets selected for the study, on an entirely confidential basis ^[22] .

The confidentiality of all case study and survey data was paramount to the project. CGA used all reasonable means to ensure strict adherence via the Data Protection Act, Market Research Society ( MRS) Code of Conduct, CGA Internal Confidentiality Policies and Non-Disclosure Agreements.

Premises clarification

The research aimed to cover premises defined as 'pubs'. The definition of a 'pub' applied in this research was based on that employed in previous research undertaken by CGA for the UK Parliament ^[23] where a pub was defined as:

" An establishment where the public can enter without payment, and consume an alcoholic beverage on the premises. The premises primary activities are the sale by retail, and consumption on site of, licensed alcoholic products".

A question was raised by the Sounding Board as to whether other types of premises such as small hotels within strict size/ operational criteria, bar-restaurants, social clubs, etc…, should be included as part of the research. The rationale behind this was recognition of the role that many small hotels play in the Scottish on trade market. Many hotels undertake similar wet-led sales activity as more traditionally defined pubs. This resulted in a review of the current CGA classification of pub premises.

CGA determined, with agreement from the Sounding Board, that some hotels (as defined in CGA's Outlet Index database) should be included in the study as pubs. The generally accepted criterion was to include a hotel with a small number of rooms (maximum of 5 letting rooms) as operating similarly to pubs.

Analysis undertaken by CGA of its hotel segment in Scotland indicated that 664 (of the 1,200 hotels) in the database were potentially of relevance to the 'pub' research. This group of hotels were letting 10 or less rooms ^[24] .

The table below provides detail on the proportion of outlets available for consideration based on number of rooms available for accommodation.

Table 2: Scottish Pub and Hotel Market with bedrooms
[Source: CGA Outlet Index Database as per November 2015 ^[25]

No of bedrooms (only data for 10 or more rooms)
	Total	0 or Not Known	More than 10	10 or less rooms
Pubs Free Tenanted Managed	4,573	4,019	107	447 225 78 144
Hotels*	1,200	-	536	664
Entertainment Bars	96	91	-	5
Other Bars	149	148	-	1

*This represents an estimate as calculated by CGA using all available letting information

During recruitment, those outlets identified as "Hotels" were subject to supplementary filtering questions asking for the number of letting bedrooms. Only those "Hotels" that had 5 rooms or less, and for whom their main source of income came from wet or food sales, were considered suitable for recruitment.

Recruitment process

The flowchart below ( Figure 1) outlines the process that CGA adopted in the recruitment of outlets to conduct the face to face interviews.

Figure 1: Process to collect outlet level data

The recruitment process took part in order to ensure that:

• All respondents fit the relevant criteria for selection. A filtering process during the initial conversation confirmed suitability.
• Interviews were administered across all pub types included within the original recruitment brief.
• Planning and logistical considerations, made on an individual basis, allowed a suitable date, time and location for the interview. It also allowed respondents the necessary time to produce all the additional information required, primarily financial data.

A challenge encountered during the recruitment process was that no tenanted free of tie outlets could be recruited. The main reasons for this was that less than 5% of the pub trade in Scotland (excluding entertainment and other late bars) were of this type. This operational equivalent of the MRO option had not been a key part of Pub Company contractual offers except where there was a specific business rationale for its agreement. Examples of this include a lease to a micro/ local brewery, or small independent regional pub groups ^[26] .

Due to the low representation of free of tie tenanted pubs in Scotland, it proved extremely difficult to recruit relevant respondents who were prepared to fully participate in the survey. As such, CGA decided to recruit five Partially Tied pubs instead.

The recruitment survey took place from 11 ^th January to 11 ^th March 2016 to select 25 pubs for the case study interviews.

The paragraphs that follow explain the methodology employed to collect data from publicans, Managed Pub businesses and Pub Companies. There is also a section summarising the challenges encountered through the research exercise.

Face to face interviews with publicans

The research focused on Scottish tenanted and independent pubs. The operational types covered through the face to face interviews were; Independent Free Trade, Fully Tied and Partially Tied tenanted/ leased pubs.

Independent free trade ( IFT) are those pubs that are wholly operated by the licensee and free to purchase all drinks from independent sources.

Within the tenanted pubs, those Fully Tied represent pubs that are Leased/ Tenanted with a total tie to their Pub Company for drinks.

Pubs that are Partially Tied are defined as those pubs that are Leased/ Tenanted with a partial tie to their Pub Company for drinks (some agreed drinks can be purchased outside their agreement).

For the face to face interviewing, the research required the sample sets to include pubs that operated as identified in Table 3. This was an important part of the recruitment process and ensured that appropriate coverage of all key pub types was included at that stage in the process (see below).

Table 3 Scottish Pub Market and initial research target framework
[Source: CGA Outlet Index Database - November 2015]

	Total number of pubs in Scotland	Target Sample points
Independent Free Trade	2,853	10
Tenanted/ leased pubs [TOTAL] - fully tied - partially tied	1,096	10 5
Total number of pubs (including hotels with 5 or less bedrooms) under tenancy in Scotland where England & Wales MRO legislation could potentially apply	518

In England & Wales, pubs operated by companies with over 500 outlets would be eligible for MRO legislation. In Scotland, ca. 47% of the tenanted market would be included in MRO through operational criteria. When considering the whole pub market, MRO legislation would apply to 10% of all pubs in Scotland ^[27] .

CGA produced three structured bespoke questionnaires for each type of pub interviewed: Independent Free Trade, Partially Tied and Fully Tied. The questionnaires were relevant to each pub type but at the same time maximised comparability across them. The questionnaires were professionally printed in an A4 booklet. A summary of the questionnaire coverage for each type of pub is in Annex 4.

Two experienced CGA field researchers conducted the face to face interviews. The field researchers planned and arranged the appointments directly with the individual interviewees. Each interview took around 90 to 120 minutes to complete. Fieldwork took place from April to July 2016.

On completion of fieldwork, data was collected from 10 Independent Free Trade outlets, 10 tenanted Fully Tied and 5 from Partially Tied tenants.

Data collection from Managed Pubs

In order to obtain a full representation of the market, Managed Pubs were also included in the data collection process. Managed pubs are a pub where, by broad industry definition, a company owns a pub, specifies what it sells, and hires a salaried manager to run it. ^[28]

CGA made requests for Managed outlet case study data from the relevant managed operators in Scotland and with support from the relevant trade associations.

Ideally, the research would have managed operator's select five pub businesses in their estates.

It was not realistically possible for individual pub managers to complete surveys of this type without the tacit agreement of the pub business itself. The head office controls the financial and operational requirements of each pub within their portfolio.

To ensure no bias from this methodology, CGA would obtain data from three sources to provide as broad a range of managed pub business performance data as possible. CGA would then cross-reference all data sources available.

Data would then be analysed and compared against responses given from the tenanted sector, to establish whether any part of the sector was particularly disadvantaged against another.

Nine Managed Pub businesses received an initial contact. CGA requested that they complete a template survey (see Annex 5) to provide detailed financial information and supplemental comments.

Research requests took place from 27 ^th May to 20 ^th June 2016. No managed businesses agreed to take part in the self-completion survey but other primary and secondary data sources were included.

Case Study data collection from Pub Companies

Further to the stated requirement to ensure the broadest coverage and representation of all pub business types in the Scottish trade, there was a necessity to include Pub Companies within the study.

At this point, in conjunction with CGA, both the BBPA, other trade bodies in Scotland and representatives of key Pub Companies within the Sounding Board assisted in identifying those businesses that would form a basis of the research. CGA approached eight Pub Companies. These companies accounted for the vast majority of tenanted pubs in Scotland.

The basis of the criteria set included total number of pubs in Scotland (within all business types), overall size of business (financial and geographical), perceived influence within the sector, profit and turnover.

The Pub Company questionnaire took into account the primary function of this operational type to maximise comparability with the tenanted sample (see Annex 6).

Pub Companies received initial interviews over the phone and then self-completion template surveys to provide detailed financial information and supplemental comments.

Research collection took place from 20 ^th May to 26 ^th August 2016. On completion of fieldwork, five Pub Companies had participated in the survey.

Summary of issues encountered during the research process

During the course of the research process an extensive number of challenges arose that required alternative strategies to be developed and implemented by CGA.

The challenges encountered were broad and varied but focused on the difficulties faced in recruiting suitable outlets and businesses for the case study surveys. This had a detrimental effect on resource planning and extended timelines beyond those originally planned. In addition, (as previously outlined), the lack of Free of Tie Tenanted pubs within the available outlets required a reconsideration and restructure of the sample base.

Table 4 below provides a summary of the challenges encountered alongside the subsequent potential solutions, their implications and outcomes for the research process. For further detail, a full chronology of the research programme can be found in Annex 2.

Table 4: Challenges encountered and solutions provided during the research process

Identified Issue	Potential Solution	Implications & Outcomes
Initial telephone recruitment research failed to recruit a single outlet after 87 calls	- Senior member of the team allocated for the recruitment - Additional piece of communication -Ministerial letter of support distributed to all remaining potential outlets [29] - Review of recruitment questionnaire and criteria	- Recruitment process was delayed and took around 8 weeks to complete (vs the two weeks originally allocated) - 40 potential interviewees recruited - Individual interviewees then further vetted to ensure suitability and commitment
Resource plan was severely affected as a result of the delays encountered in the initial recruitment process and had implications on pre-planned CGA work	CGA reassigned an additional field interviewer. Overall, two researchers were required to cover the interviews.	Data collection was unaffected
No availability of Tenanted Free of Tie outlets during the recruitment process (to note that this represent less than 5% of the overall pub sector)	After conducting internal research to clarify, the number and role of free of tie tenanted pubs in Scotland the cohort was reassigned Instead, the research recruited Tenanted pubs with partial ties.	Overall, five partially tied pubs were recruited and interviewed
Delays in interviewing of several respondents - last minute requested re-schedules/ 'no shows', etc...	Persistence of the field researcher and changes to existing schedules. It was necessary to find several replacements.	Face to face interviews took 11 weeks longer than scheduled - albeit for the last Independent Free Trade interviewee
Field research questionnaire preparation and modifications	Extended timelines and schedule were planned	A fully rounded review process was undertaken. And a consensus questionnaire produced to match the original objectives of the study.

A reluctance of licensees to contribute to the survey, either from the perspective of understanding the necessity for them to contribute based on a broad mistrust, general misinformation, or from concerns regarding the provision of confidential financial documentation, arose. This meant that the overall number of potential pub tie contributors was significantly smaller than expected.

In the case of tied outlets, the numbers of businesses operating from a 'free of tie' contract within Scotland were negligible; as a result, it proved impossible to recruit any willing participants. Independent Free Trade operators also proved difficult to obtain agreement from. Many did not see a valid rationale for engaging in a process that they considered unlikely to affect their individual situation.

In the case of Pub Companies and Managed businesses there was some concern as to whether any contributions made could have a longer term negative impact on them from a legislative angle.

All the above issues had a negative effect on the ability of CGA to recruit, collect, and process the required data in a timely and efficient manner.

At this point (even with the assistance of the Sounding Board) CGA experienced significant difficulties in obtaining the wide ranging selection of data required to complete this report. Complicated further by time delays and varying levels of reluctance on the part of some individual licensees, businesses and Pub Companies to provide the in depth level of (particularly financial) information necessary.

These issues are stratified into several key areas:

• Scottish Licensed Trade Association ( SLTA) & CAMRA membership contacts resulted in disappointingly few directly attributable pubs with which the research could engage.
• Mistrust of Scottish Government and/ or Pub Companies - based on individual licensee experiences. These were different in each case but the primary rationale often related to concerns regarding the effect additional regulations could have on their business - often with negative connotations.
• Concerns over the use of the data and its intentions.
• General unwillingness to provide sensitive personal business financial information.
• Lack of interest/ engagement or failure to see/ understand need for legislation. Especially from the Independent Free Trade sector, and Managed operators, who felt that the key issues did not, or would not, directly affect their businesses.
• Differences in tie implementation in Scotland. For example, spirits are significantly more likely to be 'free to buy' in Scotland than in England & Wales within the context of a standard 'inclusive' tie agreement.

As a consequence of the issues raised above, and concerns about the capability to recruit and collect satisfactory responses, this report recommends not to extend the project to the second phase of reporting (which was to expand upon the key results of the scoping study). This decision is based on the general perceived lack of engagement from within the sector and a consensus that further data collection will not significantly improve the knowledge and insight already gained within Phase One.

The section that follows contains the main findings derived from the Scoping Study, Phase One.